linux/tools/perf/builtin-script.c

3727 lines
96 KiB
C
Raw Normal View History

License cleanup: add SPDX GPL-2.0 license identifier to files with no license Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-01 14:07:57 +00:00
// SPDX-License-Identifier: GPL-2.0
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
#include "builtin.h"
#include "perf.h"
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
#include "util/cache.h"
#include "util/debug.h"
#include <subcmd/exec-cmd.h>
#include "util/header.h"
#include <subcmd/parse-options.h>
#include "util/perf_regs.h"
#include "util/session.h"
#include "util/tool.h"
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
#include "util/symbol.h"
#include "util/thread.h"
#include "util/trace-event.h"
#include "util/util.h"
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
#include "util/evlist.h"
#include "util/evsel.h"
#include "util/sort.h"
#include "util/data.h"
#include "util/auxtrace.h"
#include "util/cpumap.h"
#include "util/thread_map.h"
#include "util/stat.h"
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
#include "util/color.h"
#include "util/string2.h"
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
#include "util/thread-stack.h"
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
#include "util/time-utils.h"
#include "util/path.h"
#include "print_binary.h"
#include <linux/bitmap.h>
#include <linux/kernel.h>
#include <linux/stringify.h>
#include <linux/time64.h>
#include "asm/bug.h"
#include "util/mem-events.h"
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
#include "util/dump-insn.h"
#include <dirent.h>
#include <errno.h>
#include <inttypes.h>
#include <signal.h>
#include <sys/param.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
perf script: Add --insn-trace for instruction decoding Add a --insn-trace short hand option for decoding and disassembling instruction streams for intel_pt. This automatically pipes the output into the xed disassembler to generate disassembled instructions. This just makes this use model much nicer to use. Before % perf record -e intel_pt// ... % perf script --itrace=i0ns --ns -F +insn,-event,-period | xed -F insn: -A -64 swapper 0 [000] 17276.429606186: ffffffff81010486 pt_config ([kernel.kallsyms]) nopl %eax, (%rax,%rax,1) swapper 0 [000] 17276.429606186: ffffffff8101048b pt_config ([kernel.kallsyms]) add $0x10, %rsp swapper 0 [000] 17276.429606186: ffffffff8101048f pt_config ([kernel.kallsyms]) popq %rbx swapper 0 [000] 17276.429606186: ffffffff81010490 pt_config ([kernel.kallsyms]) popq %rbp swapper 0 [000] 17276.429606186: ffffffff81010491 pt_config ([kernel.kallsyms]) popq %r12 swapper 0 [000] 17276.429606186: ffffffff81010493 pt_config ([kernel.kallsyms]) popq %r13 swapper 0 [000] 17276.429606186: ffffffff81010495 pt_config ([kernel.kallsyms]) popq %r14 swapper 0 [000] 17276.429606186: ffffffff81010497 pt_config ([kernel.kallsyms]) popq %r15 swapper 0 [000] 17276.429606186: ffffffff81010499 pt_config ([kernel.kallsyms]) retq swapper 0 [000] 17276.429606186: ffffffff8101063e pt_event_add ([kernel.kallsyms]) cmpl $0x1, 0x1b0(%rbx) swapper 0 [000] 17276.429606186: ffffffff81010645 pt_event_add ([kernel.kallsyms]) mov $0xffffffea, %eax swapper 0 [000] 17276.429606186: ffffffff8101064a pt_event_add ([kernel.kallsyms]) mov $0x0, %edx swapper 0 [000] 17276.429606186: ffffffff8101064f pt_event_add ([kernel.kallsyms]) popq %rbx swapper 0 [000] 17276.429606186: ffffffff81010650 pt_event_add ([kernel.kallsyms]) cmovnz %edx, %eax swapper 0 [000] 17276.429606186: ffffffff81010653 pt_event_add ([kernel.kallsyms]) jmp 0xffffffff81010635 swapper 0 [000] 17276.429606186: ffffffff81010635 pt_event_add ([kernel.kallsyms]) retq swapper 0 [000] 17276.429606186: ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) test %eax, %eax Now: % perf record -e intel_pt// ... % perf script --insn-trace --xed ... same output ... XED needs to be installed with: $ git clone https://github.com/intelxed/mbuild.git mbuild $ git clone https://github.com/intelxed/xed $ cd xed $ ./mfile.py $ ./mfile.py examples $ sudo ./mfile.py --prefix=/usr/local install $ sudo cp obj/examples/xed /usr/local/bin $ xed | head -3 ERROR: required argument(s) were missing Copyright (C) 2017, Intel Corporation. All rights reserved. XED version: [v10.0-328-g7d62c8c49b7b] $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20180920180540.14039-2-andi@firstfloor.org [ Fixed up whitespace damage, added the 'mfile.py examples + cp obj/examples/xed ... ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:36 +00:00
#include <subcmd/pager.h>
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
#include "sane_ctype.h"
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
static char const *script_name;
static char const *generate_script_lang;
static bool debug_mode;
static u64 last_timestamp;
static u64 nr_unordered;
perf script: Add support for dumping symbols Add option to dump symbols found in events. e.g., perf script -f comm,pid,tid,time,trace,sym swapper 0/0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff8100134a cpu_idle ([kernel.kallsyms]) ffffffff81370b39 rest_init ([kernel.kallsyms]) ffffffff81696c23 start_kernel ([kernel.kallsyms].init.text) ffffffff816962af x86_64_start_reservations ([kernel.kallsyms].init.text) ffffffff816963b9 x86_64_start_kernel ([kernel.kallsyms].init.text) sshd 1675/1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff813837aa schedule_hrtimeout_range_clock ([kernel.kallsyms]) ffffffff81383886 schedule_hrtimeout_range ([kernel.kallsyms]) ffffffff8110c4f9 poll_schedule_timeout ([kernel.kallsyms]) ffffffff8110cd20 do_select ([kernel.kallsyms]) ffffffff8110ced8 core_sys_select ([kernel.kallsyms]) ffffffff8110d00d sys_select ([kernel.kallsyms]) ffffffff81002bc2 system_call ([kernel.kallsyms]) 7f1647e56e93 __GI_select (/lib64/libc-2.12.90.so) netstat 1692/1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff81002c3a sysret_careful ([kernel.kallsyms]) 7f7a6cd1b210 __GI___libc_read (/lib64/libc-2.12.90.so) Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-6-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:27 +00:00
static bool no_callchain;
static bool latency_format;
static bool system_wide;
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
static bool print_flags;
static bool nanosecs;
static const char *cpu_list;
static DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
static struct perf_stat_config stat_config;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
static int max_blocks;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
unsigned int scripting_max_stack = PERF_MAX_STACK_DEPTH;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
enum perf_output_field {
PERF_OUTPUT_COMM = 1U << 0,
PERF_OUTPUT_TID = 1U << 1,
PERF_OUTPUT_PID = 1U << 2,
PERF_OUTPUT_TIME = 1U << 3,
PERF_OUTPUT_CPU = 1U << 4,
PERF_OUTPUT_EVNAME = 1U << 5,
PERF_OUTPUT_TRACE = 1U << 6,
PERF_OUTPUT_IP = 1U << 7,
PERF_OUTPUT_SYM = 1U << 8,
PERF_OUTPUT_DSO = 1U << 9,
PERF_OUTPUT_ADDR = 1U << 10,
PERF_OUTPUT_SYMOFFSET = 1U << 11,
perf script: Add an option to print the source line number Add field 'srcline' that displays the source file name and line number associated with the sample ip. The information displayed is the same as from addr2line. $ perf script -f comm,tid,pid,time,ip,sym,dso,symoff,srcline grep 10701/10701 2497321.421013: ffffffff81043ffa native_write_msr_safe+0xa ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/msr.h:95 grep 10701/10701 2497321.421984: ffffffff8165b6b3 _raw_spin_lock+0x13 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/spinlock.h:54 grep 10701/10701 2497321.421990: ffffffff810b64b3 tick_sched_timer+0x53 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/time/tick-sched.c:840 grep 10701/10701 2497321.421992: ffffffff8106f63f run_timer_softirq+0x2f ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/timer.c:1372 Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1386315778-11633-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-12-06 07:42:57 +00:00
PERF_OUTPUT_SRCLINE = 1U << 12,
perf script: Add period data column Adding period data column to be displayed in perf script. It's possible to get period values using -f option, like: $ perf script -f comm,tid,time,period,ip,sym,dso :26019 26019 52414.329088: 3707 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329088: 44 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 1987 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 6 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) ls 26019 52414.329442: 537558 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.329442: 2099 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.330181: 1242100 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.330181: 3774 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.331427: 1083662 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) ls 26019 52414.331427: 360 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) Signed-off-by: Jiri Olsa <jolsa@kernel.org> Acked-by: David Ahern <dsahern@gmail.com> Cc: "Jen-Cheng(Tommy) Huang" <tommy24@gatech.edu> Cc: Andi Kleen <andi@firstfloor.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jen-Cheng(Tommy) Huang <tommy24@gatech.edu> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1408977943-16594-9-git-send-email-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2014-08-25 14:45:42 +00:00
PERF_OUTPUT_PERIOD = 1U << 13,
PERF_OUTPUT_IREGS = 1U << 14,
PERF_OUTPUT_BRSTACK = 1U << 15,
PERF_OUTPUT_BRSTACKSYM = 1U << 16,
PERF_OUTPUT_DATA_SRC = 1U << 17,
PERF_OUTPUT_WEIGHT = 1U << 18,
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
PERF_OUTPUT_BPF_OUTPUT = 1U << 19,
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
PERF_OUTPUT_CALLINDENT = 1U << 20,
PERF_OUTPUT_INSN = 1U << 21,
PERF_OUTPUT_INSNLEN = 1U << 22,
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
PERF_OUTPUT_BRSTACKINSN = 1U << 23,
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
PERF_OUTPUT_BRSTACKOFF = 1U << 24,
PERF_OUTPUT_SYNTH = 1U << 25,
PERF_OUTPUT_PHYS_ADDR = 1U << 26,
PERF_OUTPUT_UREGS = 1U << 27,
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
PERF_OUTPUT_METRIC = 1U << 28,
PERF_OUTPUT_MISC = 1U << 29,
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
PERF_OUTPUT_SRCCODE = 1U << 30,
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
};
struct output_option {
const char *str;
enum perf_output_field field;
} all_output_options[] = {
{.str = "comm", .field = PERF_OUTPUT_COMM},
{.str = "tid", .field = PERF_OUTPUT_TID},
{.str = "pid", .field = PERF_OUTPUT_PID},
{.str = "time", .field = PERF_OUTPUT_TIME},
{.str = "cpu", .field = PERF_OUTPUT_CPU},
{.str = "event", .field = PERF_OUTPUT_EVNAME},
{.str = "trace", .field = PERF_OUTPUT_TRACE},
{.str = "ip", .field = PERF_OUTPUT_IP},
perf script: Add support for dumping symbols Add option to dump symbols found in events. e.g., perf script -f comm,pid,tid,time,trace,sym swapper 0/0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff8100134a cpu_idle ([kernel.kallsyms]) ffffffff81370b39 rest_init ([kernel.kallsyms]) ffffffff81696c23 start_kernel ([kernel.kallsyms].init.text) ffffffff816962af x86_64_start_reservations ([kernel.kallsyms].init.text) ffffffff816963b9 x86_64_start_kernel ([kernel.kallsyms].init.text) sshd 1675/1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff813837aa schedule_hrtimeout_range_clock ([kernel.kallsyms]) ffffffff81383886 schedule_hrtimeout_range ([kernel.kallsyms]) ffffffff8110c4f9 poll_schedule_timeout ([kernel.kallsyms]) ffffffff8110cd20 do_select ([kernel.kallsyms]) ffffffff8110ced8 core_sys_select ([kernel.kallsyms]) ffffffff8110d00d sys_select ([kernel.kallsyms]) ffffffff81002bc2 system_call ([kernel.kallsyms]) 7f1647e56e93 __GI_select (/lib64/libc-2.12.90.so) netstat 1692/1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff81002c3a sysret_careful ([kernel.kallsyms]) 7f7a6cd1b210 __GI___libc_read (/lib64/libc-2.12.90.so) Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-6-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:27 +00:00
{.str = "sym", .field = PERF_OUTPUT_SYM},
{.str = "dso", .field = PERF_OUTPUT_DSO},
{.str = "addr", .field = PERF_OUTPUT_ADDR},
{.str = "symoff", .field = PERF_OUTPUT_SYMOFFSET},
perf script: Add an option to print the source line number Add field 'srcline' that displays the source file name and line number associated with the sample ip. The information displayed is the same as from addr2line. $ perf script -f comm,tid,pid,time,ip,sym,dso,symoff,srcline grep 10701/10701 2497321.421013: ffffffff81043ffa native_write_msr_safe+0xa ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/msr.h:95 grep 10701/10701 2497321.421984: ffffffff8165b6b3 _raw_spin_lock+0x13 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/spinlock.h:54 grep 10701/10701 2497321.421990: ffffffff810b64b3 tick_sched_timer+0x53 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/time/tick-sched.c:840 grep 10701/10701 2497321.421992: ffffffff8106f63f run_timer_softirq+0x2f ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/timer.c:1372 Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1386315778-11633-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-12-06 07:42:57 +00:00
{.str = "srcline", .field = PERF_OUTPUT_SRCLINE},
perf script: Add period data column Adding period data column to be displayed in perf script. It's possible to get period values using -f option, like: $ perf script -f comm,tid,time,period,ip,sym,dso :26019 26019 52414.329088: 3707 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329088: 44 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 1987 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 6 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) ls 26019 52414.329442: 537558 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.329442: 2099 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.330181: 1242100 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.330181: 3774 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.331427: 1083662 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) ls 26019 52414.331427: 360 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) Signed-off-by: Jiri Olsa <jolsa@kernel.org> Acked-by: David Ahern <dsahern@gmail.com> Cc: "Jen-Cheng(Tommy) Huang" <tommy24@gatech.edu> Cc: Andi Kleen <andi@firstfloor.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jen-Cheng(Tommy) Huang <tommy24@gatech.edu> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1408977943-16594-9-git-send-email-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2014-08-25 14:45:42 +00:00
{.str = "period", .field = PERF_OUTPUT_PERIOD},
{.str = "iregs", .field = PERF_OUTPUT_IREGS},
{.str = "uregs", .field = PERF_OUTPUT_UREGS},
{.str = "brstack", .field = PERF_OUTPUT_BRSTACK},
{.str = "brstacksym", .field = PERF_OUTPUT_BRSTACKSYM},
{.str = "data_src", .field = PERF_OUTPUT_DATA_SRC},
{.str = "weight", .field = PERF_OUTPUT_WEIGHT},
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
{.str = "bpf-output", .field = PERF_OUTPUT_BPF_OUTPUT},
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
{.str = "callindent", .field = PERF_OUTPUT_CALLINDENT},
{.str = "insn", .field = PERF_OUTPUT_INSN},
{.str = "insnlen", .field = PERF_OUTPUT_INSNLEN},
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
{.str = "brstackinsn", .field = PERF_OUTPUT_BRSTACKINSN},
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
{.str = "brstackoff", .field = PERF_OUTPUT_BRSTACKOFF},
{.str = "synth", .field = PERF_OUTPUT_SYNTH},
{.str = "phys_addr", .field = PERF_OUTPUT_PHYS_ADDR},
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
{.str = "metric", .field = PERF_OUTPUT_METRIC},
{.str = "misc", .field = PERF_OUTPUT_MISC},
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
{.str = "srccode", .field = PERF_OUTPUT_SRCCODE},
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
};
enum {
OUTPUT_TYPE_SYNTH = PERF_TYPE_MAX,
OUTPUT_TYPE_MAX
};
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
/* default set to maintain compatibility with current format */
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
static struct {
bool user_set;
bool wildcard_set;
unsigned int print_ip_opts;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
u64 fields;
u64 invalid_fields;
} output[OUTPUT_TYPE_MAX] = {
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
[PERF_TYPE_HARDWARE] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
perf script: Show symbol offsets by default Since the ip shown for a symbol is now always a virtual address, it becomes difficult to correlate this with objdump output and determine the exact instruction address. So, we always show the offset from the start of the symbol. This can be verified on a powerpc64le system running Fedora 27 as follows: # perf probe -a sys_write # perf record -e probe:sys_write -g ~/test Before applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so) 10000440 main (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... After applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so) 10000440 main+0x20 (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 06:33:26 +00:00
PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD,
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
},
[PERF_TYPE_SOFTWARE] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
perf script: Show symbol offsets by default Since the ip shown for a symbol is now always a virtual address, it becomes difficult to correlate this with objdump output and determine the exact instruction address. So, we always show the offset from the start of the symbol. This can be verified on a powerpc64le system running Fedora 27 as follows: # perf probe -a sys_write # perf record -e probe:sys_write -g ~/test Before applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so) 10000440 main (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... After applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so) 10000440 main+0x20 (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 06:33:26 +00:00
PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD |
PERF_OUTPUT_BPF_OUTPUT,
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
.invalid_fields = PERF_OUTPUT_TRACE,
},
[PERF_TYPE_TRACEPOINT] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
PERF_OUTPUT_EVNAME | PERF_OUTPUT_TRACE
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
},
perf script: Show hw-cache events 'perf script' fails to report hardware cache events (PERF_TYPE_HW_CACHE) where as 'perf report' shows the samples. Fix it. Ex, # perf record -e L1-dcache-loads ./a.out [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.008 MB perf.data (11 samples)] Before patch: # perf script | wc -l 0 After patch: # perf script | wc -l 11 Committer testing: [root@jouet ~]# perf script | head -30 | tail Timer 9803 [2] 8.963330: 1554 L1-dcache-loads: 7ffef89baae4 __vdso_clock_gettime+0xf4 ([vdso]) swapper 0 [2] 8.963343: 5626 L1-dcache-loads: ffffffffa66f4f6b cpuidle_not_av+0xb (/lib/modules/4.17.0-rc5/build/vmlinux) firefox 4853 [2] 8.964070: 18935 L1-dcache-loads: 7f0b9a00dc30 xcb_poll_for_event+0x0 (/usr/lib64/libxcb.so.1.1.0) Softwar~cTh 4928 [2] 8.964548: 15928 L1-dcache-loads: ffffffffa60d795c update_curr+0x10c (/lib/modules/4.17.0-rc5/build/vmlinux) firefox 4853 [2] 8.964675: 14978 L1-dcache-loads: ffffffffa6897018 mutex_unlock+0x18 (/lib/modules/4.17.0-rc5/build/vmlinux) gnome-shell 2026 [3] 8.964693: 50670 L1-dcache-loads: 7fa08854de6d g_source_iter_next+0x6d (/usr/lib64/libglib-2.0.so.0.5400.3) Compositor 4929 [1] 8.964784: 71772 L1-dcache-loads: 7f0b936bf078 [unknown] (/usr/lib64/firefox/libxul.so) Xwayland 2096 [2] 8.964919: 16799 L1-dcache-loads: 7f68ce2fcb8a glXGetCurrentContext+0x1a (/usr/lib64/libGLX.so.0.0.0) gnome-shell 2026 [3] 8.964997: 50670 L1-dcache-loads: 7fa08854de6d g_source_iter_next+0x6d (/usr/lib64/libglib-2.0.so.0.5400.3) [root@jouet ~]# Signed-off-by: Seeteena Thoufeek <s1seetee@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1528455748-20087-1-git-send-email-s1seetee@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-06-08 11:02:28 +00:00
[PERF_TYPE_HW_CACHE] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD,
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
[PERF_TYPE_RAW] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
perf script: Show symbol offsets by default Since the ip shown for a symbol is now always a virtual address, it becomes difficult to correlate this with objdump output and determine the exact instruction address. So, we always show the offset from the start of the symbol. This can be verified on a powerpc64le system running Fedora 27 as follows: # perf probe -a sys_write # perf record -e probe:sys_write -g ~/test Before applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so) 10000440 main (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... After applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so) 10000440 main+0x20 (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 06:33:26 +00:00
PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD |
PERF_OUTPUT_ADDR | PERF_OUTPUT_DATA_SRC |
PERF_OUTPUT_WEIGHT | PERF_OUTPUT_PHYS_ADDR,
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
perf script: Add support for PERF_TYPE_BREAKPOINT Useful for getting stack traces for hardware breakpoint events. Test result: Before this patch: # ~/perf record -g -e mem:0x600980 ./sample [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.011 MB perf.data (12 samples) ] # ~/perf script # ~/perf script -F comm,tid,pid,time,event,ip,sym,dso sample 22520/22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520/22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) ... After this patch: # ~/perf script sample 22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) Committer note: So, further testing, lets do it for a kernel global variable, tcp_hashinfo: # grep -w tcp_hashinfo /proc/kallsyms ffffffff8202fc00 B tcp_hashinfo # Note: allow specifying mem:tcp_hashinfo: # perf record -g -e mem:0xffffffff81c65ac0 -a ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.790 MB perf.data ] # # perf evlist mem:0xffffffff8202fc00 # perf evlist -v mem:0xffffffff8202fc00: type: 5, size: 112, { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CALLCHAIN|CPU, disabled: 1, inherit: 1, mmap: 1, comm: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, bp_type: 3, { bp_addr, config1 }: 0xffffffff8202fc00, { bp_len, config2 }: 0x4 # Then, after this patch: # perf script swapper 0 [000] 171036.986988: mem:0xffffffff8202fc00: 8a0fb5 __inet_lookup_established (/lib/modules/4.3.0+/build/vmlinux) 8bc09d tcp_v4_early_demux (/lib/modules/4.3.0+/build/vmlinux) 896def ip_rcv_finish (/lib/modules/4.3.0+/build/vmlinux) 8976c2 ip_rcv (/lib/modules/4.3.0+/build/vmlinux) 855eba __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux) 8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux) 8572a8 process_backlog (/lib/modules/4.3.0+/build/vmlinux) 856b11 net_rx_action (/lib/modules/4.3.0+/build/vmlinux) 2a284b __do_softirq (/lib/modules/4.3.0+/build/vmlinux) 2a2ba3 irq_exit (/lib/modules/4.3.0+/build/vmlinux) 96b7a4 do_IRQ (/lib/modules/4.3.0+/build/vmlinux) 969807 ret_from_intr (/lib/modules/4.3.0+/build/vmlinux) 804c27 cpuidle_enter (/lib/modules/4.3.0+/build/vmlinux) 2ded22 call_cpuidle (/lib/modules/4.3.0+/build/vmlinux) 2defb6 cpu_startup_entry (/lib/modules/4.3.0+/build/vmlinux) 95d5bc rest_init (/lib/modules/4.3.0+/build/vmlinux) 1163ffa start_kernel ([kernel.vmlinux].init.text) 11634d7 x86_64_start_reservations ([kernel.vmlinux].init.text) 1163623 x86_64_start_kernel ([kernel.vmlinux].init.text) Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David S. Miller <davem@davemloft.net> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Zefan Li <lizefan@huawei.com> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1449541544-67621-16-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-12-08 02:25:43 +00:00
[PERF_TYPE_BREAKPOINT] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
perf script: Show symbol offsets by default Since the ip shown for a symbol is now always a virtual address, it becomes difficult to correlate this with objdump output and determine the exact instruction address. So, we always show the offset from the start of the symbol. This can be verified on a powerpc64le system running Fedora 27 as follows: # perf probe -a sys_write # perf record -e probe:sys_write -g ~/test Before applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so) 10000440 main (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... After applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so) 10000440 main+0x20 (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 06:33:26 +00:00
PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
PERF_OUTPUT_DSO | PERF_OUTPUT_PERIOD,
perf script: Add support for PERF_TYPE_BREAKPOINT Useful for getting stack traces for hardware breakpoint events. Test result: Before this patch: # ~/perf record -g -e mem:0x600980 ./sample [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.011 MB perf.data (12 samples) ] # ~/perf script # ~/perf script -F comm,tid,pid,time,event,ip,sym,dso sample 22520/22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520/22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) ... After this patch: # ~/perf script sample 22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) Committer note: So, further testing, lets do it for a kernel global variable, tcp_hashinfo: # grep -w tcp_hashinfo /proc/kallsyms ffffffff8202fc00 B tcp_hashinfo # Note: allow specifying mem:tcp_hashinfo: # perf record -g -e mem:0xffffffff81c65ac0 -a ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.790 MB perf.data ] # # perf evlist mem:0xffffffff8202fc00 # perf evlist -v mem:0xffffffff8202fc00: type: 5, size: 112, { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CALLCHAIN|CPU, disabled: 1, inherit: 1, mmap: 1, comm: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, bp_type: 3, { bp_addr, config1 }: 0xffffffff8202fc00, { bp_len, config2 }: 0x4 # Then, after this patch: # perf script swapper 0 [000] 171036.986988: mem:0xffffffff8202fc00: 8a0fb5 __inet_lookup_established (/lib/modules/4.3.0+/build/vmlinux) 8bc09d tcp_v4_early_demux (/lib/modules/4.3.0+/build/vmlinux) 896def ip_rcv_finish (/lib/modules/4.3.0+/build/vmlinux) 8976c2 ip_rcv (/lib/modules/4.3.0+/build/vmlinux) 855eba __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux) 8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux) 8572a8 process_backlog (/lib/modules/4.3.0+/build/vmlinux) 856b11 net_rx_action (/lib/modules/4.3.0+/build/vmlinux) 2a284b __do_softirq (/lib/modules/4.3.0+/build/vmlinux) 2a2ba3 irq_exit (/lib/modules/4.3.0+/build/vmlinux) 96b7a4 do_IRQ (/lib/modules/4.3.0+/build/vmlinux) 969807 ret_from_intr (/lib/modules/4.3.0+/build/vmlinux) 804c27 cpuidle_enter (/lib/modules/4.3.0+/build/vmlinux) 2ded22 call_cpuidle (/lib/modules/4.3.0+/build/vmlinux) 2defb6 cpu_startup_entry (/lib/modules/4.3.0+/build/vmlinux) 95d5bc rest_init (/lib/modules/4.3.0+/build/vmlinux) 1163ffa start_kernel ([kernel.vmlinux].init.text) 11634d7 x86_64_start_reservations ([kernel.vmlinux].init.text) 1163623 x86_64_start_kernel ([kernel.vmlinux].init.text) Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David S. Miller <davem@davemloft.net> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Zefan Li <lizefan@huawei.com> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1449541544-67621-16-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-12-08 02:25:43 +00:00
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
perf script: Add support for PERF_TYPE_BREAKPOINT Useful for getting stack traces for hardware breakpoint events. Test result: Before this patch: # ~/perf record -g -e mem:0x600980 ./sample [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.011 MB perf.data (12 samples) ] # ~/perf script # ~/perf script -F comm,tid,pid,time,event,ip,sym,dso sample 22520/22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520/22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) ... After this patch: # ~/perf script sample 22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) Committer note: So, further testing, lets do it for a kernel global variable, tcp_hashinfo: # grep -w tcp_hashinfo /proc/kallsyms ffffffff8202fc00 B tcp_hashinfo # Note: allow specifying mem:tcp_hashinfo: # perf record -g -e mem:0xffffffff81c65ac0 -a ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.790 MB perf.data ] # # perf evlist mem:0xffffffff8202fc00 # perf evlist -v mem:0xffffffff8202fc00: type: 5, size: 112, { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CALLCHAIN|CPU, disabled: 1, inherit: 1, mmap: 1, comm: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, bp_type: 3, { bp_addr, config1 }: 0xffffffff8202fc00, { bp_len, config2 }: 0x4 # Then, after this patch: # perf script swapper 0 [000] 171036.986988: mem:0xffffffff8202fc00: 8a0fb5 __inet_lookup_established (/lib/modules/4.3.0+/build/vmlinux) 8bc09d tcp_v4_early_demux (/lib/modules/4.3.0+/build/vmlinux) 896def ip_rcv_finish (/lib/modules/4.3.0+/build/vmlinux) 8976c2 ip_rcv (/lib/modules/4.3.0+/build/vmlinux) 855eba __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux) 8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux) 8572a8 process_backlog (/lib/modules/4.3.0+/build/vmlinux) 856b11 net_rx_action (/lib/modules/4.3.0+/build/vmlinux) 2a284b __do_softirq (/lib/modules/4.3.0+/build/vmlinux) 2a2ba3 irq_exit (/lib/modules/4.3.0+/build/vmlinux) 96b7a4 do_IRQ (/lib/modules/4.3.0+/build/vmlinux) 969807 ret_from_intr (/lib/modules/4.3.0+/build/vmlinux) 804c27 cpuidle_enter (/lib/modules/4.3.0+/build/vmlinux) 2ded22 call_cpuidle (/lib/modules/4.3.0+/build/vmlinux) 2defb6 cpu_startup_entry (/lib/modules/4.3.0+/build/vmlinux) 95d5bc rest_init (/lib/modules/4.3.0+/build/vmlinux) 1163ffa start_kernel ([kernel.vmlinux].init.text) 11634d7 x86_64_start_reservations ([kernel.vmlinux].init.text) 1163623 x86_64_start_kernel ([kernel.vmlinux].init.text) Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David S. Miller <davem@davemloft.net> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Zefan Li <lizefan@huawei.com> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1449541544-67621-16-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-12-08 02:25:43 +00:00
},
[OUTPUT_TYPE_SYNTH] = {
.user_set = false,
.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
perf script: Show symbol offsets by default Since the ip shown for a symbol is now always a virtual address, it becomes difficult to correlate this with objdump output and determine the exact instruction address. So, we always show the offset from the start of the symbol. This can be verified on a powerpc64le system running Fedora 27 as follows: # perf probe -a sys_write # perf record -e probe:sys_write -g ~/test Before applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so) 10000440 main (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... After applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so) 10000440 main+0x20 (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 06:33:26 +00:00
PERF_OUTPUT_SYM | PERF_OUTPUT_SYMOFFSET |
PERF_OUTPUT_DSO | PERF_OUTPUT_SYNTH,
.invalid_fields = PERF_OUTPUT_TRACE | PERF_OUTPUT_BPF_OUTPUT,
},
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
};
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
struct perf_evsel_script {
char *filename;
FILE *fp;
u64 samples;
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
/* For metric output */
u64 val;
int gnum;
};
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
static inline struct perf_evsel_script *evsel_script(struct perf_evsel *evsel)
{
return (struct perf_evsel_script *)evsel->priv;
}
static struct perf_evsel_script *perf_evsel_script__new(struct perf_evsel *evsel,
struct perf_data *data)
{
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
struct perf_evsel_script *es = zalloc(sizeof(*es));
if (es != NULL) {
if (asprintf(&es->filename, "%s.%s.dump", data->file.path, perf_evsel__name(evsel)) < 0)
goto out_free;
es->fp = fopen(es->filename, "w");
if (es->fp == NULL)
goto out_free_filename;
}
return es;
out_free_filename:
zfree(&es->filename);
out_free:
free(es);
return NULL;
}
static void perf_evsel_script__delete(struct perf_evsel_script *es)
{
zfree(&es->filename);
fclose(es->fp);
es->fp = NULL;
free(es);
}
static int perf_evsel_script__fprintf(struct perf_evsel_script *es, FILE *fp)
{
struct stat st;
fstat(fileno(es->fp), &st);
return fprintf(fp, "[ perf script: Wrote %.3f MB %s (%" PRIu64 " samples) ]\n",
st.st_size / 1024.0 / 1024.0, es->filename, es->samples);
}
static inline int output_type(unsigned int type)
{
switch (type) {
case PERF_TYPE_SYNTH:
return OUTPUT_TYPE_SYNTH;
default:
return type;
}
}
static inline unsigned int attr_type(unsigned int type)
{
switch (type) {
case OUTPUT_TYPE_SYNTH:
return PERF_TYPE_SYNTH;
default:
return type;
}
}
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
static bool output_set_by_user(void)
{
int j;
for (j = 0; j < OUTPUT_TYPE_MAX; ++j) {
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (output[j].user_set)
return true;
}
return false;
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
static const char *output_field2str(enum perf_output_field field)
{
int i, imax = ARRAY_SIZE(all_output_options);
const char *str = "";
for (i = 0; i < imax; ++i) {
if (all_output_options[i].field == field) {
str = all_output_options[i].str;
break;
}
}
return str;
}
#define PRINT_FIELD(x) (output[output_type(attr->type)].fields & PERF_OUTPUT_##x)
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
static int perf_evsel__do_check_stype(struct perf_evsel *evsel,
u64 sample_type, const char *sample_msg,
enum perf_output_field field,
bool allow_user_set)
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
{
struct perf_event_attr *attr = &evsel->attr;
int type = output_type(attr->type);
const char *evname;
if (attr->sample_type & sample_type)
return 0;
if (output[type].user_set) {
if (allow_user_set)
return 0;
evname = perf_evsel__name(evsel);
pr_err("Samples for '%s' event do not have %s attribute set. "
"Cannot print '%s' field.\n",
evname, sample_msg, output_field2str(field));
return -1;
}
/* user did not ask for it explicitly so remove from the default list */
output[type].fields &= ~field;
evname = perf_evsel__name(evsel);
pr_debug("Samples for '%s' event do not have %s attribute set. "
"Skipping '%s' field.\n",
evname, sample_msg, output_field2str(field));
return 0;
}
static int perf_evsel__check_stype(struct perf_evsel *evsel,
u64 sample_type, const char *sample_msg,
enum perf_output_field field)
{
return perf_evsel__do_check_stype(evsel, sample_type, sample_msg, field,
false);
}
static int perf_evsel__check_attr(struct perf_evsel *evsel,
struct perf_session *session)
{
struct perf_event_attr *attr = &evsel->attr;
bool allow_user_set;
if (perf_header__has_feat(&session->header, HEADER_STAT))
return 0;
allow_user_set = perf_header__has_feat(&session->header,
HEADER_AUXTRACE);
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
if (PRINT_FIELD(TRACE) &&
!perf_session__has_traces(session, "record -R"))
return -EINVAL;
if (PRINT_FIELD(IP)) {
if (perf_evsel__check_stype(evsel, PERF_SAMPLE_IP, "IP",
PERF_OUTPUT_IP))
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
return -EINVAL;
}
if (PRINT_FIELD(ADDR) &&
perf_evsel__do_check_stype(evsel, PERF_SAMPLE_ADDR, "ADDR",
PERF_OUTPUT_ADDR, allow_user_set))
return -EINVAL;
if (PRINT_FIELD(DATA_SRC) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_DATA_SRC, "DATA_SRC",
PERF_OUTPUT_DATA_SRC))
return -EINVAL;
if (PRINT_FIELD(WEIGHT) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_WEIGHT, "WEIGHT",
PERF_OUTPUT_WEIGHT))
return -EINVAL;
perf script: Allow sym and dso without ip, addr Currently sym and dso require printing ip and addr because the print function is tied to those outputs. With callindent it makes sense to print the symbol or dso without numerical IP or ADDR. So change the dependency check to only check the underlying attribute. Also the branch target output relies on the user_set flag to determine if the branch target should be implicitely printed. When modifying the fields with + or - also set user_set, so that ADDR can be removed. We also need to set wildcard_set to make the initial sanity check pass. This allows to remove a lot of noise in callindent output by dropping the numerical addresses, which are not all that useful. Before % perf script --itrace=cr -F +callindent swapper 0 [000] 156546.354971: 1 branches: pt_config 0 [unknown] ([unknown]) => ffffffff81010486 pt_config ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_config ffffffff81010499 pt_config ([kernel.kallsyms]) => ffffffff8101063e pt_event_add ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_event_add ffffffff81010635 pt_event_add ([kernel.kallsyms]) => ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable ffffffff8115e726 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff811579b0 perf_pmu_enable ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void ffffffff81151730 perf_pmu_nop_void ([kernel.kallsyms]) => ffffffff8115e72b event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 ffffffff8115e737 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff8115e7a5 group_sched_in ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax ffffffff8115e7f6 group_sched_in ([kernel.kallsyms]) => ffffffff81a03000 __x86_indirect_thunk_rax ([kernel.kallsyms]) After % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_event_add swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_int swapper 0 [000] 156546.354971: 1 branches: group_sched_in swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: group_sched_in Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:09 +00:00
if (PRINT_FIELD(SYM) &&
!(evsel->attr.sample_type & (PERF_SAMPLE_IP|PERF_SAMPLE_ADDR))) {
pr_err("Display of symbols requested but neither sample IP nor "
perf script: Allow sym and dso without ip, addr Currently sym and dso require printing ip and addr because the print function is tied to those outputs. With callindent it makes sense to print the symbol or dso without numerical IP or ADDR. So change the dependency check to only check the underlying attribute. Also the branch target output relies on the user_set flag to determine if the branch target should be implicitely printed. When modifying the fields with + or - also set user_set, so that ADDR can be removed. We also need to set wildcard_set to make the initial sanity check pass. This allows to remove a lot of noise in callindent output by dropping the numerical addresses, which are not all that useful. Before % perf script --itrace=cr -F +callindent swapper 0 [000] 156546.354971: 1 branches: pt_config 0 [unknown] ([unknown]) => ffffffff81010486 pt_config ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_config ffffffff81010499 pt_config ([kernel.kallsyms]) => ffffffff8101063e pt_event_add ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_event_add ffffffff81010635 pt_event_add ([kernel.kallsyms]) => ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable ffffffff8115e726 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff811579b0 perf_pmu_enable ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void ffffffff81151730 perf_pmu_nop_void ([kernel.kallsyms]) => ffffffff8115e72b event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 ffffffff8115e737 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff8115e7a5 group_sched_in ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax ffffffff8115e7f6 group_sched_in ([kernel.kallsyms]) => ffffffff81a03000 __x86_indirect_thunk_rax ([kernel.kallsyms]) After % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_event_add swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_int swapper 0 [000] 156546.354971: 1 branches: group_sched_in swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: group_sched_in Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:09 +00:00
"sample address\navailable. Hence, no addresses to convert "
"to symbols.\n");
return -EINVAL;
}
if (PRINT_FIELD(SYMOFFSET) && !PRINT_FIELD(SYM)) {
pr_err("Display of offsets requested but symbol is not"
"selected.\n");
return -EINVAL;
}
perf script: Allow sym and dso without ip, addr Currently sym and dso require printing ip and addr because the print function is tied to those outputs. With callindent it makes sense to print the symbol or dso without numerical IP or ADDR. So change the dependency check to only check the underlying attribute. Also the branch target output relies on the user_set flag to determine if the branch target should be implicitely printed. When modifying the fields with + or - also set user_set, so that ADDR can be removed. We also need to set wildcard_set to make the initial sanity check pass. This allows to remove a lot of noise in callindent output by dropping the numerical addresses, which are not all that useful. Before % perf script --itrace=cr -F +callindent swapper 0 [000] 156546.354971: 1 branches: pt_config 0 [unknown] ([unknown]) => ffffffff81010486 pt_config ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_config ffffffff81010499 pt_config ([kernel.kallsyms]) => ffffffff8101063e pt_event_add ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_event_add ffffffff81010635 pt_event_add ([kernel.kallsyms]) => ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable ffffffff8115e726 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff811579b0 perf_pmu_enable ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void ffffffff81151730 perf_pmu_nop_void ([kernel.kallsyms]) => ffffffff8115e72b event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 ffffffff8115e737 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff8115e7a5 group_sched_in ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax ffffffff8115e7f6 group_sched_in ([kernel.kallsyms]) => ffffffff81a03000 __x86_indirect_thunk_rax ([kernel.kallsyms]) After % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_event_add swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_int swapper 0 [000] 156546.354971: 1 branches: group_sched_in swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: group_sched_in Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:09 +00:00
if (PRINT_FIELD(DSO) &&
!(evsel->attr.sample_type & (PERF_SAMPLE_IP|PERF_SAMPLE_ADDR))) {
pr_err("Display of DSO requested but no address to convert.\n");
return -EINVAL;
}
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if ((PRINT_FIELD(SRCLINE) || PRINT_FIELD(SRCCODE)) && !PRINT_FIELD(IP)) {
perf script: Add an option to print the source line number Add field 'srcline' that displays the source file name and line number associated with the sample ip. The information displayed is the same as from addr2line. $ perf script -f comm,tid,pid,time,ip,sym,dso,symoff,srcline grep 10701/10701 2497321.421013: ffffffff81043ffa native_write_msr_safe+0xa ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/msr.h:95 grep 10701/10701 2497321.421984: ffffffff8165b6b3 _raw_spin_lock+0x13 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/spinlock.h:54 grep 10701/10701 2497321.421990: ffffffff810b64b3 tick_sched_timer+0x53 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/time/tick-sched.c:840 grep 10701/10701 2497321.421992: ffffffff8106f63f run_timer_softirq+0x2f ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/timer.c:1372 Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1386315778-11633-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-12-06 07:42:57 +00:00
pr_err("Display of source line number requested but sample IP is not\n"
"selected. Hence, no address to lookup the source line number.\n");
return -EINVAL;
}
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (PRINT_FIELD(BRSTACKINSN) &&
!(perf_evlist__combined_branch_type(session->evlist) &
PERF_SAMPLE_BRANCH_ANY)) {
pr_err("Display of branch stack assembler requested, but non all-branch filter set\n"
"Hint: run 'perf record -b ...'\n");
return -EINVAL;
}
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
if ((PRINT_FIELD(PID) || PRINT_FIELD(TID)) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_TID, "TID",
PERF_OUTPUT_TID|PERF_OUTPUT_PID))
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
return -EINVAL;
if (PRINT_FIELD(TIME) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_TIME, "TIME",
PERF_OUTPUT_TIME))
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
return -EINVAL;
if (PRINT_FIELD(CPU) &&
perf_evsel__do_check_stype(evsel, PERF_SAMPLE_CPU, "CPU",
PERF_OUTPUT_CPU, allow_user_set))
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
return -EINVAL;
if (PRINT_FIELD(IREGS) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_REGS_INTR, "IREGS",
PERF_OUTPUT_IREGS))
return -EINVAL;
if (PRINT_FIELD(UREGS) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_REGS_USER, "UREGS",
PERF_OUTPUT_UREGS))
return -EINVAL;
if (PRINT_FIELD(PHYS_ADDR) &&
perf_evsel__check_stype(evsel, PERF_SAMPLE_PHYS_ADDR, "PHYS_ADDR",
PERF_OUTPUT_PHYS_ADDR))
return -EINVAL;
return 0;
}
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
static void set_print_ip_opts(struct perf_event_attr *attr)
{
unsigned int type = output_type(attr->type);
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
output[type].print_ip_opts = 0;
if (PRINT_FIELD(IP))
output[type].print_ip_opts |= EVSEL__PRINT_IP;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
if (PRINT_FIELD(SYM))
output[type].print_ip_opts |= EVSEL__PRINT_SYM;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
if (PRINT_FIELD(DSO))
output[type].print_ip_opts |= EVSEL__PRINT_DSO;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
if (PRINT_FIELD(SYMOFFSET))
output[type].print_ip_opts |= EVSEL__PRINT_SYMOFFSET;
perf script: Add an option to print the source line number Add field 'srcline' that displays the source file name and line number associated with the sample ip. The information displayed is the same as from addr2line. $ perf script -f comm,tid,pid,time,ip,sym,dso,symoff,srcline grep 10701/10701 2497321.421013: ffffffff81043ffa native_write_msr_safe+0xa ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/msr.h:95 grep 10701/10701 2497321.421984: ffffffff8165b6b3 _raw_spin_lock+0x13 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/arch/x86/include/asm/spinlock.h:54 grep 10701/10701 2497321.421990: ffffffff810b64b3 tick_sched_timer+0x53 ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/time/tick-sched.c:840 grep 10701/10701 2497321.421992: ffffffff8106f63f run_timer_softirq+0x2f ([kernel.kallsyms]) /usr/src/debug/kernel-3.9.fc17/linux-3.9.10-100.fc17.x86_64/kernel/timer.c:1372 Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1386315778-11633-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-12-06 07:42:57 +00:00
if (PRINT_FIELD(SRCLINE))
output[type].print_ip_opts |= EVSEL__PRINT_SRCLINE;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
}
/*
* verify all user requested events exist and the samples
* have the expected data
*/
static int perf_session__check_output_opt(struct perf_session *session)
{
perf script: Show call graphs when 1st event doesn't have it but some other has There's a display inconsistency when there are multiple tracepoint events, some of which have the 'call-graph' config option set but the first one hasn't, i.e. the whole logic for call graph processing is enabled only if the first tracepoint event has call-graph set. For instance, if we record signal_deliver with call-graph and signal_generate without: $ perf record -g -a -e signal:signal_deliver -e signal:signal_generate/call-graph=no/ [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:1 13 [000] 6563.875949: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1313 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1313 [000] 6563.877584: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Then we exchange the order of these two events in commandline, and keep signal_generate without call-graph. $ perf record -g -a -e signal:signal_generate/call-graph=no/ -e signal:signal_deliver [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 This time, the callchain of the event signal_deliver disappeared. The problem is caused by that perf only checks for the first evsel in evlist and decides if callchain should be printed. This patch traverses all evsels in evlist to see if any of them have callchains, and shows the right result: $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Signed-off-by: He Kuang <hekuang@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1463374279-97209-1-git-send-email-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-05-16 04:51:19 +00:00
unsigned int j;
struct perf_evsel *evsel;
for (j = 0; j < OUTPUT_TYPE_MAX; ++j) {
evsel = perf_session__find_first_evtype(session, attr_type(j));
/*
* even if fields is set to 0 (ie., show nothing) event must
* exist if user explicitly includes it on the command line
*/
if (!evsel && output[j].user_set && !output[j].wildcard_set &&
j != OUTPUT_TYPE_SYNTH) {
pr_err("%s events do not exist. "
"Remove corresponding -F option to proceed.\n",
event_type(j));
return -1;
}
if (evsel && output[j].fields &&
perf_evsel__check_attr(evsel, session))
return -1;
if (evsel == NULL)
continue;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
set_print_ip_opts(&evsel->attr);
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
}
if (!no_callchain) {
bool use_callchain = false;
2016-08-04 11:25:43 +00:00
bool not_pipe = false;
evlist__for_each_entry(session->evlist, evsel) {
2016-08-04 11:25:43 +00:00
not_pipe = true;
if (evsel__has_callchain(evsel)) {
use_callchain = true;
break;
}
}
2016-08-04 11:25:43 +00:00
if (not_pipe && !use_callchain)
symbol_conf.use_callchain = false;
}
/*
* set default for tracepoints to print symbols only
* if callchains are present
*/
if (symbol_conf.use_callchain &&
!output[PERF_TYPE_TRACEPOINT].user_set) {
j = PERF_TYPE_TRACEPOINT;
evlist__for_each_entry(session->evlist, evsel) {
perf script: Show call graphs when 1st event doesn't have it but some other has There's a display inconsistency when there are multiple tracepoint events, some of which have the 'call-graph' config option set but the first one hasn't, i.e. the whole logic for call graph processing is enabled only if the first tracepoint event has call-graph set. For instance, if we record signal_deliver with call-graph and signal_generate without: $ perf record -g -a -e signal:signal_deliver -e signal:signal_generate/call-graph=no/ [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:1 13 [000] 6563.875949: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1313 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1313 [000] 6563.877584: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Then we exchange the order of these two events in commandline, and keep signal_generate without call-graph. $ perf record -g -a -e signal:signal_generate/call-graph=no/ -e signal:signal_deliver [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 This time, the callchain of the event signal_deliver disappeared. The problem is caused by that perf only checks for the first evsel in evlist and decides if callchain should be printed. This patch traverses all evsels in evlist to see if any of them have callchains, and shows the right result: $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Signed-off-by: He Kuang <hekuang@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1463374279-97209-1-git-send-email-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-05-16 04:51:19 +00:00
if (evsel->attr.type != j)
continue;
if (evsel__has_callchain(evsel)) {
perf script: Show call graphs when 1st event doesn't have it but some other has There's a display inconsistency when there are multiple tracepoint events, some of which have the 'call-graph' config option set but the first one hasn't, i.e. the whole logic for call graph processing is enabled only if the first tracepoint event has call-graph set. For instance, if we record signal_deliver with call-graph and signal_generate without: $ perf record -g -a -e signal:signal_deliver -e signal:signal_generate/call-graph=no/ [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:1 13 [000] 6563.875949: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1313 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1313 [000] 6563.877584: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Then we exchange the order of these two events in commandline, and keep signal_generate without call-graph. $ perf record -g -a -e signal:signal_generate/call-graph=no/ -e signal:signal_deliver [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 This time, the callchain of the event signal_deliver disappeared. The problem is caused by that perf only checks for the first evsel in evlist and decides if callchain should be printed. This patch traverses all evsels in evlist to see if any of them have callchains, and shows the right result: $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Signed-off-by: He Kuang <hekuang@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1463374279-97209-1-git-send-email-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-05-16 04:51:19 +00:00
output[j].fields |= PERF_OUTPUT_IP;
output[j].fields |= PERF_OUTPUT_SYM;
perf script: Show symbol offsets by default Since the ip shown for a symbol is now always a virtual address, it becomes difficult to correlate this with objdump output and determine the exact instruction address. So, we always show the offset from the start of the symbol. This can be verified on a powerpc64le system running Fedora 27 as follows: # perf probe -a sys_write # perf record -e probe:sys_write -g ~/test Before applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17 (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts (/usr/lib64/libc-2.26.so) 10000440 main (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... After applying this patch: # perf script test 9710 [013] 95614.332431: probe:sys_write: (c0000000004025b0) c0000000004025b0 sys_write+0x10 (/lib/modules/4.17.0-rc4+/build/vmlinux) c00000000000b9e0 system_call+0x58 (/lib/modules/4.17.0-rc4+/build/vmlinux) 7fffb70d8234 __GI___libc_write+0x24 (/usr/lib64/libc-2.26.so) 7fffb7052c74 _IO_file_write@@GLIBC_2.17+0x44 (/usr/lib64/libc-2.26.so) 5afc1818 [unknown] ([unknown]) 7fffb7051a60 new_do_write+0x90 (/usr/lib64/libc-2.26.so) 7fffb7054638 _IO_do_write@@GLIBC_2.17+0x38 (/usr/lib64/libc-2.26.so) 7fffb7054bbc _IO_file_overflow@@GLIBC_2.17+0x14c (/usr/lib64/libc-2.26.so) 7fffb7055a24 __overflow+0x64 (/usr/lib64/libc-2.26.so) 7fffb7044548 _IO_puts+0x218 (/usr/lib64/libc-2.26.so) 10000440 main+0x20 (/home/sandipan/test) 7fffb6fe36a0 generic_start_main.isra.0+0x140 (/usr/lib64/libc-2.26.so) 7fffb6fe3898 __libc_start_main+0xb8 (/usr/lib64/libc-2.26.so) 0 [unknown] ([unknown]) ... Signed-off-by: Sandipan Das <sandipan@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Link: http://lkml.kernel.org/r/20180517063326.6319-2-sandipan@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 06:33:26 +00:00
output[j].fields |= PERF_OUTPUT_SYMOFFSET;
perf script: Show call graphs when 1st event doesn't have it but some other has There's a display inconsistency when there are multiple tracepoint events, some of which have the 'call-graph' config option set but the first one hasn't, i.e. the whole logic for call graph processing is enabled only if the first tracepoint event has call-graph set. For instance, if we record signal_deliver with call-graph and signal_generate without: $ perf record -g -a -e signal:signal_deliver -e signal:signal_generate/call-graph=no/ [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:1 13 [000] 6563.875949: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1313 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1313 [000] 6563.877584: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Then we exchange the order of these two events in commandline, and keep signal_generate without call-graph. $ perf record -g -a -e signal:signal_generate/call-graph=no/ -e signal:signal_deliver [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 This time, the callchain of the event signal_deliver disappeared. The problem is caused by that perf only checks for the first evsel in evlist and decides if callchain should be printed. This patch traverses all evsels in evlist to see if any of them have callchains, and shows the right result: $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Signed-off-by: He Kuang <hekuang@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1463374279-97209-1-git-send-email-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-05-16 04:51:19 +00:00
output[j].fields |= PERF_OUTPUT_DSO;
set_print_ip_opts(&evsel->attr);
perf script: Show call graphs when 1st event doesn't have it but some other has There's a display inconsistency when there are multiple tracepoint events, some of which have the 'call-graph' config option set but the first one hasn't, i.e. the whole logic for call graph processing is enabled only if the first tracepoint event has call-graph set. For instance, if we record signal_deliver with call-graph and signal_generate without: $ perf record -g -a -e signal:signal_deliver -e signal:signal_generate/call-graph=no/ [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:1 13 [000] 6563.875949: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1313 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1313 [000] 6563.877584: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Then we exchange the order of these two events in commandline, and keep signal_generate without call-graph. $ perf record -g -a -e signal:signal_generate/call-graph=no/ -e signal:signal_deliver [ perf record: Captured and wrote 0.017 MB perf.data (2 samples) ] $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 This time, the callchain of the event signal_deliver disappeared. The problem is caused by that perf only checks for the first evsel in evlist and decides if callchain should be printed. This patch traverses all evsels in evlist to see if any of them have callchains, and shows the right result: $ perf script kworker/u2:2 1314 [000] 6933.353060: signal:signal_generate: sig=2 errno=0 code=128 comm=perf pid=1321 grp=1 res=0 ff61cc __send_signal+0x3ec ([kernel.kallsyms]) perf 1321 [000] 6933.353872: signal:signal_deliver: sig=2 errno=0 code=128 sa_handler=43115e sa_flags=14000000 7ffff314 get_signal+0x80007f0023a4 ([kernel.kallsyms]) 7fffe358 do_signal+0x80007f002028 ([kernel.kallsyms]) 7fffa5e8 exit_to_usermode_loop+0x80007f002053 ([kernel.kallsyms]) ... Signed-off-by: He Kuang <hekuang@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1463374279-97209-1-git-send-email-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-05-16 04:51:19 +00:00
goto out;
}
}
}
out:
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
return 0;
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
perf script: Share code and output format for uregs and iregs output The iregs output was missing the newline at end as well as the leading ABI output. This made it hard to compare the iregs and uregs values. Instead, use a single function to output the register values and use it for both, iregs and uregs, to ensure the output is consistent. Before: perf 7049 [-01] 1343.354347: 1 cycles:ppp: ffffffffa7bc21ce perf_event_exec+0x18e (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7ead3 setup_new_exec+0xf3 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7cd7be5 load_elf_binary+0x395 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7e540 search_binary_handler+0x80 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f1aa __do_execve_file.isra.13+0x58a (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f561 do_execve+0x21 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f596 __x64_sys_execve+0x26 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7a041cb do_syscall_64+0x5b (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa840008c entry_SYSCALL_64+0x7c (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) AX:0x80000000 BX:0x0 CX:0x0 DX:0x7 SI:0xf DI:0x286 BP:0xffff95bc8213a460 SP:0xffffacbf0ba97d18 IP:0xffffffffa7bc21cd FLAGS:0x28e CS:0x10 SS:0x18 R8:0x2 R9:0x21440 R10:0x33816fb3b8c R11:0x1 R12:0xffff95bc8213a460 R13:0xffff95bc8213a400 R14:0xffff95bc8213a400 R15:0x1 ABI:2 AX:0xffffffffffffffda BX:0xffffffffffffffff CX:0x7f84ad85798b DX:0x560209699d50 SI:0x7ffe2c7a6820 DI:0x7ffe2c7a8c9b BP:0x7ffe2c7a20d0 SP:0x7ffe2c7a2058 IP:0x7f84ad85798b FLAGS:0x206 CS:0x33 SS:0x2b R8:0x7ffe2c7a2030 R9:0x7f84ae55f010 R10:0x8 R11:0x206 R12:0xffffffffffffffff R13:0xffffffffffffffff R14:0xffffffffffffffff R15:0xffffffffffffffff perf 7049 [-01] 1343.354363: 1 cycles:ppp: ... After: perf 7049 [-01] 1343.354347: 1 cycles:ppp: ffffffffa7bc21ce perf_event_exec+0x18e (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7ead3 setup_new_exec+0xf3 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7cd7be5 load_elf_binary+0x395 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7e540 search_binary_handler+0x80 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f1aa __do_execve_file.isra.13+0x58a (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f561 do_execve+0x21 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f596 __x64_sys_execve+0x26 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7a041cb do_syscall_64+0x5b (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa840008c entry_SYSCALL_64+0x7c (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ABI:2 AX:0x80000000 BX:0x0 CX:0x0 DX:0x7 SI:0xf DI:0x286 BP:0xffff95bc8213a460 SP:0xffffacbf0ba97d18 IP:0xffffffffa7bc21cd FLAGS:0x28e CS:0x10 SS:0x18 R8:0x2 R9:0x21440 R10:0x33816fb3b8c R11:0x1 R12:0xffff95bc8213a460 R13:0xffff95bc8213a400 R14:0xffff95bc8213a400 R15:0x1 ABI:2 AX:0xffffffffffffffda BX:0xffffffffffffffff CX:0x7f84ad85798b DX:0x560209699d50 SI:0x7ffe2c7a6820 DI:0x7ffe2c7a8c9b BP:0x7ffe2c7a20d0 SP:0x7ffe2c7a2058 IP:0x7f84ad85798b FLAGS:0x206 CS:0x33 SS:0x2b R8:0x7ffe2c7a2030 R9:0x7f84ae55f010 R10:0x8 R11:0x206 R12:0xffffffffffffffff R13:0xffffffffffffffff R14:0xffffffffffffffff R15:0xffffffffffffffff perf 7049 [-01] 1343.354363: 1 cycles:ppp: ... Signed-off-by: Milian Wolff <milian.wolff@kdab.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181107223437.9071-1-milian.wolff@kdab.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-11-07 22:34:37 +00:00
static int perf_sample__fprintf_regs(struct regs_dump *regs, uint64_t mask,
FILE *fp
)
{
unsigned i = 0, r;
int printed = 0;
if (!regs || !regs->regs)
return 0;
printed += fprintf(fp, " ABI:%" PRIu64 " ", regs->abi);
for_each_set_bit(r, (unsigned long *) &mask, sizeof(mask) * 8) {
u64 val = regs->regs[i++];
printed += fprintf(fp, "%5s:0x%"PRIx64" ", perf_reg_name(r), val);
}
fprintf(fp, "\n");
return printed;
}
perf script: Share code and output format for uregs and iregs output The iregs output was missing the newline at end as well as the leading ABI output. This made it hard to compare the iregs and uregs values. Instead, use a single function to output the register values and use it for both, iregs and uregs, to ensure the output is consistent. Before: perf 7049 [-01] 1343.354347: 1 cycles:ppp: ffffffffa7bc21ce perf_event_exec+0x18e (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7ead3 setup_new_exec+0xf3 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7cd7be5 load_elf_binary+0x395 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7e540 search_binary_handler+0x80 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f1aa __do_execve_file.isra.13+0x58a (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f561 do_execve+0x21 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f596 __x64_sys_execve+0x26 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7a041cb do_syscall_64+0x5b (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa840008c entry_SYSCALL_64+0x7c (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) AX:0x80000000 BX:0x0 CX:0x0 DX:0x7 SI:0xf DI:0x286 BP:0xffff95bc8213a460 SP:0xffffacbf0ba97d18 IP:0xffffffffa7bc21cd FLAGS:0x28e CS:0x10 SS:0x18 R8:0x2 R9:0x21440 R10:0x33816fb3b8c R11:0x1 R12:0xffff95bc8213a460 R13:0xffff95bc8213a400 R14:0xffff95bc8213a400 R15:0x1 ABI:2 AX:0xffffffffffffffda BX:0xffffffffffffffff CX:0x7f84ad85798b DX:0x560209699d50 SI:0x7ffe2c7a6820 DI:0x7ffe2c7a8c9b BP:0x7ffe2c7a20d0 SP:0x7ffe2c7a2058 IP:0x7f84ad85798b FLAGS:0x206 CS:0x33 SS:0x2b R8:0x7ffe2c7a2030 R9:0x7f84ae55f010 R10:0x8 R11:0x206 R12:0xffffffffffffffff R13:0xffffffffffffffff R14:0xffffffffffffffff R15:0xffffffffffffffff perf 7049 [-01] 1343.354363: 1 cycles:ppp: ... After: perf 7049 [-01] 1343.354347: 1 cycles:ppp: ffffffffa7bc21ce perf_event_exec+0x18e (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7ead3 setup_new_exec+0xf3 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7cd7be5 load_elf_binary+0x395 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7e540 search_binary_handler+0x80 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f1aa __do_execve_file.isra.13+0x58a (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f561 do_execve+0x21 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7c7f596 __x64_sys_execve+0x26 (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa7a041cb do_syscall_64+0x5b (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ffffffffa840008c entry_SYSCALL_64+0x7c (/lib/modules/4.20.0-rc1perf-devel-05115-gc0bc98f76e39-dirty/build/vmlinux) ABI:2 AX:0x80000000 BX:0x0 CX:0x0 DX:0x7 SI:0xf DI:0x286 BP:0xffff95bc8213a460 SP:0xffffacbf0ba97d18 IP:0xffffffffa7bc21cd FLAGS:0x28e CS:0x10 SS:0x18 R8:0x2 R9:0x21440 R10:0x33816fb3b8c R11:0x1 R12:0xffff95bc8213a460 R13:0xffff95bc8213a400 R14:0xffff95bc8213a400 R15:0x1 ABI:2 AX:0xffffffffffffffda BX:0xffffffffffffffff CX:0x7f84ad85798b DX:0x560209699d50 SI:0x7ffe2c7a6820 DI:0x7ffe2c7a8c9b BP:0x7ffe2c7a20d0 SP:0x7ffe2c7a2058 IP:0x7f84ad85798b FLAGS:0x206 CS:0x33 SS:0x2b R8:0x7ffe2c7a2030 R9:0x7f84ae55f010 R10:0x8 R11:0x206 R12:0xffffffffffffffff R13:0xffffffffffffffff R14:0xffffffffffffffff R15:0xffffffffffffffff perf 7049 [-01] 1343.354363: 1 cycles:ppp: ... Signed-off-by: Milian Wolff <milian.wolff@kdab.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181107223437.9071-1-milian.wolff@kdab.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-11-07 22:34:37 +00:00
static int perf_sample__fprintf_iregs(struct perf_sample *sample,
struct perf_event_attr *attr, FILE *fp)
{
return perf_sample__fprintf_regs(&sample->intr_regs,
attr->sample_regs_intr, fp);
}
static int perf_sample__fprintf_uregs(struct perf_sample *sample,
struct perf_event_attr *attr, FILE *fp)
{
return perf_sample__fprintf_regs(&sample->user_regs,
attr->sample_regs_user, fp);
}
static int perf_sample__fprintf_start(struct perf_sample *sample,
struct thread *thread,
struct perf_evsel *evsel,
u32 type, FILE *fp)
{
struct perf_event_attr *attr = &evsel->attr;
unsigned long secs;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
unsigned long long nsecs;
int printed = 0;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
if (PRINT_FIELD(COMM)) {
if (latency_format)
printed += fprintf(fp, "%8.8s ", thread__comm_str(thread));
else if (PRINT_FIELD(IP) && evsel__has_callchain(evsel) && symbol_conf.use_callchain)
printed += fprintf(fp, "%s ", thread__comm_str(thread));
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
else
printed += fprintf(fp, "%16s ", thread__comm_str(thread));
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
if (PRINT_FIELD(PID) && PRINT_FIELD(TID))
printed += fprintf(fp, "%5d/%-5d ", sample->pid, sample->tid);
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
else if (PRINT_FIELD(PID))
printed += fprintf(fp, "%5d ", sample->pid);
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
else if (PRINT_FIELD(TID))
printed += fprintf(fp, "%5d ", sample->tid);
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
if (PRINT_FIELD(CPU)) {
if (latency_format)
printed += fprintf(fp, "%3d ", sample->cpu);
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
else
printed += fprintf(fp, "[%03d] ", sample->cpu);
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
if (PRINT_FIELD(MISC)) {
int ret = 0;
#define has(m) \
(sample->misc & PERF_RECORD_MISC_##m) == PERF_RECORD_MISC_##m
if (has(KERNEL))
ret += fprintf(fp, "K");
if (has(USER))
ret += fprintf(fp, "U");
if (has(HYPERVISOR))
ret += fprintf(fp, "H");
if (has(GUEST_KERNEL))
ret += fprintf(fp, "G");
if (has(GUEST_USER))
ret += fprintf(fp, "g");
switch (type) {
case PERF_RECORD_MMAP:
case PERF_RECORD_MMAP2:
if (has(MMAP_DATA))
ret += fprintf(fp, "M");
break;
case PERF_RECORD_COMM:
if (has(COMM_EXEC))
ret += fprintf(fp, "E");
break;
case PERF_RECORD_SWITCH:
case PERF_RECORD_SWITCH_CPU_WIDE:
perf script: Extend misc field decoding with switch out event type Append 'p' sign to 'S' tag designating the type of context switch out event so 'Sp' means preemption context switch. Documentation is extended to cover new presentation changes. $ perf script --show-switch-events -F +misc -I -i perf.data: hdparm 4073 [004] U 762.198265: 380194 cycles:ppp: 7faf727f5a23 strchr (/usr/lib64/ld-2.26.so) hdparm 4073 [004] K 762.198366: 441572 cycles:ppp: ffffffffb9218435 alloc_set_pte (/lib/modules/4.16.0-rc6+/build/vmlinux) hdparm 4073 [004] S 762.198391: PERF_RECORD_SWITCH_CPU_WIDE OUT next pid/tid: 0/0 swapper 0 [004] 762.198392: PERF_RECORD_SWITCH_CPU_WIDE IN prev pid/tid: 4073/4073 swapper 0 [004] Sp 762.198477: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt next pid/tid: 4073/4073 hdparm 4073 [004] 762.198478: PERF_RECORD_SWITCH_CPU_WIDE IN prev pid/tid: 0/0 swapper 0 [007] K 762.198514: 2303073 cycles:ppp: ffffffffb98b0c66 intel_idle (/lib/modules/4.16.0-rc6+/build/vmlinux) swapper 0 [007] Sp 762.198561: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt next pid/tid: 1134/1134 kworker/u16:18 1134 [007] 762.198562: PERF_RECORD_SWITCH_CPU_WIDE IN prev pid/tid: 0/0 kworker/u16:18 1134 [007] S 762.198567: PERF_RECORD_SWITCH_CPU_WIDE OUT next pid/tid: 0/0 Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/5fc65ce7-8ca5-53ae-8858-8ddd27290575@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-04-09 07:26:46 +00:00
if (has(SWITCH_OUT)) {
ret += fprintf(fp, "S");
perf script: Extend misc field decoding with switch out event type Append 'p' sign to 'S' tag designating the type of context switch out event so 'Sp' means preemption context switch. Documentation is extended to cover new presentation changes. $ perf script --show-switch-events -F +misc -I -i perf.data: hdparm 4073 [004] U 762.198265: 380194 cycles:ppp: 7faf727f5a23 strchr (/usr/lib64/ld-2.26.so) hdparm 4073 [004] K 762.198366: 441572 cycles:ppp: ffffffffb9218435 alloc_set_pte (/lib/modules/4.16.0-rc6+/build/vmlinux) hdparm 4073 [004] S 762.198391: PERF_RECORD_SWITCH_CPU_WIDE OUT next pid/tid: 0/0 swapper 0 [004] 762.198392: PERF_RECORD_SWITCH_CPU_WIDE IN prev pid/tid: 4073/4073 swapper 0 [004] Sp 762.198477: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt next pid/tid: 4073/4073 hdparm 4073 [004] 762.198478: PERF_RECORD_SWITCH_CPU_WIDE IN prev pid/tid: 0/0 swapper 0 [007] K 762.198514: 2303073 cycles:ppp: ffffffffb98b0c66 intel_idle (/lib/modules/4.16.0-rc6+/build/vmlinux) swapper 0 [007] Sp 762.198561: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt next pid/tid: 1134/1134 kworker/u16:18 1134 [007] 762.198562: PERF_RECORD_SWITCH_CPU_WIDE IN prev pid/tid: 0/0 kworker/u16:18 1134 [007] S 762.198567: PERF_RECORD_SWITCH_CPU_WIDE OUT next pid/tid: 0/0 Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/5fc65ce7-8ca5-53ae-8858-8ddd27290575@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-04-09 07:26:46 +00:00
if (sample->misc & PERF_RECORD_MISC_SWITCH_OUT_PREEMPT)
ret += fprintf(fp, "p");
}
default:
break;
}
#undef has
ret += fprintf(fp, "%*s", 6 - ret, " ");
printed += ret;
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
if (PRINT_FIELD(TIME)) {
nsecs = sample->time;
secs = nsecs / NSEC_PER_SEC;
nsecs -= secs * NSEC_PER_SEC;
if (nanosecs)
printed += fprintf(fp, "%5lu.%09llu: ", secs, nsecs);
else {
char sample_time[32];
timestamp__scnprintf_usec(sample->time, sample_time, sizeof(sample_time));
printed += fprintf(fp, "%12s: ", sample_time);
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
return printed;
}
static inline char
mispred_str(struct branch_entry *br)
{
if (!(br->flags.mispred || br->flags.predicted))
return '-';
return br->flags.predicted ? 'P' : 'M';
}
static int perf_sample__fprintf_brstack(struct perf_sample *sample,
struct thread *thread,
struct perf_event_attr *attr, FILE *fp)
{
struct branch_stack *br = sample->branch_stack;
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
struct addr_location alf, alt;
u64 i, from, to;
int printed = 0;
if (!(br && br->nr))
return 0;
for (i = 0; i < br->nr; i++) {
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
from = br->entries[i].from;
to = br->entries[i].to;
if (PRINT_FIELD(DSO)) {
memset(&alf, 0, sizeof(alf));
memset(&alt, 0, sizeof(alt));
thread__find_map_fb(thread, sample->cpumode, from, &alf);
thread__find_map_fb(thread, sample->cpumode, to, &alt);
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
}
printed += fprintf(fp, " 0x%"PRIx64, from);
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, "(");
printed += map__fprintf_dsoname(alf.map, fp);
printed += fprintf(fp, ")");
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
}
printed += fprintf(fp, "/0x%"PRIx64, to);
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, "(");
printed += map__fprintf_dsoname(alt.map, fp);
printed += fprintf(fp, ")");
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
}
printed += fprintf(fp, "/%c/%c/%c/%d ",
mispred_str( br->entries + i),
br->entries[i].flags.in_tx? 'X' : '-',
br->entries[i].flags.abort? 'A' : '-',
br->entries[i].flags.cycles);
}
return printed;
}
static int perf_sample__fprintf_brstacksym(struct perf_sample *sample,
struct thread *thread,
struct perf_event_attr *attr, FILE *fp)
{
struct branch_stack *br = sample->branch_stack;
struct addr_location alf, alt;
u64 i, from, to;
int printed = 0;
if (!(br && br->nr))
return 0;
for (i = 0; i < br->nr; i++) {
memset(&alf, 0, sizeof(alf));
memset(&alt, 0, sizeof(alt));
from = br->entries[i].from;
to = br->entries[i].to;
thread__find_symbol_fb(thread, sample->cpumode, from, &alf);
thread__find_symbol_fb(thread, sample->cpumode, to, &alt);
printed += symbol__fprintf_symname_offs(alf.sym, &alf, fp);
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, "(");
printed += map__fprintf_dsoname(alf.map, fp);
printed += fprintf(fp, ")");
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
}
printed += fprintf(fp, "%c", '/');
printed += symbol__fprintf_symname_offs(alt.sym, &alt, fp);
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, "(");
printed += map__fprintf_dsoname(alt.map, fp);
printed += fprintf(fp, ")");
perf script: Support -F brstack,dso and brstacksym,dso Perf script can report the dso for "addr" and "ip" fields. This adds the same support for the "brstack" and "brstacksym" fields. This can be helpful for AutoFDO: we can ignore LBR entries unless the source and target address are both in the target module we are about to build. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution with perf record -b. Using the new perf script functionality I can easily find cases where there was a transition from one dso to another: $ perf record -a -b -- sleep 5 [ perf record: Woken up 55 times to write data ] [ perf record: Captured and wrote 18.815 MB perf.data (43593 samples) ] $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-1-marksan@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:24 +00:00
}
printed += fprintf(fp, "/%c/%c/%c/%d ",
mispred_str( br->entries + i),
br->entries[i].flags.in_tx? 'X' : '-',
br->entries[i].flags.abort? 'A' : '-',
br->entries[i].flags.cycles);
}
return printed;
}
static int perf_sample__fprintf_brstackoff(struct perf_sample *sample,
struct thread *thread,
struct perf_event_attr *attr, FILE *fp)
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
{
struct branch_stack *br = sample->branch_stack;
struct addr_location alf, alt;
u64 i, from, to;
int printed = 0;
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
if (!(br && br->nr))
return 0;
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
for (i = 0; i < br->nr; i++) {
memset(&alf, 0, sizeof(alf));
memset(&alt, 0, sizeof(alt));
from = br->entries[i].from;
to = br->entries[i].to;
if (thread__find_map_fb(thread, sample->cpumode, from, &alf) &&
!alf.map->dso->adjust_symbols)
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
from = map__map_ip(alf.map, from);
if (thread__find_map_fb(thread, sample->cpumode, to, &alt) &&
!alt.map->dso->adjust_symbols)
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
to = map__map_ip(alt.map, to);
printed += fprintf(fp, " 0x%"PRIx64, from);
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, "(");
printed += map__fprintf_dsoname(alf.map, fp);
printed += fprintf(fp, ")");
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
}
printed += fprintf(fp, "/0x%"PRIx64, to);
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, "(");
printed += map__fprintf_dsoname(alt.map, fp);
printed += fprintf(fp, ")");
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
}
printed += fprintf(fp, "/%c/%c/%c/%d ",
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
mispred_str(br->entries + i),
br->entries[i].flags.in_tx ? 'X' : '-',
br->entries[i].flags.abort ? 'A' : '-',
br->entries[i].flags.cycles);
}
return printed;
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
}
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
#define MAXBB 16384UL
static int grab_bb(u8 *buffer, u64 start, u64 end,
struct machine *machine, struct thread *thread,
bool *is64bit, u8 *cpumode, bool last)
{
long offset, len;
struct addr_location al;
bool kernel;
if (!start || !end)
return 0;
kernel = machine__kernel_ip(machine, start);
if (kernel)
*cpumode = PERF_RECORD_MISC_KERNEL;
else
*cpumode = PERF_RECORD_MISC_USER;
/*
* Block overlaps between kernel and user.
* This can happen due to ring filtering
* On Intel CPUs the entry into the kernel is filtered,
* but the exit is not. Let the caller patch it up.
*/
if (kernel != machine__kernel_ip(machine, end)) {
pr_debug("\tblock %" PRIx64 "-%" PRIx64 " transfers between kernel and user\n", start, end);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
return -ENXIO;
}
memset(&al, 0, sizeof(al));
if (end - start > MAXBB - MAXINSN) {
if (last)
pr_debug("\tbrstack does not reach to final jump (%" PRIx64 "-%" PRIx64 ")\n", start, end);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
else
pr_debug("\tblock %" PRIx64 "-%" PRIx64 " (%" PRIu64 ") too long to dump\n", start, end, end - start);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
return 0;
}
if (!thread__find_map(thread, *cpumode, start, &al) || !al.map->dso) {
pr_debug("\tcannot resolve %" PRIx64 "-%" PRIx64 "\n", start, end);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
return 0;
}
if (al.map->dso->data.status == DSO_DATA_STATUS_ERROR) {
pr_debug("\tcannot resolve %" PRIx64 "-%" PRIx64 "\n", start, end);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
return 0;
}
/* Load maps to ensure dso->is_64_bit has been updated */
map__load(al.map);
offset = al.map->map_ip(al.map, start);
len = dso__data_read_offset(al.map->dso, machine, offset, (u8 *)buffer,
end - start + MAXINSN);
*is64bit = al.map->dso->is_64_bit;
if (len <= 0)
pr_debug("\tcannot fetch code for block at %" PRIx64 "-%" PRIx64 "\n",
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
start, end);
return len;
}
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
static int print_srccode(struct thread *thread, u8 cpumode, uint64_t addr)
{
struct addr_location al;
int ret = 0;
memset(&al, 0, sizeof(al));
thread__find_map(thread, cpumode, addr, &al);
if (!al.map)
return 0;
ret = map__fprintf_srccode(al.map, al.addr, stdout,
&thread->srccode_state);
if (ret)
ret += printf("\n");
return ret;
}
static int ip__fprintf_jump(uint64_t ip, struct branch_entry *en,
struct perf_insn *x, u8 *inbuf, int len,
int insn, FILE *fp, int *total_cycles)
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
{
int printed = fprintf(fp, "\t%016" PRIx64 "\t%-30s\t#%s%s%s%s", ip,
dump_insn(x, ip, inbuf, len, NULL),
en->flags.predicted ? " PRED" : "",
en->flags.mispred ? " MISPRED" : "",
en->flags.in_tx ? " INTX" : "",
en->flags.abort ? " ABORT" : "");
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (en->flags.cycles) {
*total_cycles += en->flags.cycles;
printed += fprintf(fp, " %d cycles [%d]", en->flags.cycles, *total_cycles);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (insn)
printed += fprintf(fp, " %.2f IPC", (float)insn / en->flags.cycles);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
return printed + fprintf(fp, "\n");
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
static int ip__fprintf_sym(uint64_t addr, struct thread *thread,
u8 cpumode, int cpu, struct symbol **lastsym,
struct perf_event_attr *attr, FILE *fp)
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
{
struct addr_location al;
int off, printed = 0;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
memset(&al, 0, sizeof(al));
thread__find_map(thread, cpumode, addr, &al);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if ((*lastsym) && al.addr >= (*lastsym)->start && al.addr < (*lastsym)->end)
return 0;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
al.cpu = cpu;
al.sym = NULL;
if (al.map)
al.sym = map__find_symbol(al.map, al.addr);
if (!al.sym)
return 0;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (al.addr < al.sym->end)
off = al.addr - al.sym->start;
else
off = al.addr - al.map->start - al.sym->start;
printed += fprintf(fp, "\t%s", al.sym->name);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (off)
printed += fprintf(fp, "%+d", off);
printed += fprintf(fp, ":");
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (PRINT_FIELD(SRCLINE))
printed += map__fprintf_srcline(al.map, al.addr, "\t", fp);
printed += fprintf(fp, "\n");
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
*lastsym = al.sym;
return printed;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
static int perf_sample__fprintf_brstackinsn(struct perf_sample *sample,
struct thread *thread,
struct perf_event_attr *attr,
struct machine *machine, FILE *fp)
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
{
struct branch_stack *br = sample->branch_stack;
u64 start, end;
int i, insn, len, nr, ilen, printed = 0;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
struct perf_insn x;
u8 buffer[MAXBB];
unsigned off;
struct symbol *lastsym = NULL;
int total_cycles = 0;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (!(br && br->nr))
return 0;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
nr = br->nr;
if (max_blocks && nr > max_blocks + 1)
nr = max_blocks + 1;
x.thread = thread;
x.cpu = sample->cpu;
printed += fprintf(fp, "%c", '\n');
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
/* Handle first from jump, of which we don't know the entry. */
len = grab_bb(buffer, br->entries[nr-1].from,
br->entries[nr-1].from,
machine, thread, &x.is64bit, &x.cpumode, false);
if (len > 0) {
printed += ip__fprintf_sym(br->entries[nr - 1].from, thread,
x.cpumode, x.cpu, &lastsym, attr, fp);
printed += ip__fprintf_jump(br->entries[nr - 1].from, &br->entries[nr - 1],
&x, buffer, len, 0, fp, &total_cycles);
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if (PRINT_FIELD(SRCCODE))
printed += print_srccode(thread, x.cpumode, br->entries[nr - 1].from);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
/* Print all blocks */
for (i = nr - 2; i >= 0; i--) {
if (br->entries[i].from || br->entries[i].to)
pr_debug("%d: %" PRIx64 "-%" PRIx64 "\n", i,
br->entries[i].from,
br->entries[i].to);
start = br->entries[i + 1].to;
end = br->entries[i].from;
len = grab_bb(buffer, start, end, machine, thread, &x.is64bit, &x.cpumode, false);
/* Patch up missing kernel transfers due to ring filters */
if (len == -ENXIO && i > 0) {
end = br->entries[--i].from;
pr_debug("\tpatching up to %" PRIx64 "-%" PRIx64 "\n", start, end);
len = grab_bb(buffer, start, end, machine, thread, &x.is64bit, &x.cpumode, false);
}
if (len <= 0)
continue;
insn = 0;
for (off = 0;; off += ilen) {
uint64_t ip = start + off;
printed += ip__fprintf_sym(ip, thread, x.cpumode, x.cpu, &lastsym, attr, fp);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (ip == end) {
printed += ip__fprintf_jump(ip, &br->entries[i], &x, buffer + off, len - off, insn, fp,
&total_cycles);
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if (PRINT_FIELD(SRCCODE))
printed += print_srccode(thread, x.cpumode, ip);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
break;
} else {
printed += fprintf(fp, "\t%016" PRIx64 "\t%s\n", ip,
dump_insn(&x, ip, buffer + off, len - off, &ilen));
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (ilen == 0)
break;
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if (PRINT_FIELD(SRCCODE))
print_srccode(thread, x.cpumode, ip);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
insn++;
}
}
}
/*
* Hit the branch? In this case we are already done, and the target
* has not been executed yet.
*/
if (br->entries[0].from == sample->ip)
goto out;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (br->entries[0].flags.abort)
goto out;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
/*
* Print final block upto sample
*/
start = br->entries[0].to;
end = sample->ip;
len = grab_bb(buffer, start, end, machine, thread, &x.is64bit, &x.cpumode, true);
printed += ip__fprintf_sym(start, thread, x.cpumode, x.cpu, &lastsym, attr, fp);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (len <= 0) {
/* Print at least last IP if basic block did not work */
len = grab_bb(buffer, sample->ip, sample->ip,
machine, thread, &x.is64bit, &x.cpumode, false);
if (len <= 0)
goto out;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
printed += fprintf(fp, "\t%016" PRIx64 "\t%s\n", sample->ip,
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
dump_insn(&x, sample->ip, buffer, len, NULL));
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if (PRINT_FIELD(SRCCODE))
print_srccode(thread, x.cpumode, sample->ip);
goto out;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
for (off = 0; off <= end - start; off += ilen) {
printed += fprintf(fp, "\t%016" PRIx64 "\t%s\n", start + off,
dump_insn(&x, start + off, buffer + off, len - off, &ilen));
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (ilen == 0)
break;
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if (PRINT_FIELD(SRCCODE))
print_srccode(thread, x.cpumode, start + off);
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
out:
return printed;
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
}
static int perf_sample__fprintf_addr(struct perf_sample *sample,
struct thread *thread,
struct perf_event_attr *attr, FILE *fp)
{
struct addr_location al;
int printed = fprintf(fp, "%16" PRIx64, sample->addr);
if (!sample_addr_correlates_sym(attr))
goto out;
thread__resolve(thread, &al, sample);
if (PRINT_FIELD(SYM)) {
printed += fprintf(fp, " ");
if (PRINT_FIELD(SYMOFFSET))
printed += symbol__fprintf_symname_offs(al.sym, &al, fp);
else
printed += symbol__fprintf_symname(al.sym, fp);
}
if (PRINT_FIELD(DSO)) {
printed += fprintf(fp, " (");
printed += map__fprintf_dsoname(al.map, fp);
printed += fprintf(fp, ")");
}
out:
return printed;
}
perf script: Implement --graph-function Add a ftrace style --graph-function argument to 'perf script' that allows to print itrace function calls only below a given function. This makes it easier to find the code of interest in a large trace. % perf record -e intel_pt//k -a sleep 1 % perf script --graph-function group_sched_in --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) calc_timer_values perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_cpu perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) arch_perf_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __fentry__ perf 900 [000] 194167.205652203: ([kernel.kallsyms]) using_native_sched_clock perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_stable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) group_sched_in swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_set_state.part.71 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_time swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_pmu_disable swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_log_itrace_start swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) calc_timer_values swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_cpu swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) arch_perf_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __fentry__ swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) using_native_sched_clock swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_stable Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:39 +00:00
static const char *resolve_branch_sym(struct perf_sample *sample,
struct perf_evsel *evsel,
struct thread *thread,
struct addr_location *al,
u64 *ip)
{
struct addr_location addr_al;
struct perf_event_attr *attr = &evsel->attr;
const char *name = NULL;
if (sample->flags & (PERF_IP_FLAG_CALL | PERF_IP_FLAG_TRACE_BEGIN)) {
if (sample_addr_correlates_sym(attr)) {
thread__resolve(thread, &addr_al, sample);
if (addr_al.sym)
name = addr_al.sym->name;
else
*ip = sample->addr;
} else {
*ip = sample->addr;
}
} else if (sample->flags & (PERF_IP_FLAG_RETURN | PERF_IP_FLAG_TRACE_END)) {
if (al->sym)
name = al->sym->name;
else
*ip = sample->ip;
}
return name;
}
static int perf_sample__fprintf_callindent(struct perf_sample *sample,
struct perf_evsel *evsel,
struct thread *thread,
struct addr_location *al, FILE *fp)
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
{
struct perf_event_attr *attr = &evsel->attr;
size_t depth = thread_stack__depth(thread);
const char *name = NULL;
static int spacing;
int len = 0;
perf script: Print DSO for callindent Now that we don't need to print the IP/ADDR for callindent the DSO is also not printed. It's useful for some cases, so add an own DSO printout for callindent for the case when IP/ADDR is not enabled. Before: % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff,-addr swapper 0 [000] 3377.917072: 1 branches: pt_config swapper 0 [000] 3377.917072: 1 branches: pt_config swapper 0 [000] 3377.917072: 1 branches: pt_event_add swapper 0 [000] 3377.917072: 1 branches: perf_pmu_enable swapper 0 [000] 3377.917072: 1 branches: perf_pmu_nop_void swapper 0 [000] 3377.917072: 1 branches: event_sched_in.isra.107 swapper 0 [000] 3377.917072: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: perf_pmu_nop_int swapper 0 [000] 3377.917072: 1 branches: group_sched_in swapper 0 [000] 3377.917072: 1 branches: event_filter_match swapper 0 [000] 3377.917072: 1 branches: event_filter_match swapper 0 [000] 3377.917072: 1 branches: group_sched_in After: swapper 0 [000] 3377.917072: 1 branches: ([unknown]) pt_config swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) pt_config swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) pt_event_add swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_enable swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_void swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_int swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) group_sched_in swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_filter_match swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_filter_match swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) group_sched_in swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_txn swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_sched_in.isra.107 (in the kernel case of course it's not very useful, but it's important with user programs where symbols are not unique) Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-6-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:10 +00:00
int dlen = 0;
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
u64 ip = 0;
/*
* The 'return' has already been popped off the stack so the depth has
* to be adjusted to match the 'call'.
*/
if (thread->ts && sample->flags & PERF_IP_FLAG_RETURN)
depth += 1;
perf script: Implement --graph-function Add a ftrace style --graph-function argument to 'perf script' that allows to print itrace function calls only below a given function. This makes it easier to find the code of interest in a large trace. % perf record -e intel_pt//k -a sleep 1 % perf script --graph-function group_sched_in --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) calc_timer_values perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_cpu perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) arch_perf_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __fentry__ perf 900 [000] 194167.205652203: ([kernel.kallsyms]) using_native_sched_clock perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_stable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) group_sched_in swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_set_state.part.71 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_time swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_pmu_disable swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_log_itrace_start swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) calc_timer_values swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_cpu swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) arch_perf_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __fentry__ swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) using_native_sched_clock swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_stable Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:39 +00:00
name = resolve_branch_sym(sample, evsel, thread, al, &ip);
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
perf script: Print DSO for callindent Now that we don't need to print the IP/ADDR for callindent the DSO is also not printed. It's useful for some cases, so add an own DSO printout for callindent for the case when IP/ADDR is not enabled. Before: % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff,-addr swapper 0 [000] 3377.917072: 1 branches: pt_config swapper 0 [000] 3377.917072: 1 branches: pt_config swapper 0 [000] 3377.917072: 1 branches: pt_event_add swapper 0 [000] 3377.917072: 1 branches: perf_pmu_enable swapper 0 [000] 3377.917072: 1 branches: perf_pmu_nop_void swapper 0 [000] 3377.917072: 1 branches: event_sched_in.isra.107 swapper 0 [000] 3377.917072: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: perf_pmu_nop_int swapper 0 [000] 3377.917072: 1 branches: group_sched_in swapper 0 [000] 3377.917072: 1 branches: event_filter_match swapper 0 [000] 3377.917072: 1 branches: event_filter_match swapper 0 [000] 3377.917072: 1 branches: group_sched_in After: swapper 0 [000] 3377.917072: 1 branches: ([unknown]) pt_config swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) pt_config swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) pt_event_add swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_enable swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_void swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_int swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) group_sched_in swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_filter_match swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_filter_match swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) group_sched_in swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_txn swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_sched_in.isra.107 (in the kernel case of course it's not very useful, but it's important with user programs where symbols are not unique) Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-6-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:10 +00:00
if (PRINT_FIELD(DSO) && !(PRINT_FIELD(IP) || PRINT_FIELD(ADDR))) {
dlen += fprintf(fp, "(");
dlen += map__fprintf_dsoname(al->map, fp);
dlen += fprintf(fp, ")\t");
}
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
if (name)
len = fprintf(fp, "%*s%s", (int)depth * 4, "", name);
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
else if (ip)
len = fprintf(fp, "%*s%16" PRIx64, (int)depth * 4, "", ip);
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
if (len < 0)
return len;
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
/*
* Try to keep the output length from changing frequently so that the
* output lines up more nicely.
*/
if (len > spacing || (len && len < spacing - 52))
spacing = round_up(len + 4, 32);
if (len < spacing)
len += fprintf(fp, "%*s", spacing - len, "");
perf script: Print DSO for callindent Now that we don't need to print the IP/ADDR for callindent the DSO is also not printed. It's useful for some cases, so add an own DSO printout for callindent for the case when IP/ADDR is not enabled. Before: % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff,-addr swapper 0 [000] 3377.917072: 1 branches: pt_config swapper 0 [000] 3377.917072: 1 branches: pt_config swapper 0 [000] 3377.917072: 1 branches: pt_event_add swapper 0 [000] 3377.917072: 1 branches: perf_pmu_enable swapper 0 [000] 3377.917072: 1 branches: perf_pmu_nop_void swapper 0 [000] 3377.917072: 1 branches: event_sched_in.isra.107 swapper 0 [000] 3377.917072: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: perf_pmu_nop_int swapper 0 [000] 3377.917072: 1 branches: group_sched_in swapper 0 [000] 3377.917072: 1 branches: event_filter_match swapper 0 [000] 3377.917072: 1 branches: event_filter_match swapper 0 [000] 3377.917072: 1 branches: group_sched_in After: swapper 0 [000] 3377.917072: 1 branches: ([unknown]) pt_config swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) pt_config swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) pt_event_add swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_enable swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_void swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_int swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) group_sched_in swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_filter_match swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_filter_match swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) group_sched_in swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) perf_pmu_nop_txn swapper 0 [000] 3377.917072: 1 branches: ([kernel.kallsyms]) event_sched_in.isra.107 (in the kernel case of course it's not very useful, but it's important with user programs where symbols are not unique) Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-6-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:10 +00:00
return len + dlen;
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
}
static int perf_sample__fprintf_insn(struct perf_sample *sample,
struct perf_event_attr *attr,
struct thread *thread,
struct machine *machine, FILE *fp)
{
int printed = 0;
if (PRINT_FIELD(INSNLEN))
printed += fprintf(fp, " ilen: %d", sample->insn_len);
if (PRINT_FIELD(INSN)) {
int i;
printed += fprintf(fp, " insn:");
for (i = 0; i < sample->insn_len; i++)
printed += fprintf(fp, " %02x", (unsigned char)sample->insn[i]);
}
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
if (PRINT_FIELD(BRSTACKINSN))
printed += perf_sample__fprintf_brstackinsn(sample, thread, attr, machine, fp);
return printed;
}
static int perf_sample__fprintf_bts(struct perf_sample *sample,
struct perf_evsel *evsel,
struct thread *thread,
struct addr_location *al,
struct machine *machine, FILE *fp)
{
struct perf_event_attr *attr = &evsel->attr;
unsigned int type = output_type(attr->type);
bool print_srcline_last = false;
int printed = 0;
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
if (PRINT_FIELD(CALLINDENT))
printed += perf_sample__fprintf_callindent(sample, evsel, thread, al, fp);
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
/* print branch_from information */
if (PRINT_FIELD(IP)) {
unsigned int print_opts = output[type].print_ip_opts;
perf script: Fix segfault when printing callchains This fixes a bug caused by an unitialized callchain cursor. The crash frist appeared in: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") The callchain cursor is a struct that contains pointers, that when uninitialized will cause unpredictable behavior (usually a crash) when trying to append to the callchain. The existing implementation has the following issues: 1. The callchain cursor used is not initialized, resulting in unpredictable behavior when used. 2. The cursor is declared on the stack. Even if it is properly initalized, the implmentation will leak memory when the function returns, since all the references to the callchain_nodes allocated by callchain_cursor_append will be lost when the cursor goes out of scope. 3. Storing the cursor on the stack is inefficient. Even if memory is properly freed when it goes out of scope, a performance penalty will be incurred due to reallocation of callchain nodes. callchain_cursor_append is designed to avoid these reallocations when an existing cursor is reused. This patch fixes the crash by replacing cursor_callchain with a reference to the global callchain_cursor which also resolves all 3 issues mentioned above. How to reproduce the crash: $ perf record --call-graph=dwarf stress -t 1 -c 1 $ perf script > /dev/null Segfault Signed-off-by: Chris Phlipot <cphlipot0@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-20 02:32:11 +00:00
struct callchain_cursor *cursor = NULL;
if (symbol_conf.use_callchain && sample->callchain &&
perf script: Fix segfault when printing callchains This fixes a bug caused by an unitialized callchain cursor. The crash frist appeared in: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") The callchain cursor is a struct that contains pointers, that when uninitialized will cause unpredictable behavior (usually a crash) when trying to append to the callchain. The existing implementation has the following issues: 1. The callchain cursor used is not initialized, resulting in unpredictable behavior when used. 2. The cursor is declared on the stack. Even if it is properly initalized, the implmentation will leak memory when the function returns, since all the references to the callchain_nodes allocated by callchain_cursor_append will be lost when the cursor goes out of scope. 3. Storing the cursor on the stack is inefficient. Even if memory is properly freed when it goes out of scope, a performance penalty will be incurred due to reallocation of callchain nodes. callchain_cursor_append is designed to avoid these reallocations when an existing cursor is reused. This patch fixes the crash by replacing cursor_callchain with a reference to the global callchain_cursor which also resolves all 3 issues mentioned above. How to reproduce the crash: $ perf record --call-graph=dwarf stress -t 1 -c 1 $ perf script > /dev/null Segfault Signed-off-by: Chris Phlipot <cphlipot0@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-20 02:32:11 +00:00
thread__resolve_callchain(al->thread, &callchain_cursor, evsel,
sample, NULL, NULL, scripting_max_stack) == 0)
perf script: Fix segfault when printing callchains This fixes a bug caused by an unitialized callchain cursor. The crash frist appeared in: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") The callchain cursor is a struct that contains pointers, that when uninitialized will cause unpredictable behavior (usually a crash) when trying to append to the callchain. The existing implementation has the following issues: 1. The callchain cursor used is not initialized, resulting in unpredictable behavior when used. 2. The cursor is declared on the stack. Even if it is properly initalized, the implmentation will leak memory when the function returns, since all the references to the callchain_nodes allocated by callchain_cursor_append will be lost when the cursor goes out of scope. 3. Storing the cursor on the stack is inefficient. Even if memory is properly freed when it goes out of scope, a performance penalty will be incurred due to reallocation of callchain nodes. callchain_cursor_append is designed to avoid these reallocations when an existing cursor is reused. This patch fixes the crash by replacing cursor_callchain with a reference to the global callchain_cursor which also resolves all 3 issues mentioned above. How to reproduce the crash: $ perf record --call-graph=dwarf stress -t 1 -c 1 $ perf script > /dev/null Segfault Signed-off-by: Chris Phlipot <cphlipot0@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-20 02:32:11 +00:00
cursor = &callchain_cursor;
if (cursor == NULL) {
printed += fprintf(fp, " ");
if (print_opts & EVSEL__PRINT_SRCLINE) {
print_srcline_last = true;
print_opts &= ~EVSEL__PRINT_SRCLINE;
}
} else
printed += fprintf(fp, "\n");
printed += sample__fprintf_sym(sample, al, 0, print_opts, cursor, fp);
}
/* print branch_to information */
if (PRINT_FIELD(ADDR) ||
((evsel->attr.sample_type & PERF_SAMPLE_ADDR) &&
!output[type].user_set)) {
printed += fprintf(fp, " => ");
printed += perf_sample__fprintf_addr(sample, thread, attr, fp);
}
if (print_srcline_last)
printed += map__fprintf_srcline(al->map, al->addr, "\n ", fp);
printed += perf_sample__fprintf_insn(sample, attr, thread, machine, fp);
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
printed += fprintf(fp, "\n");
if (PRINT_FIELD(SRCCODE)) {
int ret = map__fprintf_srccode(al->map, al->addr, stdout,
&thread->srccode_state);
if (ret) {
printed += ret;
printed += printf("\n");
}
}
return printed;
}
perf script: Print sample flags more nicely The flags field is synthesized and may have a value when Instruction Trace decoding. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Change the display so that known combinations of flags are printed more nicely e.g.: "call" for "bc", "return" for "br", "jcc" for "bo", "jmp" for "b", "int" for "bci", "iret" for "bri", "syscall" for "bcs", "sysret" for "brs", "async" for "by", "hw int" for "bcyi", "tx abrt" for "bA", "tr strt" for "bB", "tr end" for "bE". However the "x" flag will be displayed separately in those cases e.g. "jcc (x)" for a condition branch within a transaction. Example: perf record -e intel_pt//u ls perf script --ns -F comm,cpu,pid,tid,time,ip,addr,sym,dso,symoff,flags ... ls 3689/3689 [001] 2062.020965237: jcc 7f06a958847a _dl_sysdep_start+0xfa (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588450 _dl_sysdep_start+0xd0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a9588461 _dl_sysdep_start+0xe1 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885a0 _dl_sysdep_start+0x220 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a95885a4 _dl_sysdep_start+0x224 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588470 _dl_sysdep_start+0xf0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: call 7f06a95884c3 _dl_sysdep_start+0x143 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9589140 brk+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: syscall 7f06a958914a brk+0xa (/lib/x86_64-linux-gnu/ld-2.19.so) => 0 [unknown] ([unknown]) ls 3689/3689 [001] 2062.020966237: tr strt 0 [unknown] ([unknown]) => 7f06a958914c brk+0xc (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: return 7f06a9589165 brk+0x25 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95884c8 _dl_sysdep_start+0x148 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a95884d7 _dl_sysdep_start+0x157 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: call 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac50 strlen+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a958ac6e strlen+0x1e (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac60 strlen+0x10 (/lib/x86_64-linux-gnu/ld-2.19.so) ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:56 +00:00
static struct {
u32 flags;
const char *name;
} sample_flags[] = {
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL, "call"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN, "return"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CONDITIONAL, "jcc"},
{PERF_IP_FLAG_BRANCH, "jmp"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_INTERRUPT, "int"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN | PERF_IP_FLAG_INTERRUPT, "iret"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_SYSCALLRET, "syscall"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN | PERF_IP_FLAG_SYSCALLRET, "sysret"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_ASYNC, "async"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_ASYNC | PERF_IP_FLAG_INTERRUPT, "hw int"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TX_ABORT, "tx abrt"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TRACE_BEGIN, "tr strt"},
{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TRACE_END, "tr end"},
{0, NULL}
};
static const char *sample_flags_to_name(u32 flags)
{
int i;
for (i = 0; sample_flags[i].name ; i++) {
if (sample_flags[i].flags == flags)
return sample_flags[i].name;
}
return NULL;
}
static int perf_sample__fprintf_flags(u32 flags, FILE *fp)
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
{
const char *chars = PERF_IP_FLAG_CHARS;
const int n = strlen(PERF_IP_FLAG_CHARS);
perf script: Print sample flags more nicely The flags field is synthesized and may have a value when Instruction Trace decoding. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Change the display so that known combinations of flags are printed more nicely e.g.: "call" for "bc", "return" for "br", "jcc" for "bo", "jmp" for "b", "int" for "bci", "iret" for "bri", "syscall" for "bcs", "sysret" for "brs", "async" for "by", "hw int" for "bcyi", "tx abrt" for "bA", "tr strt" for "bB", "tr end" for "bE". However the "x" flag will be displayed separately in those cases e.g. "jcc (x)" for a condition branch within a transaction. Example: perf record -e intel_pt//u ls perf script --ns -F comm,cpu,pid,tid,time,ip,addr,sym,dso,symoff,flags ... ls 3689/3689 [001] 2062.020965237: jcc 7f06a958847a _dl_sysdep_start+0xfa (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588450 _dl_sysdep_start+0xd0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a9588461 _dl_sysdep_start+0xe1 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885a0 _dl_sysdep_start+0x220 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a95885a4 _dl_sysdep_start+0x224 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588470 _dl_sysdep_start+0xf0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: call 7f06a95884c3 _dl_sysdep_start+0x143 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9589140 brk+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: syscall 7f06a958914a brk+0xa (/lib/x86_64-linux-gnu/ld-2.19.so) => 0 [unknown] ([unknown]) ls 3689/3689 [001] 2062.020966237: tr strt 0 [unknown] ([unknown]) => 7f06a958914c brk+0xc (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: return 7f06a9589165 brk+0x25 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95884c8 _dl_sysdep_start+0x148 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a95884d7 _dl_sysdep_start+0x157 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: call 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac50 strlen+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a958ac6e strlen+0x1e (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac60 strlen+0x10 (/lib/x86_64-linux-gnu/ld-2.19.so) ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:56 +00:00
bool in_tx = flags & PERF_IP_FLAG_IN_TX;
const char *name = NULL;
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
char str[33];
int i, pos = 0;
name = sample_flags_to_name(flags & ~PERF_IP_FLAG_IN_TX);
if (name)
return fprintf(fp, " %-15s%4s ", name, in_tx ? "(x)" : "");
if (flags & PERF_IP_FLAG_TRACE_BEGIN) {
name = sample_flags_to_name(flags & ~(PERF_IP_FLAG_IN_TX | PERF_IP_FLAG_TRACE_BEGIN));
if (name)
return fprintf(fp, " tr strt %-7s%4s ", name, in_tx ? "(x)" : "");
}
if (flags & PERF_IP_FLAG_TRACE_END) {
name = sample_flags_to_name(flags & ~(PERF_IP_FLAG_IN_TX | PERF_IP_FLAG_TRACE_END));
if (name)
return fprintf(fp, " tr end %-7s%4s ", name, in_tx ? "(x)" : "");
perf script: Print sample flags more nicely The flags field is synthesized and may have a value when Instruction Trace decoding. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Change the display so that known combinations of flags are printed more nicely e.g.: "call" for "bc", "return" for "br", "jcc" for "bo", "jmp" for "b", "int" for "bci", "iret" for "bri", "syscall" for "bcs", "sysret" for "brs", "async" for "by", "hw int" for "bcyi", "tx abrt" for "bA", "tr strt" for "bB", "tr end" for "bE". However the "x" flag will be displayed separately in those cases e.g. "jcc (x)" for a condition branch within a transaction. Example: perf record -e intel_pt//u ls perf script --ns -F comm,cpu,pid,tid,time,ip,addr,sym,dso,symoff,flags ... ls 3689/3689 [001] 2062.020965237: jcc 7f06a958847a _dl_sysdep_start+0xfa (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588450 _dl_sysdep_start+0xd0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a9588461 _dl_sysdep_start+0xe1 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885a0 _dl_sysdep_start+0x220 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a95885a4 _dl_sysdep_start+0x224 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588470 _dl_sysdep_start+0xf0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: call 7f06a95884c3 _dl_sysdep_start+0x143 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9589140 brk+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: syscall 7f06a958914a brk+0xa (/lib/x86_64-linux-gnu/ld-2.19.so) => 0 [unknown] ([unknown]) ls 3689/3689 [001] 2062.020966237: tr strt 0 [unknown] ([unknown]) => 7f06a958914c brk+0xc (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: return 7f06a9589165 brk+0x25 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95884c8 _dl_sysdep_start+0x148 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a95884d7 _dl_sysdep_start+0x157 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: call 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac50 strlen+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a958ac6e strlen+0x1e (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac60 strlen+0x10 (/lib/x86_64-linux-gnu/ld-2.19.so) ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:56 +00:00
}
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
for (i = 0; i < n; i++, flags >>= 1) {
if (flags & 1)
str[pos++] = chars[i];
}
for (; i < 32; i++, flags >>= 1) {
if (flags & 1)
str[pos++] = '?';
}
str[pos] = 0;
perf script: Print sample flags more nicely The flags field is synthesized and may have a value when Instruction Trace decoding. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Change the display so that known combinations of flags are printed more nicely e.g.: "call" for "bc", "return" for "br", "jcc" for "bo", "jmp" for "b", "int" for "bci", "iret" for "bri", "syscall" for "bcs", "sysret" for "brs", "async" for "by", "hw int" for "bcyi", "tx abrt" for "bA", "tr strt" for "bB", "tr end" for "bE". However the "x" flag will be displayed separately in those cases e.g. "jcc (x)" for a condition branch within a transaction. Example: perf record -e intel_pt//u ls perf script --ns -F comm,cpu,pid,tid,time,ip,addr,sym,dso,symoff,flags ... ls 3689/3689 [001] 2062.020965237: jcc 7f06a958847a _dl_sysdep_start+0xfa (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588450 _dl_sysdep_start+0xd0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a9588461 _dl_sysdep_start+0xe1 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885a0 _dl_sysdep_start+0x220 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965237: jmp 7f06a95885a4 _dl_sysdep_start+0x224 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9588470 _dl_sysdep_start+0xf0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: call 7f06a95884c3 _dl_sysdep_start+0x143 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a9589140 brk+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020965904: syscall 7f06a958914a brk+0xa (/lib/x86_64-linux-gnu/ld-2.19.so) => 0 [unknown] ([unknown]) ls 3689/3689 [001] 2062.020966237: tr strt 0 [unknown] ([unknown]) => 7f06a958914c brk+0xc (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: return 7f06a9589165 brk+0x25 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95884c8 _dl_sysdep_start+0x148 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a95884d7 _dl_sysdep_start+0x157 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: call 7f06a95885f0 _dl_sysdep_start+0x270 (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac50 strlen+0x0 (/lib/x86_64-linux-gnu/ld-2.19.so) ls 3689/3689 [001] 2062.020966237: jcc 7f06a958ac6e strlen+0x1e (/lib/x86_64-linux-gnu/ld-2.19.so) => 7f06a958ac60 strlen+0x10 (/lib/x86_64-linux-gnu/ld-2.19.so) ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:56 +00:00
return fprintf(fp, " %-19s ", str);
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
}
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
struct printer_data {
int line_no;
bool hit_nul;
bool is_printable;
};
static int sample__fprintf_bpf_output(enum binary_printer_ops op,
unsigned int val,
void *extra, FILE *fp)
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
{
unsigned char ch = (unsigned char)val;
struct printer_data *printer_data = extra;
int printed = 0;
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
switch (op) {
case BINARY_PRINT_DATA_BEGIN:
printed += fprintf(fp, "\n");
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
break;
case BINARY_PRINT_LINE_BEGIN:
printed += fprintf(fp, "%17s", !printer_data->line_no ? "BPF output:" :
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
" ");
break;
case BINARY_PRINT_ADDR:
printed += fprintf(fp, " %04x:", val);
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
break;
case BINARY_PRINT_NUM_DATA:
printed += fprintf(fp, " %02x", val);
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
break;
case BINARY_PRINT_NUM_PAD:
printed += fprintf(fp, " ");
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
break;
case BINARY_PRINT_SEP:
printed += fprintf(fp, " ");
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
break;
case BINARY_PRINT_CHAR_DATA:
if (printer_data->hit_nul && ch)
printer_data->is_printable = false;
if (!isprint(ch)) {
printed += fprintf(fp, "%c", '.');
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
if (!printer_data->is_printable)
break;
if (ch == '\0')
printer_data->hit_nul = true;
else
printer_data->is_printable = false;
} else {
printed += fprintf(fp, "%c", ch);
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
}
break;
case BINARY_PRINT_CHAR_PAD:
printed += fprintf(fp, " ");
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
break;
case BINARY_PRINT_LINE_END:
printed += fprintf(fp, "\n");
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
printer_data->line_no++;
break;
case BINARY_PRINT_DATA_END:
default:
break;
}
return printed;
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
}
static int perf_sample__fprintf_bpf_output(struct perf_sample *sample, FILE *fp)
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
{
unsigned int nr_bytes = sample->raw_size;
struct printer_data printer_data = {0, false, true};
int printed = binary__fprintf(sample->raw_data, nr_bytes, 8,
sample__fprintf_bpf_output, &printer_data, fp);
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
if (printer_data.is_printable && printer_data.hit_nul)
printed += fprintf(fp, "%17s \"%s\"\n", "BPF string:", (char *)(sample->raw_data));
return printed;
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
}
static int perf_sample__fprintf_spacing(int len, int spacing, FILE *fp)
{
if (len > 0 && len < spacing)
return fprintf(fp, "%*s", spacing - len, "");
return 0;
}
static int perf_sample__fprintf_pt_spacing(int len, FILE *fp)
{
return perf_sample__fprintf_spacing(len, 34, fp);
}
static int perf_sample__fprintf_synth_ptwrite(struct perf_sample *sample, FILE *fp)
{
struct perf_synth_intel_ptwrite *data = perf_sample__synth_ptr(sample);
int len;
if (perf_sample__bad_synth_size(sample, *data))
return 0;
len = fprintf(fp, " IP: %u payload: %#" PRIx64 " ",
data->ip, le64_to_cpu(data->payload));
return len + perf_sample__fprintf_pt_spacing(len, fp);
}
static int perf_sample__fprintf_synth_mwait(struct perf_sample *sample, FILE *fp)
{
struct perf_synth_intel_mwait *data = perf_sample__synth_ptr(sample);
int len;
if (perf_sample__bad_synth_size(sample, *data))
return 0;
len = fprintf(fp, " hints: %#x extensions: %#x ",
data->hints, data->extensions);
return len + perf_sample__fprintf_pt_spacing(len, fp);
}
static int perf_sample__fprintf_synth_pwre(struct perf_sample *sample, FILE *fp)
{
struct perf_synth_intel_pwre *data = perf_sample__synth_ptr(sample);
int len;
if (perf_sample__bad_synth_size(sample, *data))
return 0;
len = fprintf(fp, " hw: %u cstate: %u sub-cstate: %u ",
data->hw, data->cstate, data->subcstate);
return len + perf_sample__fprintf_pt_spacing(len, fp);
}
static int perf_sample__fprintf_synth_exstop(struct perf_sample *sample, FILE *fp)
{
struct perf_synth_intel_exstop *data = perf_sample__synth_ptr(sample);
int len;
if (perf_sample__bad_synth_size(sample, *data))
return 0;
len = fprintf(fp, " IP: %u ", data->ip);
return len + perf_sample__fprintf_pt_spacing(len, fp);
}
static int perf_sample__fprintf_synth_pwrx(struct perf_sample *sample, FILE *fp)
{
struct perf_synth_intel_pwrx *data = perf_sample__synth_ptr(sample);
int len;
if (perf_sample__bad_synth_size(sample, *data))
return 0;
len = fprintf(fp, " deepest cstate: %u last cstate: %u wake reason: %#x ",
data->deepest_cstate, data->last_cstate,
data->wake_reason);
return len + perf_sample__fprintf_pt_spacing(len, fp);
}
static int perf_sample__fprintf_synth_cbr(struct perf_sample *sample, FILE *fp)
{
struct perf_synth_intel_cbr *data = perf_sample__synth_ptr(sample);
unsigned int percent, freq;
int len;
if (perf_sample__bad_synth_size(sample, *data))
return 0;
freq = (le32_to_cpu(data->freq) + 500) / 1000;
len = fprintf(fp, " cbr: %2u freq: %4u MHz ", data->cbr, freq);
if (data->max_nonturbo) {
percent = (5 + (1000 * data->cbr) / data->max_nonturbo) / 10;
len += fprintf(fp, "(%3u%%) ", percent);
}
return len + perf_sample__fprintf_pt_spacing(len, fp);
}
static int perf_sample__fprintf_synth(struct perf_sample *sample,
struct perf_evsel *evsel, FILE *fp)
{
switch (evsel->attr.config) {
case PERF_SYNTH_INTEL_PTWRITE:
return perf_sample__fprintf_synth_ptwrite(sample, fp);
case PERF_SYNTH_INTEL_MWAIT:
return perf_sample__fprintf_synth_mwait(sample, fp);
case PERF_SYNTH_INTEL_PWRE:
return perf_sample__fprintf_synth_pwre(sample, fp);
case PERF_SYNTH_INTEL_EXSTOP:
return perf_sample__fprintf_synth_exstop(sample, fp);
case PERF_SYNTH_INTEL_PWRX:
return perf_sample__fprintf_synth_pwrx(sample, fp);
case PERF_SYNTH_INTEL_CBR:
return perf_sample__fprintf_synth_cbr(sample, fp);
default:
break;
}
return 0;
}
struct perf_script {
struct perf_tool tool;
struct perf_session *session;
bool show_task_events;
bool show_mmap_events;
bool show_switch_events;
perf script: Add script print support for namespace events Introduce a new option to display events of type PERF_RECORD_NAMESPACES and update perf-script documentation accordingly. Shown below is output (trimmed) of perf script command with the newly introduced option, on perf.data generated with perf record command using --namespaces option. $ perf script --show-namespace-events swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 1/1 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 2/2 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] Commiter notes: Testing it: Investigating that double PERF_RECORD_NAMESPACES for the 19155 pid/tid... Its more than that, there are two PERF_RECORD_COMM as well, and with zeroed timestamps, so probably a synthesizing artifact... # perf script --show-task --show-namespace <SNIP> perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19154/19154 perf 0 [000] 0.000000: PERF_RECORD_FORK(19155:19155):(19154:19154) perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 3110.881834: 1 cycles: ffffffffa7060bf6 native_write_msr (/lib/modules/4.11.0-rc1+/build/vmlinux) <SNIP> Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com> Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Eric Biederman <ebiederm@xmission.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sargun Dhillon <sargun@sargun.me> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/148891932627.25309.1941587059154176221.stgit@hbathini.in.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-03-07 20:42:06 +00:00
bool show_namespace_events;
bool show_lost_events;
bool show_round_events;
bool allocated;
bool per_event_dump;
struct cpu_map *cpus;
struct thread_map *threads;
int name_width;
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
const char *time_str;
struct perf_time_interval *ptime_range;
int range_size;
int range_num;
};
static int perf_evlist__max_name_len(struct perf_evlist *evlist)
{
struct perf_evsel *evsel;
int max = 0;
evlist__for_each_entry(evlist, evsel) {
int len = strlen(perf_evsel__name(evsel));
max = MAX(len, max);
}
return max;
}
static int data_src__fprintf(u64 data_src, FILE *fp)
{
struct mem_info mi = { .data_src.val = data_src };
char decode[100];
char out[100];
static int maxlen;
int len;
perf_script__meminfo_scnprintf(decode, 100, &mi);
len = scnprintf(out, 100, "%16" PRIx64 " %s", data_src, decode);
if (maxlen < len)
maxlen = len;
return fprintf(fp, "%-*s", maxlen, out);
}
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
struct metric_ctx {
struct perf_sample *sample;
struct thread *thread;
struct perf_evsel *evsel;
FILE *fp;
};
static void script_print_metric(struct perf_stat_config *config __maybe_unused,
void *ctx, const char *color,
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
const char *fmt,
const char *unit, double val)
{
struct metric_ctx *mctx = ctx;
if (!fmt)
return;
perf_sample__fprintf_start(mctx->sample, mctx->thread, mctx->evsel,
PERF_RECORD_SAMPLE, mctx->fp);
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
fputs("\tmetric: ", mctx->fp);
if (color)
color_fprintf(mctx->fp, color, fmt, val);
else
printf(fmt, val);
fprintf(mctx->fp, " %s\n", unit);
}
static void script_new_line(struct perf_stat_config *config __maybe_unused,
void *ctx)
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
{
struct metric_ctx *mctx = ctx;
perf_sample__fprintf_start(mctx->sample, mctx->thread, mctx->evsel,
PERF_RECORD_SAMPLE, mctx->fp);
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
fputs("\tmetric: ", mctx->fp);
}
static void perf_sample__fprint_metric(struct perf_script *script,
struct thread *thread,
struct perf_evsel *evsel,
struct perf_sample *sample,
FILE *fp)
{
struct perf_stat_output_ctx ctx = {
.print_metric = script_print_metric,
.new_line = script_new_line,
.ctx = &(struct metric_ctx) {
.sample = sample,
.thread = thread,
.evsel = evsel,
.fp = fp,
},
.force_header = false,
};
struct perf_evsel *ev2;
static bool init;
u64 val;
if (!init) {
perf_stat__init_shadow_stats();
init = true;
}
if (!evsel->stats)
perf_evlist__alloc_stats(script->session->evlist, false);
if (evsel_script(evsel->leader)->gnum++ == 0)
perf_stat__reset_shadow_stats();
val = sample->period * evsel->scale;
perf_stat__update_shadow_stats(evsel,
val,
sample->cpu,
&rt_stat);
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
evsel_script(evsel)->val = val;
if (evsel_script(evsel->leader)->gnum == evsel->leader->nr_members) {
for_each_group_member (ev2, evsel->leader) {
perf_stat__print_shadow_stats(&stat_config, ev2,
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
evsel_script(ev2)->val,
sample->cpu,
&ctx,
NULL,
&rt_stat);
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
}
evsel_script(evsel->leader)->gnum = 0;
}
}
perf script: Implement --graph-function Add a ftrace style --graph-function argument to 'perf script' that allows to print itrace function calls only below a given function. This makes it easier to find the code of interest in a large trace. % perf record -e intel_pt//k -a sleep 1 % perf script --graph-function group_sched_in --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) calc_timer_values perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_cpu perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) arch_perf_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __fentry__ perf 900 [000] 194167.205652203: ([kernel.kallsyms]) using_native_sched_clock perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_stable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) group_sched_in swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_set_state.part.71 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_time swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_pmu_disable swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_log_itrace_start swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) calc_timer_values swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_cpu swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) arch_perf_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __fentry__ swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) using_native_sched_clock swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_stable Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:39 +00:00
static bool show_event(struct perf_sample *sample,
struct perf_evsel *evsel,
struct thread *thread,
struct addr_location *al)
{
int depth = thread_stack__depth(thread);
if (!symbol_conf.graph_function)
return true;
if (thread->filter) {
if (depth <= thread->filter_entry_depth) {
thread->filter = false;
return false;
}
return true;
} else {
const char *s = symbol_conf.graph_function;
u64 ip;
const char *name = resolve_branch_sym(sample, evsel, thread, al,
&ip);
unsigned nlen;
if (!name)
return false;
nlen = strlen(name);
while (*s) {
unsigned len = strcspn(s, ",");
if (nlen == len && !strncmp(name, s, len)) {
thread->filter = true;
thread->filter_entry_depth = depth;
return true;
}
s += len;
if (*s == ',')
s++;
}
return false;
}
}
static void process_event(struct perf_script *script,
struct perf_sample *sample, struct perf_evsel *evsel,
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
struct addr_location *al,
struct machine *machine)
{
struct thread *thread = al->thread;
struct perf_event_attr *attr = &evsel->attr;
unsigned int type = output_type(attr->type);
struct perf_evsel_script *es = evsel->priv;
FILE *fp = es->fp;
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
if (output[type].fields == 0)
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
return;
perf script: Implement --graph-function Add a ftrace style --graph-function argument to 'perf script' that allows to print itrace function calls only below a given function. This makes it easier to find the code of interest in a large trace. % perf record -e intel_pt//k -a sleep 1 % perf script --graph-function group_sched_in --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) calc_timer_values perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_cpu perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) arch_perf_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __fentry__ perf 900 [000] 194167.205652203: ([kernel.kallsyms]) using_native_sched_clock perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_stable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) group_sched_in swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_set_state.part.71 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_time swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_pmu_disable swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_log_itrace_start swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) calc_timer_values swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_cpu swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) arch_perf_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __fentry__ swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) using_native_sched_clock swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_stable Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:39 +00:00
if (!show_event(sample, evsel, thread, al))
return;
++es->samples;
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_SAMPLE, fp);
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
perf script: Add period data column Adding period data column to be displayed in perf script. It's possible to get period values using -f option, like: $ perf script -f comm,tid,time,period,ip,sym,dso :26019 26019 52414.329088: 3707 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329088: 44 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 1987 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 6 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) ls 26019 52414.329442: 537558 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.329442: 2099 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.330181: 1242100 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.330181: 3774 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.331427: 1083662 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) ls 26019 52414.331427: 360 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) Signed-off-by: Jiri Olsa <jolsa@kernel.org> Acked-by: David Ahern <dsahern@gmail.com> Cc: "Jen-Cheng(Tommy) Huang" <tommy24@gatech.edu> Cc: Andi Kleen <andi@firstfloor.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jen-Cheng(Tommy) Huang <tommy24@gatech.edu> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1408977943-16594-9-git-send-email-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2014-08-25 14:45:42 +00:00
if (PRINT_FIELD(PERIOD))
fprintf(fp, "%10" PRIu64 " ", sample->period);
perf script: Add period data column Adding period data column to be displayed in perf script. It's possible to get period values using -f option, like: $ perf script -f comm,tid,time,period,ip,sym,dso :26019 26019 52414.329088: 3707 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329088: 44 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 1987 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) :26019 26019 52414.329093: 6 ffffffff8105443a native_write_msr_safe ([kernel.kallsyms]) ls 26019 52414.329442: 537558 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.329442: 2099 3407c0639c _dl_map_object_from_fd (/usr/lib64/ld-2.17.so) ls 26019 52414.330181: 1242100 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.330181: 3774 34080917bb get_next_seq (/usr/lib64/libc-2.17.so) ls 26019 52414.331427: 1083662 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) ls 26019 52414.331427: 360 ffffffff810c7dc2 update_curr ([kernel.kallsyms]) Signed-off-by: Jiri Olsa <jolsa@kernel.org> Acked-by: David Ahern <dsahern@gmail.com> Cc: "Jen-Cheng(Tommy) Huang" <tommy24@gatech.edu> Cc: Andi Kleen <andi@firstfloor.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jen-Cheng(Tommy) Huang <tommy24@gatech.edu> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1408977943-16594-9-git-send-email-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2014-08-25 14:45:42 +00:00
if (PRINT_FIELD(EVNAME)) {
const char *evname = perf_evsel__name(evsel);
if (!script->name_width)
script->name_width = perf_evlist__max_name_len(script->session->evlist);
fprintf(fp, "%*s: ", script->name_width, evname ?: "[unknown]");
}
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
if (print_flags)
perf_sample__fprintf_flags(sample->flags, fp);
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
if (is_bts_event(attr)) {
perf_sample__fprintf_bts(sample, evsel, thread, al, machine, fp);
return;
}
if (PRINT_FIELD(TRACE)) {
event_format__fprintf(evsel->tp_format, sample->cpu,
sample->raw_data, sample->raw_size, fp);
}
if (attr->type == PERF_TYPE_SYNTH && PRINT_FIELD(SYNTH))
perf_sample__fprintf_synth(sample, evsel, fp);
if (PRINT_FIELD(ADDR))
perf_sample__fprintf_addr(sample, thread, attr, fp);
if (PRINT_FIELD(DATA_SRC))
data_src__fprintf(sample->data_src, fp);
if (PRINT_FIELD(WEIGHT))
fprintf(fp, "%16" PRIu64, sample->weight);
if (PRINT_FIELD(IP)) {
perf script: Fix segfault when printing callchains This fixes a bug caused by an unitialized callchain cursor. The crash frist appeared in: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") The callchain cursor is a struct that contains pointers, that when uninitialized will cause unpredictable behavior (usually a crash) when trying to append to the callchain. The existing implementation has the following issues: 1. The callchain cursor used is not initialized, resulting in unpredictable behavior when used. 2. The cursor is declared on the stack. Even if it is properly initalized, the implmentation will leak memory when the function returns, since all the references to the callchain_nodes allocated by callchain_cursor_append will be lost when the cursor goes out of scope. 3. Storing the cursor on the stack is inefficient. Even if memory is properly freed when it goes out of scope, a performance penalty will be incurred due to reallocation of callchain nodes. callchain_cursor_append is designed to avoid these reallocations when an existing cursor is reused. This patch fixes the crash by replacing cursor_callchain with a reference to the global callchain_cursor which also resolves all 3 issues mentioned above. How to reproduce the crash: $ perf record --call-graph=dwarf stress -t 1 -c 1 $ perf script > /dev/null Segfault Signed-off-by: Chris Phlipot <cphlipot0@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-20 02:32:11 +00:00
struct callchain_cursor *cursor = NULL;
if (symbol_conf.use_callchain && sample->callchain &&
perf script: Fix segfault when printing callchains This fixes a bug caused by an unitialized callchain cursor. The crash frist appeared in: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") The callchain cursor is a struct that contains pointers, that when uninitialized will cause unpredictable behavior (usually a crash) when trying to append to the callchain. The existing implementation has the following issues: 1. The callchain cursor used is not initialized, resulting in unpredictable behavior when used. 2. The cursor is declared on the stack. Even if it is properly initalized, the implmentation will leak memory when the function returns, since all the references to the callchain_nodes allocated by callchain_cursor_append will be lost when the cursor goes out of scope. 3. Storing the cursor on the stack is inefficient. Even if memory is properly freed when it goes out of scope, a performance penalty will be incurred due to reallocation of callchain nodes. callchain_cursor_append is designed to avoid these reallocations when an existing cursor is reused. This patch fixes the crash by replacing cursor_callchain with a reference to the global callchain_cursor which also resolves all 3 issues mentioned above. How to reproduce the crash: $ perf record --call-graph=dwarf stress -t 1 -c 1 $ perf script > /dev/null Segfault Signed-off-by: Chris Phlipot <cphlipot0@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-20 02:32:11 +00:00
thread__resolve_callchain(al->thread, &callchain_cursor, evsel,
sample, NULL, NULL, scripting_max_stack) == 0)
perf script: Fix segfault when printing callchains This fixes a bug caused by an unitialized callchain cursor. The crash frist appeared in: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") The callchain cursor is a struct that contains pointers, that when uninitialized will cause unpredictable behavior (usually a crash) when trying to append to the callchain. The existing implementation has the following issues: 1. The callchain cursor used is not initialized, resulting in unpredictable behavior when used. 2. The cursor is declared on the stack. Even if it is properly initalized, the implmentation will leak memory when the function returns, since all the references to the callchain_nodes allocated by callchain_cursor_append will be lost when the cursor goes out of scope. 3. Storing the cursor on the stack is inefficient. Even if memory is properly freed when it goes out of scope, a performance penalty will be incurred due to reallocation of callchain nodes. callchain_cursor_append is designed to avoid these reallocations when an existing cursor is reused. This patch fixes the crash by replacing cursor_callchain with a reference to the global callchain_cursor which also resolves all 3 issues mentioned above. How to reproduce the crash: $ perf record --call-graph=dwarf stress -t 1 -c 1 $ perf script > /dev/null Segfault Signed-off-by: Chris Phlipot <cphlipot0@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}") Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-20 02:32:11 +00:00
cursor = &callchain_cursor;
fputc(cursor ? '\n' : ' ', fp);
sample__fprintf_sym(sample, al, 0, output[type].print_ip_opts, cursor, fp);
perf script: Add support for dumping symbols Add option to dump symbols found in events. e.g., perf script -f comm,pid,tid,time,trace,sym swapper 0/0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff8100134a cpu_idle ([kernel.kallsyms]) ffffffff81370b39 rest_init ([kernel.kallsyms]) ffffffff81696c23 start_kernel ([kernel.kallsyms].init.text) ffffffff816962af x86_64_start_reservations ([kernel.kallsyms].init.text) ffffffff816963b9 x86_64_start_kernel ([kernel.kallsyms].init.text) sshd 1675/1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff813837aa schedule_hrtimeout_range_clock ([kernel.kallsyms]) ffffffff81383886 schedule_hrtimeout_range ([kernel.kallsyms]) ffffffff8110c4f9 poll_schedule_timeout ([kernel.kallsyms]) ffffffff8110cd20 do_select ([kernel.kallsyms]) ffffffff8110ced8 core_sys_select ([kernel.kallsyms]) ffffffff8110d00d sys_select ([kernel.kallsyms]) ffffffff81002bc2 system_call ([kernel.kallsyms]) 7f1647e56e93 __GI_select (/lib64/libc-2.12.90.so) netstat 1692/1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff81002c3a sysret_careful ([kernel.kallsyms]) 7f7a6cd1b210 __GI___libc_read (/lib64/libc-2.12.90.so) Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-6-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:27 +00:00
}
if (PRINT_FIELD(IREGS))
perf_sample__fprintf_iregs(sample, attr, fp);
if (PRINT_FIELD(UREGS))
perf_sample__fprintf_uregs(sample, attr, fp);
if (PRINT_FIELD(BRSTACK))
perf_sample__fprintf_brstack(sample, thread, attr, fp);
else if (PRINT_FIELD(BRSTACKSYM))
perf_sample__fprintf_brstacksym(sample, thread, attr, fp);
perf script: Support -F brstackoff,dso The idea here is to make AutoFDO easier in cloud environment with ASLR. It's easiest to show how this is useful by example. I built a small test akin to "while(1) { do_nothing(); }" where the do_nothing function is loaded from a dso: $ cat burncpu.cpp #include <dlfcn.h> int main() { void* handle = dlopen("./dso.so", RTLD_LAZY); if (!handle) return -1; typedef void (*fp)(); fp do_nothing = (fp) dlsym(handle, "do_nothing"); while(1) { do_nothing(); } } $ cat dso.cpp extern "C" void do_nothing() {} $ cat build.sh #!/bin/bash g++ -shared dso.cpp -o dso.so g++ burncpu.cpp -o burncpu -ldl I sampled the execution of this program with perf record -b. Using the existing "brstack,dso", we get absolute addresses that are affected by ASLR, and could be different on different hosts. The address does not uniquely identify a branch/target in the binary: $ perf script -F brstack,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x7f967139b6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Using the existing "brstacksym,dso" is a little better, because the symbol plus offset and dso name *does* uniquely identify a branch/target in the binary. Ultimately, however, AutoFDO wants a simple offset into the binary, so we'd have to undo all the work perf did to symbolize in the first place: $ perf script -F brstacksym,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 do_nothing+0x5(/tmp/burncpu/dso.so)/main+0x44(/tmp/burncpu/exe)/P/-/-/0 With the new "brstackoff,dso" we get what we need: a simple offset into a specific dso/binary that uniquely identifies a branch/target: $ perf script -F brstackoff,dso | sed 's/\/0 /\/0\n/g' | grep burncpu | grep dso.so | head -n 1 0x6aa(/tmp/burncpu/dso.so)/0x4006b1(/tmp/burncpu/exe)/P/-/-/0 Signed-off-by: Mark Santaniello <marksan@fb.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20170619163825.2012979-2-marksan@fb.com [ Updated documentation about 'brstackoff' using text from above ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-19 16:38:25 +00:00
else if (PRINT_FIELD(BRSTACKOFF))
perf_sample__fprintf_brstackoff(sample, thread, attr, fp);
perf script: Print bpf-output events in 'perf script' This patch allows 'perf script' output messages from BPF program. For example, use test_bpf_output_3.c at the end of this commit message, # ./perf record -e bpf-output/no-inherit,name=evt/ \ -e ./test_bpf_output_3.c/map:channel.event=evt/ \ usleep 100000 # ./perf script usleep 4882 21384.532523: evt: ffffffff810e97d1 sys_nanosleep ([kernel.kallsyms]) BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" usleep 4882 21384.632606: evt: ffffffff8105c609 kretprobe_trampoline_holder ([kernel.kallsyms BPF output: 0000: 52 61 69 73 65 20 61 20 Raise a 0008: 42 50 46 20 65 76 65 6e BPF even 0010: 74 21 00 00 t!.. BPF string: "Raise a BPF event!" Two samples from BPF output are printed by both binary and string format. If BPF program output something unprintable, string format is suppressed. /************************ BEGIN **************************/ #include <uapi/linux/bpf.h> struct bpf_map_def { unsigned int type; unsigned int key_size; unsigned int value_size; unsigned int max_entries; }; #define SEC(NAME) __attribute__((section(NAME), used)) static u64 (*ktime_get_ns)(void) = (void *)BPF_FUNC_ktime_get_ns; static int (*trace_printk)(const char *fmt, int fmt_size, ...) = (void *)BPF_FUNC_trace_printk; static int (*get_smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id; static int (*perf_event_output)(void *, struct bpf_map_def *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output; struct bpf_map_def SEC("maps") channel = { .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, .key_size = sizeof(int), .value_size = sizeof(u32), .max_entries = __NR_CPUS__, }; static inline int __attribute__((always_inline)) func(void *ctx, int type) { char output_str[] = "Raise a BPF event!"; perf_event_output(ctx, &channel, get_smp_processor_id(), &output_str, sizeof(output_str)); return 0; } SEC("func_begin=sys_nanosleep") int func_begin(void *ctx) {return func(ctx, 1);} SEC("func_end=sys_nanosleep%return") int func_end(void *ctx) { return func(ctx, 2);} char _license[] SEC("license") = "GPL"; int _version SEC("version") = LINUX_VERSION_CODE; /************************* END ***************************/ Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1456312845-111583-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-02-24 11:20:45 +00:00
if (perf_evsel__is_bpf_output(evsel) && PRINT_FIELD(BPF_OUTPUT))
perf_sample__fprintf_bpf_output(sample, fp);
perf_sample__fprintf_insn(sample, attr, thread, machine, fp);
if (PRINT_FIELD(PHYS_ADDR))
fprintf(fp, "%16" PRIx64, sample->phys_addr);
fprintf(fp, "\n");
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
perf tools: Support 'srccode' output When looking at PT or brstackinsn traces with 'perf script' it can be very useful to see the source code. This adds a simple facility to print them with 'perf script', if the information is available through dwarf % perf record ... % perf script -F insn,ip,sym,srccode ... 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004c6 main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004cd main 5 for (i = 0; i < 10000000; i++) 4004b3 main 6 v++; % perf record -b ... % perf script -F insn,ip,sym,srccode,brstackinsn ... main+22: 0000000000400543 insn: e8 ca ff ff ff # PRED |18 f1(); f1: 0000000000400512 insn: 55 |10 { 0000000000400513 insn: 48 89 e5 0000000000400516 insn: b8 00 00 00 00 |11 f2(); 000000000040051b insn: e8 d6 ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; 0000000000400500 insn: 8b 0d 2a 0b 20 00 0000000000400506 insn: 99 0000000000400507 insn: f7 f9 0000000000400509 insn: 89 05 29 0b 20 00 000000000040050f insn: 90 |7 } 0000000000400510 insn: 5d 0000000000400511 insn: c3 # PRED f1+14: 0000000000400520 insn: b8 00 00 00 00 |12 f2(); 0000000000400525 insn: e8 cc ff ff ff # PRED f2: 00000000004004f6 insn: 55 |5 { 00000000004004f7 insn: 48 89 e5 00000000004004fa insn: 8b 05 2c 0b 20 00 |6 c = a / b; Not supported for callchains currently, would need some layout changes there. Committer notes: Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this warning: In file included from util/srccode.c:19:0: /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp] #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> ^~~~~~~ cc1: all warnings being treated as errors Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-12-04 00:18:48 +00:00
if (PRINT_FIELD(SRCCODE)) {
if (map__fprintf_srccode(al->map, al->addr, stdout,
&thread->srccode_state))
printf("\n");
}
perf script: Allow computing 'perf stat' style metrics Add support for computing 'perf stat' style metrics in 'perf script'. When using leader sampling we can get metrics for each sampling period by computing formulas over the values of the different group members. This allows things like fine grained IPC tracking through sampling, much more fine grained than with 'perf stat'. The metric is still averaged over the sampling period, it is not just for the sampling point. This patch adds a new metric output field for 'perf script' that uses the existing 'perf stat' metrics infrastructure to compute any metrics supported by 'perf stat'. For example to sample IPC: $ perf record -e '{ref-cycles,cycles,instructions}:S' -a sleep 1 $ perf script -F metric,ip,sym,time,cpu,comm ... alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: 7fd65937d6cc [unknown] alsa-sink-ALC32 [000] 42815.856074: metric: 0.13 insn per cycle swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: ffffffff81655df0 __schedule swapper [000] 42815.857961: metric: 0.23 insn per cycle qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: ffffffff8165ad0e _raw_spin_unlock_irqrestore qemu-system-x86 [000] 42815.858130: metric: 0.46 insn per cycle :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: ffffffffa080e5f2 vmx_vcpu_run :4972 [000] 42815.858312: metric: 0.45 insn per cycle TopDown: This requires disabling SMT if you have it enabled, because SMT would require sampling per core, which is not supported. $ perf record -e '{ref-cycles,topdown-fetch-bubbles,\ topdown-recovery-bubbles,\ topdown-slots-retired,topdown-total-slots,\ topdown-slots-issued}:S' -a sleep 1 $ perf script --header -I -F cpu,ip,sym,event,metric,period ... [000] 121108 ref-cycles: ffffffff8165222e copy_user_enhanced_fast_string [000] 190350 topdown-fetch-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 2055 topdown-recovery-bubbles: ffffffff8165222e copy_user_enhanced_fast_string [000] 148729 topdown-slots-retired: ffffffff8165222e copy_user_enhanced_fast_string [000] 144324 topdown-total-slots: ffffffff8165222e copy_user_enhanced_fast_string [000] 160852 topdown-slots-issued: ffffffff8165222e copy_user_enhanced_fast_string [000] metric: 33.0% frontend bound [000] metric: 3.5% bad speculation [000] metric: 25.8% retiring [000] metric: 37.7% backend bound [000] 112112 ref-cycles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 357222 topdown-fetch-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 3325 topdown-recovery-bubbles: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 323553 topdown-slots-retired: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 270507 topdown-total-slots: ffffffff8165aec8 _raw_spin_lock_irqsave [000] 341226 topdown-slots-issued: ffffffff8165aec8 _raw_spin_lock_irqsave [000] metric: 33.0% frontend bound [000] metric: 2.9% bad speculation [000] metric: 29.9% retiring [000] metric: 34.2% backend bound ... v2: Use evsel->priv for new fields Port to new base line, support fp output. Handle stats in ->stats, not ->priv Minor cleanups Extra explanation about the use of the term 'averaging', from Andi in the thread in the Link: tag below: <quote Andi> The current samples contains the sum of event counts for a sampling period. EventA-1 EventA-2 EventA-3 EventA-4 EventB-1 EventB-2 EventC-3 gap with no events overflow |-----------------------------------------------------------------| period-start period-end ^ ^ | | previous sample current sample So EventA = 4 and EventB = 3 at the sample point I generate a metric, let's say EventA / EventB. It applies to the whole period. But the metric is over a longer time which does not have the same behavior. For example the gap above doesn't have any events, while they are clustered at the beginning and end of the sample period. But we're summing everything together. The metric doesn't know that the gap is different than the busy period. That's what I'm trying to express with averaging. </quote> Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20171117214300.32746-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-11-17 21:43:00 +00:00
if (PRINT_FIELD(METRIC))
perf_sample__fprint_metric(script, thread, evsel, sample, fp);
perf script: Flush output stream after events in verbose mode When the perf script output is written to a terminal stream, the normal output of `perf script` would get buffered, but its debug output would be written directly. This made it quite hard to figure out where a given debug output is coming from. We can improve on this by flushing the output buffer after processing an event. To see the value, compare the following output for a `perf script -v` run: Before this patch: ``` unwind: reg 16, val 7faf7dfdc000 unwind: reg 7, val 7ffc80811e30 unwind: find_proc_info dso /usr/lib/ld-2.28.so unwind: reg 6, val 0 unwind: _start:ip = 0x7faf7dfdc000 (0x2000) unwind: reg 16, val 7faf7dfdc000 unwind: reg 7, val 7ffc80811e30 unwind: find_proc_info dso /usr/lib/ld-2.28.so unwind: reg 6, val 0 unwind: _start:ip = 0x7faf7dfdc000 (0x2000) unwind: reg 16, val 7faf7dfdc000 unwind: reg 7, val 7ffc80811e30 unwind: find_proc_info dso /usr/lib/ld-2.28.so unwind: reg 6, val 0 unwind: _start:ip = 0x7faf7dfdc000 (0x2000) unwind: reg 16, val 7faf7dfdc000 unwind: reg 7, val 7ffc80811e30 ... lots and lots of verbose debug output cpp-inlining 24617 90229.122036534: 1 cycles:uppp: 7faf7dfdc000 _start+0x0 (/usr/lib/ld-2.28.so) cpp-inlining 24617 90229.122043974: 1 cycles:uppp: 7faf7dfdc000 _start+0x0 (/usr/lib/ld-2.28.so) ... ``` After this patch: ``` ... unwind: reg 16, val 7faf7dfdc000 unwind: reg 7, val 7ffc80811e30 unwind: find_proc_info dso /usr/lib/ld-2.28.so unwind: reg 6, val 0 unwind: _start:ip = 0x7faf7dfdc000 (0x2000) cpp-inlining 24617 90229.122036534: 1 cycles:uppp: 7faf7dfdc000 _start+0x0 (/usr/lib/ld-2.28.so) unwind: reg 16, val 7faf7dfdc000 unwind: reg 7, val 7ffc80811e30 unwind: find_proc_info dso /usr/lib/ld-2.28.so unwind: reg 6, val 0 unwind: _start:ip = 0x7faf7dfdc000 (0x2000) cpp-inlining 24617 90229.122043974: 1 cycles:uppp: 7faf7dfdc000 _start+0x0 (/usr/lib/ld-2.28.so) ... ``` This new output format makes it much easier to use perf script output for debugging purposes, e.g. to investigate broken dwarf unwinding. Signed-off-by: Milian Wolff <milian.wolff@kdab.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20181021191424.16183-2-milian.wolff@kdab.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-10-21 19:14:24 +00:00
if (verbose)
fflush(fp);
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
static struct scripting_ops *scripting_ops;
perf script: Display stat events by default If no script is specified for stat data, display stat events in raw form. $ perf stat record ls SNIP Performance counter stats for 'ls': 0.851585 task-clock (msec) # 0.717 CPUs utilized 0 context-switches # 0.000 K/sec 0 cpu-migrations # 0.000 K/sec 114 page-faults # 0.134 M/sec 2,620,918 cycles # 3.078 GHz <not supported> stalled-cycles-frontend <not supported> stalled-cycles-backend 2,714,111 instructions # 1.04 insns per cycle 542,434 branches # 636.970 M/sec 15,946 branch-misses # 2.94% of all branches 0.001186954 seconds time elapsed $ perf script CPU THREAD VAL ENA RUN TIME EVENT -1 26185 851585 851585 851585 1186954 task-clock -1 26185 0 851585 851585 1186954 context-switches -1 26185 0 851585 851585 1186954 cpu-migrations -1 26185 114 851585 851585 1186954 page-faults -1 26185 2620918 853340 853340 1186954 cycles -1 26185 0 0 0 1186954 stalled-cycles-frontend -1 26185 0 0 0 1186954 stalled-cycles-backend -1 26185 2714111 853340 853340 1186954 instructions -1 26185 542434 853340 853340 1186954 branches -1 26185 15946 853340 853340 1186954 branch-misses Signed-off-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Kan Liang <kan.liang@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1452077397-31958-3-git-send-email-jolsa@kernel.org [ Rename 'time' parameter to 'tstamp' to fix build on older distros ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-01-06 10:49:56 +00:00
static void __process_stat(struct perf_evsel *counter, u64 tstamp)
{
int nthreads = thread_map__nr(counter->threads);
int ncpus = perf_evsel__nr_cpus(counter);
int cpu, thread;
static int header_printed;
if (counter->system_wide)
nthreads = 1;
if (!header_printed) {
printf("%3s %8s %15s %15s %15s %15s %s\n",
"CPU", "THREAD", "VAL", "ENA", "RUN", "TIME", "EVENT");
header_printed = 1;
}
for (thread = 0; thread < nthreads; thread++) {
for (cpu = 0; cpu < ncpus; cpu++) {
struct perf_counts_values *counts;
counts = perf_counts(counter->counts, cpu, thread);
printf("%3d %8d %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %s\n",
counter->cpus->map[cpu],
thread_map__pid(counter->threads, thread),
counts->val,
counts->ena,
counts->run,
tstamp,
perf_evsel__name(counter));
}
}
}
static void process_stat(struct perf_evsel *counter, u64 tstamp)
{
if (scripting_ops && scripting_ops->process_stat)
scripting_ops->process_stat(&stat_config, counter, tstamp);
perf script: Display stat events by default If no script is specified for stat data, display stat events in raw form. $ perf stat record ls SNIP Performance counter stats for 'ls': 0.851585 task-clock (msec) # 0.717 CPUs utilized 0 context-switches # 0.000 K/sec 0 cpu-migrations # 0.000 K/sec 114 page-faults # 0.134 M/sec 2,620,918 cycles # 3.078 GHz <not supported> stalled-cycles-frontend <not supported> stalled-cycles-backend 2,714,111 instructions # 1.04 insns per cycle 542,434 branches # 636.970 M/sec 15,946 branch-misses # 2.94% of all branches 0.001186954 seconds time elapsed $ perf script CPU THREAD VAL ENA RUN TIME EVENT -1 26185 851585 851585 851585 1186954 task-clock -1 26185 0 851585 851585 1186954 context-switches -1 26185 0 851585 851585 1186954 cpu-migrations -1 26185 114 851585 851585 1186954 page-faults -1 26185 2620918 853340 853340 1186954 cycles -1 26185 0 0 0 1186954 stalled-cycles-frontend -1 26185 0 0 0 1186954 stalled-cycles-backend -1 26185 2714111 853340 853340 1186954 instructions -1 26185 542434 853340 853340 1186954 branches -1 26185 15946 853340 853340 1186954 branch-misses Signed-off-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Kan Liang <kan.liang@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1452077397-31958-3-git-send-email-jolsa@kernel.org [ Rename 'time' parameter to 'tstamp' to fix build on older distros ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-01-06 10:49:56 +00:00
else
__process_stat(counter, tstamp);
}
static void process_stat_interval(u64 tstamp)
{
if (scripting_ops && scripting_ops->process_stat_interval)
scripting_ops->process_stat_interval(tstamp);
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
static void setup_scripting(void)
{
setup_perl_scripting();
setup_python_scripting();
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
static int flush_scripting(void)
{
return scripting_ops ? scripting_ops->flush_script() : 0;
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
static int cleanup_scripting(void)
{
pr_debug("\nperf script stopped\n");
return scripting_ops ? scripting_ops->stop_script() : 0;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
static int process_sample_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct perf_evsel *evsel,
struct machine *machine)
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
{
struct perf_script *scr = container_of(tool, struct perf_script, tool);
struct addr_location al;
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
if (perf_time__ranges_skip_sample(scr->ptime_range, scr->range_num,
sample->time)) {
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
return 0;
}
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
if (debug_mode) {
if (sample->time < last_timestamp) {
pr_err("Samples misordered, previous: %" PRIu64
" this: %" PRIu64 "\n", last_timestamp,
sample->time);
nr_unordered++;
}
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
last_timestamp = sample->time;
return 0;
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
}
if (machine__resolve(machine, &al, sample) < 0) {
pr_err("problem processing %d event, skipping it.\n",
event->header.type);
return -1;
}
if (al.filtered)
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
goto out_put;
if (cpu_list && !test_bit(sample->cpu, cpu_bitmap))
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
goto out_put;
if (scripting_ops)
scripting_ops->process_event(event, sample, evsel, &al);
else
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
process_event(scr, sample, evsel, &al, machine);
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
out_put:
addr_location__put(&al);
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
return 0;
}
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
static int process_attr(struct perf_tool *tool, union perf_event *event,
struct perf_evlist **pevlist)
{
struct perf_script *scr = container_of(tool, struct perf_script, tool);
struct perf_evlist *evlist;
struct perf_evsel *evsel, *pos;
int err;
static struct perf_evsel_script *es;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
err = perf_event__process_attr(tool, event, pevlist);
if (err)
return err;
evlist = *pevlist;
evsel = perf_evlist__last(*pevlist);
if (!evsel->priv) {
if (scr->per_event_dump) {
evsel->priv = perf_evsel_script__new(evsel,
scr->session->data);
} else {
es = zalloc(sizeof(*es));
if (!es)
return -ENOMEM;
es->fp = stdout;
evsel->priv = es;
}
}
if (evsel->attr.type >= PERF_TYPE_MAX &&
evsel->attr.type != PERF_TYPE_SYNTH)
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
return 0;
evlist__for_each_entry(evlist, pos) {
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
if (pos->attr.type == evsel->attr.type && pos != evsel)
return 0;
}
set_print_ip_opts(&evsel->attr);
if (evsel->attr.sample_type)
err = perf_evsel__check_attr(evsel, scr->session);
return err;
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
}
static int process_comm_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
int ret = -1;
thread = machine__findnew_thread(machine, event->comm.pid, event->comm.tid);
if (thread == NULL) {
pr_debug("problem processing COMM event, skipping it.\n");
return -1;
}
if (perf_event__process_comm(tool, event, sample, machine) < 0)
goto out;
if (!evsel->attr.sample_id_all) {
sample->cpu = 0;
sample->time = 0;
sample->tid = event->comm.tid;
sample->pid = event->comm.pid;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_COMM, stdout);
perf_event__fprintf(event, stdout);
ret = 0;
out:
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
thread__put(thread);
return ret;
}
perf script: Add script print support for namespace events Introduce a new option to display events of type PERF_RECORD_NAMESPACES and update perf-script documentation accordingly. Shown below is output (trimmed) of perf script command with the newly introduced option, on perf.data generated with perf record command using --namespaces option. $ perf script --show-namespace-events swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 1/1 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 2/2 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] Commiter notes: Testing it: Investigating that double PERF_RECORD_NAMESPACES for the 19155 pid/tid... Its more than that, there are two PERF_RECORD_COMM as well, and with zeroed timestamps, so probably a synthesizing artifact... # perf script --show-task --show-namespace <SNIP> perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19154/19154 perf 0 [000] 0.000000: PERF_RECORD_FORK(19155:19155):(19154:19154) perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 3110.881834: 1 cycles: ffffffffa7060bf6 native_write_msr (/lib/modules/4.11.0-rc1+/build/vmlinux) <SNIP> Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com> Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Eric Biederman <ebiederm@xmission.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sargun Dhillon <sargun@sargun.me> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/148891932627.25309.1941587059154176221.stgit@hbathini.in.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-03-07 20:42:06 +00:00
static int process_namespaces_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
int ret = -1;
thread = machine__findnew_thread(machine, event->namespaces.pid,
event->namespaces.tid);
if (thread == NULL) {
pr_debug("problem processing NAMESPACES event, skipping it.\n");
return -1;
}
if (perf_event__process_namespaces(tool, event, sample, machine) < 0)
goto out;
if (!evsel->attr.sample_id_all) {
sample->cpu = 0;
sample->time = 0;
sample->tid = event->namespaces.tid;
sample->pid = event->namespaces.pid;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_NAMESPACES, stdout);
perf script: Add script print support for namespace events Introduce a new option to display events of type PERF_RECORD_NAMESPACES and update perf-script documentation accordingly. Shown below is output (trimmed) of perf script command with the newly introduced option, on perf.data generated with perf record command using --namespaces option. $ perf script --show-namespace-events swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 1/1 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 2/2 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] Commiter notes: Testing it: Investigating that double PERF_RECORD_NAMESPACES for the 19155 pid/tid... Its more than that, there are two PERF_RECORD_COMM as well, and with zeroed timestamps, so probably a synthesizing artifact... # perf script --show-task --show-namespace <SNIP> perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19154/19154 perf 0 [000] 0.000000: PERF_RECORD_FORK(19155:19155):(19154:19154) perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 3110.881834: 1 cycles: ffffffffa7060bf6 native_write_msr (/lib/modules/4.11.0-rc1+/build/vmlinux) <SNIP> Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com> Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Eric Biederman <ebiederm@xmission.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sargun Dhillon <sargun@sargun.me> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/148891932627.25309.1941587059154176221.stgit@hbathini.in.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-03-07 20:42:06 +00:00
perf_event__fprintf(event, stdout);
ret = 0;
out:
thread__put(thread);
return ret;
}
static int process_fork_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
if (perf_event__process_fork(tool, event, sample, machine) < 0)
return -1;
thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
if (thread == NULL) {
pr_debug("problem processing FORK event, skipping it.\n");
return -1;
}
if (!evsel->attr.sample_id_all) {
sample->cpu = 0;
sample->time = event->fork.time;
sample->tid = event->fork.tid;
sample->pid = event->fork.pid;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_FORK, stdout);
perf_event__fprintf(event, stdout);
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
thread__put(thread);
return 0;
}
static int process_exit_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
int err = 0;
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
if (thread == NULL) {
pr_debug("problem processing EXIT event, skipping it.\n");
return -1;
}
if (!evsel->attr.sample_id_all) {
sample->cpu = 0;
sample->time = 0;
perf tools: Fix use of wrong event when processing exit events In a couple of cases the 'comm' member of 'union event' has been used instead of the correct member ('fork') when processing exit events. In the cases where it has been used incorrectly, only the 'pid' and 'tid' are affected. The 'pid' value would be correct anyway because it is in the same position in 'comm' and 'fork' events, but the 'tid' would have been incorrectly assigned from 'ppid'. However, for exit events, the kernel puts the current task in the 'ppid' and 'ttid' which is the same as the exiting task. That is 'ppid' == 'pid' and if the task is not multi-threaded, 'pid' == 'tid' i.e. the data goes wrong only when tracing multi-threaded programs. It is hard to find an example of how this would produce an error in practice. There are 3 occurences of the fix: 1. perf script is only affected if !sample_id_all which only happens on old kernels. 2. intel_pt is only affected when decoding without timestamps and would probably still decode correctly - the exit event is only used to flush out data which anyway gets flushed at the end of the session 3. intel_bts also uses the exit event to flush data which would probably not cause errors as it would get flushed at the end of the session instead Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1439888825-27708-1-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-08-18 09:07:05 +00:00
sample->tid = event->fork.tid;
sample->pid = event->fork.pid;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_EXIT, stdout);
perf_event__fprintf(event, stdout);
if (perf_event__process_exit(tool, event, sample, machine) < 0)
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
err = -1;
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
thread__put(thread);
return err;
}
static int process_mmap_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
if (perf_event__process_mmap(tool, event, sample, machine) < 0)
return -1;
thread = machine__findnew_thread(machine, event->mmap.pid, event->mmap.tid);
if (thread == NULL) {
pr_debug("problem processing MMAP event, skipping it.\n");
return -1;
}
if (!evsel->attr.sample_id_all) {
sample->cpu = 0;
sample->time = 0;
sample->tid = event->mmap.tid;
sample->pid = event->mmap.pid;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_MMAP, stdout);
perf_event__fprintf(event, stdout);
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
thread__put(thread);
return 0;
}
static int process_mmap2_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
if (perf_event__process_mmap2(tool, event, sample, machine) < 0)
return -1;
thread = machine__findnew_thread(machine, event->mmap2.pid, event->mmap2.tid);
if (thread == NULL) {
pr_debug("problem processing MMAP2 event, skipping it.\n");
return -1;
}
if (!evsel->attr.sample_id_all) {
sample->cpu = 0;
sample->time = 0;
sample->tid = event->mmap2.tid;
sample->pid = event->mmap2.pid;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_MMAP2, stdout);
perf_event__fprintf(event, stdout);
perf machine: Protect the machine->threads with a rwlock In addition to using refcounts for the struct thread lifetime management, we need to protect access to machine->threads from concurrent access. That happens in 'perf top', where a thread processes events, inserting and deleting entries from that rb_tree while another thread decays hist_entries, that end up dropping references and ultimately deleting threads from the rb_tree and releasing its resources when no further hist_entry (or other data structures, like in 'perf sched') references it. So the rule is the same for refcounts + protected trees in the kernel, get the tree lock, find object, bump the refcount, drop the tree lock, return, use object, drop the refcount if no more use of it is needed, keep it if storing it in some other data structure, drop when releasing that data structure. I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and "perf_event__preprocess_sample(&al)" with "addr_location__put(&al)". The addr_location__put() one is because as we return references to several data structures, we may end up adding more reference counting for the other data structures and then we'll drop it at addr_location__put() time. Acked-by: David Ahern <dsahern@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-06 23:43:22 +00:00
thread__put(thread);
return 0;
}
static int process_switch_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct thread *thread;
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
if (perf_event__process_switch(tool, event, sample, machine) < 0)
return -1;
thread = machine__findnew_thread(machine, sample->pid,
sample->tid);
if (thread == NULL) {
pr_debug("problem processing SWITCH event, skipping it.\n");
return -1;
}
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_SWITCH, stdout);
perf_event__fprintf(event, stdout);
thread__put(thread);
return 0;
}
static int
process_lost_event(struct perf_tool *tool,
union perf_event *event,
struct perf_sample *sample,
struct machine *machine)
{
struct perf_script *script = container_of(tool, struct perf_script, tool);
struct perf_session *session = script->session;
struct perf_evsel *evsel = perf_evlist__id2evsel(session->evlist, sample->id);
struct thread *thread;
thread = machine__findnew_thread(machine, sample->pid,
sample->tid);
if (thread == NULL)
return -1;
perf_sample__fprintf_start(sample, thread, evsel,
PERF_RECORD_LOST, stdout);
perf_event__fprintf(event, stdout);
thread__put(thread);
return 0;
}
static int
process_finished_round_event(struct perf_tool *tool __maybe_unused,
union perf_event *event,
struct ordered_events *oe __maybe_unused)
{
perf_event__fprintf(event, stdout);
return 0;
}
perf tools: Use __maybe_used for unused variables perf defines both __used and __unused variables to use for marking unused variables. The variable __used is defined to __attribute__((__unused__)), which contradicts the kernel definition to __attribute__((__used__)) for new gcc versions. On Android, __used is also defined in system headers and this leads to warnings like: warning: '__used__' attribute ignored __unused is not defined in the kernel and is not a standard definition. If __unused is included everywhere instead of __used, this leads to conflicts with glibc headers, since glibc has a variables with this name in its headers. The best approach is to use __maybe_unused, the definition used in the kernel for __attribute__((unused)). In this way there is only one definition in perf sources (instead of 2 definitions that point to the same thing: __used and __unused) and it works on both Linux and Android. This patch simply replaces all instances of __used and __unused with __maybe_unused. Signed-off-by: Irina Tirdea <irina.tirdea@intel.com> Acked-by: Pekka Enberg <penberg@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Namhyung Kim <namhyung.kim@lge.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/1347315303-29906-7-git-send-email-irina.tirdea@intel.com [ committer note: fixed up conflict with a116e05 in builtin-sched.c ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2012-09-10 22:15:03 +00:00
static void sig_handler(int sig __maybe_unused)
{
session_done = 1;
}
static void perf_script__fclose_per_event_dump(struct perf_script *script)
{
struct perf_evlist *evlist = script->session->evlist;
struct perf_evsel *evsel;
evlist__for_each_entry(evlist, evsel) {
if (!evsel->priv)
break;
perf_evsel_script__delete(evsel->priv);
evsel->priv = NULL;
}
}
static int perf_script__fopen_per_event_dump(struct perf_script *script)
{
struct perf_evsel *evsel;
evlist__for_each_entry(script->session->evlist, evsel) {
/*
* Already setup? I.e. we may be called twice in cases like
* Intel PT, one for the intel_pt// and dummy events, then
* for the evsels syntheized from the auxtrace info.
*
* Ses perf_script__process_auxtrace_info.
*/
if (evsel->priv != NULL)
continue;
evsel->priv = perf_evsel_script__new(evsel, script->session->data);
if (evsel->priv == NULL)
goto out_err_fclose;
}
return 0;
out_err_fclose:
perf_script__fclose_per_event_dump(script);
return -1;
}
static int perf_script__setup_per_event_dump(struct perf_script *script)
{
struct perf_evsel *evsel;
static struct perf_evsel_script es_stdout;
if (script->per_event_dump)
return perf_script__fopen_per_event_dump(script);
es_stdout.fp = stdout;
evlist__for_each_entry(script->session->evlist, evsel)
evsel->priv = &es_stdout;
return 0;
}
static void perf_script__exit_per_event_dump_stats(struct perf_script *script)
{
struct perf_evsel *evsel;
evlist__for_each_entry(script->session->evlist, evsel) {
struct perf_evsel_script *es = evsel->priv;
perf_evsel_script__fprintf(es, stdout);
perf_evsel_script__delete(es);
evsel->priv = NULL;
}
}
static int __cmd_script(struct perf_script *script)
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
{
int ret;
signal(SIGINT, sig_handler);
/* override event processing functions */
if (script->show_task_events) {
script->tool.comm = process_comm_event;
script->tool.fork = process_fork_event;
script->tool.exit = process_exit_event;
}
if (script->show_mmap_events) {
script->tool.mmap = process_mmap_event;
script->tool.mmap2 = process_mmap2_event;
}
if (script->show_switch_events)
script->tool.context_switch = process_switch_event;
perf script: Add script print support for namespace events Introduce a new option to display events of type PERF_RECORD_NAMESPACES and update perf-script documentation accordingly. Shown below is output (trimmed) of perf script command with the newly introduced option, on perf.data generated with perf record command using --namespaces option. $ perf script --show-namespace-events swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 1/1 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 2/2 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] Commiter notes: Testing it: Investigating that double PERF_RECORD_NAMESPACES for the 19155 pid/tid... Its more than that, there are two PERF_RECORD_COMM as well, and with zeroed timestamps, so probably a synthesizing artifact... # perf script --show-task --show-namespace <SNIP> perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19154/19154 perf 0 [000] 0.000000: PERF_RECORD_FORK(19155:19155):(19154:19154) perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 3110.881834: 1 cycles: ffffffffa7060bf6 native_write_msr (/lib/modules/4.11.0-rc1+/build/vmlinux) <SNIP> Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com> Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Eric Biederman <ebiederm@xmission.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sargun Dhillon <sargun@sargun.me> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/148891932627.25309.1941587059154176221.stgit@hbathini.in.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-03-07 20:42:06 +00:00
if (script->show_namespace_events)
script->tool.namespaces = process_namespaces_event;
if (script->show_lost_events)
script->tool.lost = process_lost_event;
if (script->show_round_events) {
script->tool.ordered_events = false;
script->tool.finished_round = process_finished_round_event;
}
if (perf_script__setup_per_event_dump(script)) {
pr_err("Couldn't create the per event dump files\n");
return -1;
}
ret = perf_session__process_events(script->session);
if (script->per_event_dump)
perf_script__exit_per_event_dump_stats(script);
if (debug_mode)
pr_err("Misordered timestamps: %" PRIu64 "\n", nr_unordered);
return ret;
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
struct script_spec {
struct list_head node;
struct scripting_ops *ops;
char spec[0];
};
static LIST_HEAD(script_specs);
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
static struct script_spec *script_spec__new(const char *spec,
struct scripting_ops *ops)
{
struct script_spec *s = malloc(sizeof(*s) + strlen(spec) + 1);
if (s != NULL) {
strcpy(s->spec, spec);
s->ops = ops;
}
return s;
}
static void script_spec__add(struct script_spec *s)
{
list_add_tail(&s->node, &script_specs);
}
static struct script_spec *script_spec__find(const char *spec)
{
struct script_spec *s;
list_for_each_entry(s, &script_specs, node)
if (strcasecmp(s->spec, spec) == 0)
return s;
return NULL;
}
int script_spec_register(const char *spec, struct scripting_ops *ops)
{
struct script_spec *s;
s = script_spec__find(spec);
if (s)
return -1;
s = script_spec__new(spec, ops);
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
if (!s)
return -1;
else
script_spec__add(s);
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
return 0;
}
static struct scripting_ops *script_spec__lookup(const char *spec)
{
struct script_spec *s = script_spec__find(spec);
if (!s)
return NULL;
return s->ops;
}
static void list_available_languages(void)
{
struct script_spec *s;
fprintf(stderr, "\n");
fprintf(stderr, "Scripting language extensions (used in "
"perf script -s [spec:]script.[spec]):\n\n");
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
list_for_each_entry(s, &script_specs, node)
fprintf(stderr, " %-42s [%s]\n", s->spec, s->ops->name);
fprintf(stderr, "\n");
}
perf tools: Use __maybe_used for unused variables perf defines both __used and __unused variables to use for marking unused variables. The variable __used is defined to __attribute__((__unused__)), which contradicts the kernel definition to __attribute__((__used__)) for new gcc versions. On Android, __used is also defined in system headers and this leads to warnings like: warning: '__used__' attribute ignored __unused is not defined in the kernel and is not a standard definition. If __unused is included everywhere instead of __used, this leads to conflicts with glibc headers, since glibc has a variables with this name in its headers. The best approach is to use __maybe_unused, the definition used in the kernel for __attribute__((unused)). In this way there is only one definition in perf sources (instead of 2 definitions that point to the same thing: __used and __unused) and it works on both Linux and Android. This patch simply replaces all instances of __used and __unused with __maybe_unused. Signed-off-by: Irina Tirdea <irina.tirdea@intel.com> Acked-by: Pekka Enberg <penberg@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Namhyung Kim <namhyung.kim@lge.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/1347315303-29906-7-git-send-email-irina.tirdea@intel.com [ committer note: fixed up conflict with a116e05 in builtin-sched.c ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2012-09-10 22:15:03 +00:00
static int parse_scriptname(const struct option *opt __maybe_unused,
const char *str, int unset __maybe_unused)
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
{
char spec[PATH_MAX];
const char *script, *ext;
int len;
if (strcmp(str, "lang") == 0) {
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
list_available_languages();
exit(0);
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
script = strchr(str, ':');
if (script) {
len = script - str;
if (len >= PATH_MAX) {
fprintf(stderr, "invalid language specifier");
return -1;
}
strncpy(spec, str, len);
spec[len] = '\0';
scripting_ops = script_spec__lookup(spec);
if (!scripting_ops) {
fprintf(stderr, "invalid language specifier");
return -1;
}
script++;
} else {
script = str;
ext = strrchr(script, '.');
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
if (!ext) {
fprintf(stderr, "invalid script extension");
return -1;
}
scripting_ops = script_spec__lookup(++ext);
if (!scripting_ops) {
fprintf(stderr, "invalid script extension");
return -1;
}
}
script_name = strdup(script);
return 0;
}
perf tools: Use __maybe_used for unused variables perf defines both __used and __unused variables to use for marking unused variables. The variable __used is defined to __attribute__((__unused__)), which contradicts the kernel definition to __attribute__((__used__)) for new gcc versions. On Android, __used is also defined in system headers and this leads to warnings like: warning: '__used__' attribute ignored __unused is not defined in the kernel and is not a standard definition. If __unused is included everywhere instead of __used, this leads to conflicts with glibc headers, since glibc has a variables with this name in its headers. The best approach is to use __maybe_unused, the definition used in the kernel for __attribute__((unused)). In this way there is only one definition in perf sources (instead of 2 definitions that point to the same thing: __used and __unused) and it works on both Linux and Android. This patch simply replaces all instances of __used and __unused with __maybe_unused. Signed-off-by: Irina Tirdea <irina.tirdea@intel.com> Acked-by: Pekka Enberg <penberg@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Namhyung Kim <namhyung.kim@lge.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/1347315303-29906-7-git-send-email-irina.tirdea@intel.com [ committer note: fixed up conflict with a116e05 in builtin-sched.c ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2012-09-10 22:15:03 +00:00
static int parse_output_fields(const struct option *opt __maybe_unused,
const char *arg, int unset __maybe_unused)
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
{
char *tok, *strtok_saveptr = NULL;
int i, imax = ARRAY_SIZE(all_output_options);
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
int j;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
int rc = 0;
char *str = strdup(arg);
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
int type = -1;
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
enum { DEFAULT, SET, ADD, REMOVE } change = DEFAULT;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
if (!str)
return -ENOMEM;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
/* first word can state for which event type the user is specifying
* the fields. If no type exists, the specified fields apply to all
* event types found in the file minus the invalid fields for a type.
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
*/
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
tok = strchr(str, ':');
if (tok) {
*tok = '\0';
tok++;
if (!strcmp(str, "hw"))
type = PERF_TYPE_HARDWARE;
else if (!strcmp(str, "sw"))
type = PERF_TYPE_SOFTWARE;
else if (!strcmp(str, "trace"))
type = PERF_TYPE_TRACEPOINT;
else if (!strcmp(str, "raw"))
type = PERF_TYPE_RAW;
perf script: Add support for PERF_TYPE_BREAKPOINT Useful for getting stack traces for hardware breakpoint events. Test result: Before this patch: # ~/perf record -g -e mem:0x600980 ./sample [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.011 MB perf.data (12 samples) ] # ~/perf script # ~/perf script -F comm,tid,pid,time,event,ip,sym,dso sample 22520/22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520/22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) ... After this patch: # ~/perf script sample 22520 97457.836294: mem:0x600980: 5a4ad8 __clear_user (/lib/modules/4.3.0-rc4+/build/vmlinux) ... 3f41ba sys_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 979395 return_from_execve (/lib/modules/4.3.0-rc4+/build/vmlinux) 7f1b59719cf7 [unknown] ([unknown]) sample 22520 97457.836648: mem:0x600980: 532 main (/home/w00229757/DataBreakpoints/sample) 21bd5 __libc_start_main (/tmp/oxygen_root-root/lib64/libc-2.18.so) Committer note: So, further testing, lets do it for a kernel global variable, tcp_hashinfo: # grep -w tcp_hashinfo /proc/kallsyms ffffffff8202fc00 B tcp_hashinfo # Note: allow specifying mem:tcp_hashinfo: # perf record -g -e mem:0xffffffff81c65ac0 -a ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.790 MB perf.data ] # # perf evlist mem:0xffffffff8202fc00 # perf evlist -v mem:0xffffffff8202fc00: type: 5, size: 112, { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CALLCHAIN|CPU, disabled: 1, inherit: 1, mmap: 1, comm: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, bp_type: 3, { bp_addr, config1 }: 0xffffffff8202fc00, { bp_len, config2 }: 0x4 # Then, after this patch: # perf script swapper 0 [000] 171036.986988: mem:0xffffffff8202fc00: 8a0fb5 __inet_lookup_established (/lib/modules/4.3.0+/build/vmlinux) 8bc09d tcp_v4_early_demux (/lib/modules/4.3.0+/build/vmlinux) 896def ip_rcv_finish (/lib/modules/4.3.0+/build/vmlinux) 8976c2 ip_rcv (/lib/modules/4.3.0+/build/vmlinux) 855eba __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux) 8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux) 8572a8 process_backlog (/lib/modules/4.3.0+/build/vmlinux) 856b11 net_rx_action (/lib/modules/4.3.0+/build/vmlinux) 2a284b __do_softirq (/lib/modules/4.3.0+/build/vmlinux) 2a2ba3 irq_exit (/lib/modules/4.3.0+/build/vmlinux) 96b7a4 do_IRQ (/lib/modules/4.3.0+/build/vmlinux) 969807 ret_from_intr (/lib/modules/4.3.0+/build/vmlinux) 804c27 cpuidle_enter (/lib/modules/4.3.0+/build/vmlinux) 2ded22 call_cpuidle (/lib/modules/4.3.0+/build/vmlinux) 2defb6 cpu_startup_entry (/lib/modules/4.3.0+/build/vmlinux) 95d5bc rest_init (/lib/modules/4.3.0+/build/vmlinux) 1163ffa start_kernel ([kernel.vmlinux].init.text) 11634d7 x86_64_start_reservations ([kernel.vmlinux].init.text) 1163623 x86_64_start_kernel ([kernel.vmlinux].init.text) Signed-off-by: Wang Nan <wangnan0@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David S. Miller <davem@davemloft.net> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Zefan Li <lizefan@huawei.com> Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1449541544-67621-16-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-12-08 02:25:43 +00:00
else if (!strcmp(str, "break"))
type = PERF_TYPE_BREAKPOINT;
else if (!strcmp(str, "synth"))
type = OUTPUT_TYPE_SYNTH;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
else {
fprintf(stderr, "Invalid event type in field string.\n");
rc = -EINVAL;
goto out;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
}
if (output[type].user_set)
pr_warning("Overriding previous field request for %s events.\n",
event_type(type));
output[type].fields = 0;
output[type].user_set = true;
output[type].wildcard_set = false;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
} else {
tok = str;
if (strlen(str) == 0) {
fprintf(stderr,
"Cannot set fields to 'none' for all event types.\n");
rc = -EINVAL;
goto out;
}
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
/* Don't override defaults for +- */
if (strchr(str, '+') || strchr(str, '-'))
goto parse;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (output_set_by_user())
pr_warning("Overriding previous field request for all events.\n");
for (j = 0; j < OUTPUT_TYPE_MAX; ++j) {
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
output[j].fields = 0;
output[j].user_set = true;
output[j].wildcard_set = true;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
parse:
for (tok = strtok_r(tok, ",", &strtok_saveptr); tok; tok = strtok_r(NULL, ",", &strtok_saveptr)) {
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
if (*tok == '+') {
if (change == SET)
goto out_badmix;
change = ADD;
tok++;
} else if (*tok == '-') {
if (change == SET)
goto out_badmix;
change = REMOVE;
tok++;
} else {
if (change != SET && change != DEFAULT)
goto out_badmix;
change = SET;
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
for (i = 0; i < imax; ++i) {
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (strcmp(tok, all_output_options[i].str) == 0)
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
break;
}
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
if (i == imax && strcmp(tok, "flags") == 0) {
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
print_flags = change == REMOVE ? false : true;
perf script: Add field option 'flags' to print sample flags Instruction tracing will typically have access to information about the instruction being executed for a particular ip sample. Some of that information will be available in the 'flags' member of struct perf_sample. With the addition of transactions events synthesis to Instruction Tracing options, there is a need to be able easily to see the flags because they show whether the ip is at the start, commit or abort of a tranasaction. Consequently add an option to display the flags. The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. Example using Intel PT: perf script -fip,time,event,sym,addr,flags ... 1288.721584105: branches:u: bo 401146 main => 401152 main 1288.721584105: transactions: x 0 401164 main 1288.721584105: branches:u: bx 40117c main => 40119b main 1288.721584105: branches:u: box 4011a4 main => 40117e main 1288.721584105: branches:u: bcx 401187 main => 401094 g ... 1288.721591645: branches:u: bx 4010c4 g => 4010cb g 1288.721591645: branches:u: brx 4010cc g => 401189 main 1288.721591645: transactions: 0 4011a6 main 1288.721593199: branches:u: b 4011a9 main => 4011af main 1288.721593199: branches:u: bo 4011bc main => 40113e main 1288.721593199: branches:u: b 401150 main => 40115a main 1288.721593199: transactions: x 0 401164 main 1288.721593199: branches:u: bx 40117c main => 40119b main 1288.721593199: branches:u: box 4011a4 main => 40117e main 1288.721593199: branches:u: bcx 401187 main => 40105e f ... 1288.722284747: branches:u: brx 401093 f => 401189 main 1288.722284747: branches:u: box 4011a4 main => 40117e main 1288.722284747: branches:u: bcx 401187 main => 40105e f 1288.722285883: transactions: bA 0 401071 f 1288.722285883: branches:u: bA 401071 f => 40116a main 1288.722285883: branches:u: bE 40116a main => 0 [unknown] 1288.722297174: branches:u: bB 0 [unknown] => 40116a main ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1428594864-29309-26-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-09 15:54:05 +00:00
continue;
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
if (i == imax) {
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
fprintf(stderr, "Invalid field requested.\n");
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
rc = -EINVAL;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
goto out;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (type == -1) {
/* add user option to all events types for
* which it is valid
*/
for (j = 0; j < OUTPUT_TYPE_MAX; ++j) {
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (output[j].invalid_fields & all_output_options[i].field) {
pr_warning("\'%s\' not valid for %s events. Ignoring.\n",
all_output_options[i].str, event_type(j));
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
} else {
if (change == REMOVE)
output[j].fields &= ~all_output_options[i].field;
else
output[j].fields |= all_output_options[i].field;
perf script: Allow sym and dso without ip, addr Currently sym and dso require printing ip and addr because the print function is tied to those outputs. With callindent it makes sense to print the symbol or dso without numerical IP or ADDR. So change the dependency check to only check the underlying attribute. Also the branch target output relies on the user_set flag to determine if the branch target should be implicitely printed. When modifying the fields with + or - also set user_set, so that ADDR can be removed. We also need to set wildcard_set to make the initial sanity check pass. This allows to remove a lot of noise in callindent output by dropping the numerical addresses, which are not all that useful. Before % perf script --itrace=cr -F +callindent swapper 0 [000] 156546.354971: 1 branches: pt_config 0 [unknown] ([unknown]) => ffffffff81010486 pt_config ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_config ffffffff81010499 pt_config ([kernel.kallsyms]) => ffffffff8101063e pt_event_add ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_event_add ffffffff81010635 pt_event_add ([kernel.kallsyms]) => ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable ffffffff8115e726 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff811579b0 perf_pmu_enable ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void ffffffff81151730 perf_pmu_nop_void ([kernel.kallsyms]) => ffffffff8115e72b event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 ffffffff8115e737 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff8115e7a5 group_sched_in ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax ffffffff8115e7f6 group_sched_in ([kernel.kallsyms]) => ffffffff81a03000 __x86_indirect_thunk_rax ([kernel.kallsyms]) After % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_event_add swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_int swapper 0 [000] 156546.354971: 1 branches: group_sched_in swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: group_sched_in Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:09 +00:00
output[j].user_set = true;
output[j].wildcard_set = true;
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
}
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
}
} else {
if (output[type].invalid_fields & all_output_options[i].field) {
fprintf(stderr, "\'%s\' not valid for %s events.\n",
all_output_options[i].str, event_type(type));
rc = -EINVAL;
goto out;
}
perf script: Allow sym and dso without ip, addr Currently sym and dso require printing ip and addr because the print function is tied to those outputs. With callindent it makes sense to print the symbol or dso without numerical IP or ADDR. So change the dependency check to only check the underlying attribute. Also the branch target output relies on the user_set flag to determine if the branch target should be implicitely printed. When modifying the fields with + or - also set user_set, so that ADDR can be removed. We also need to set wildcard_set to make the initial sanity check pass. This allows to remove a lot of noise in callindent output by dropping the numerical addresses, which are not all that useful. Before % perf script --itrace=cr -F +callindent swapper 0 [000] 156546.354971: 1 branches: pt_config 0 [unknown] ([unknown]) => ffffffff81010486 pt_config ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_config ffffffff81010499 pt_config ([kernel.kallsyms]) => ffffffff8101063e pt_event_add ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: pt_event_add ffffffff81010635 pt_event_add ([kernel.kallsyms]) => ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable ffffffff8115e726 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff811579b0 perf_pmu_enable ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void ffffffff81151730 perf_pmu_nop_void ([kernel.kallsyms]) => ffffffff8115e72b event_sched_in.isra.107 ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 ffffffff8115e737 event_sched_in.isra.107 ([kernel.kallsyms]) => ffffffff8115e7a5 group_sched_in ([kernel.kallsyms]) swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax ffffffff8115e7f6 group_sched_in ([kernel.kallsyms]) => ffffffff81a03000 __x86_indirect_thunk_rax ([kernel.kallsyms]) After % perf script --itrace=cr -F +callindent,-ip,-sym,-symoff swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_config swapper 0 [000] 156546.354971: 1 branches: pt_event_add swapper 0 [000] 156546.354971: 1 branches: perf_pmu_enable swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_void swapper 0 [000] 156546.354971: 1 branches: event_sched_in.isra.107 swapper 0 [000] 156546.354971: 1 branches: __x86_indirect_thunk_rax swapper 0 [000] 156546.354971: 1 branches: perf_pmu_nop_int swapper 0 [000] 156546.354971: 1 branches: group_sched_in swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: event_filter_match swapper 0 [000] 156546.354971: 1 branches: group_sched_in Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180918123214.26728-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-18 12:32:09 +00:00
output[type].user_set = true;
output[type].wildcard_set = true;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
}
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (type >= 0) {
if (output[type].fields == 0) {
pr_debug("No fields requested for %s type. "
"Events will not be displayed.\n", event_type(type));
}
}
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
goto out;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
out_badmix:
fprintf(stderr, "Cannot mix +-field with overridden fields\n");
rc = -EINVAL;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
out:
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
free(str);
return rc;
}
#define for_each_lang(scripts_path, scripts_dir, lang_dirent) \
while ((lang_dirent = readdir(scripts_dir)) != NULL) \
if ((lang_dirent->d_type == DT_DIR || \
(lang_dirent->d_type == DT_UNKNOWN && \
is_directory(scripts_path, lang_dirent))) && \
(strcmp(lang_dirent->d_name, ".")) && \
(strcmp(lang_dirent->d_name, "..")))
#define for_each_script(lang_path, lang_dir, script_dirent) \
while ((script_dirent = readdir(lang_dir)) != NULL) \
if (script_dirent->d_type != DT_DIR && \
(script_dirent->d_type != DT_UNKNOWN || \
!is_directory(lang_path, script_dirent)))
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
#define RECORD_SUFFIX "-record"
#define REPORT_SUFFIX "-report"
struct script_desc {
struct list_head node;
char *name;
char *half_liner;
char *args;
};
static LIST_HEAD(script_descs);
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
static struct script_desc *script_desc__new(const char *name)
{
struct script_desc *s = zalloc(sizeof(*s));
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (s != NULL && name)
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
s->name = strdup(name);
return s;
}
static void script_desc__delete(struct script_desc *s)
{
zfree(&s->name);
zfree(&s->half_liner);
zfree(&s->args);
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
free(s);
}
static void script_desc__add(struct script_desc *s)
{
list_add_tail(&s->node, &script_descs);
}
static struct script_desc *script_desc__find(const char *name)
{
struct script_desc *s;
list_for_each_entry(s, &script_descs, node)
if (strcasecmp(s->name, name) == 0)
return s;
return NULL;
}
static struct script_desc *script_desc__findnew(const char *name)
{
struct script_desc *s = script_desc__find(name);
if (s)
return s;
s = script_desc__new(name);
if (!s)
return NULL;
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
script_desc__add(s);
return s;
}
static const char *ends_with(const char *str, const char *suffix)
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
{
size_t suffix_len = strlen(suffix);
const char *p = str;
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
if (strlen(str) > suffix_len) {
p = str + strlen(str) - suffix_len;
if (!strncmp(p, suffix, suffix_len))
return p;
}
return NULL;
}
static int read_script_info(struct script_desc *desc, const char *filename)
{
char line[BUFSIZ], *p;
FILE *fp;
fp = fopen(filename, "r");
if (!fp)
return -1;
while (fgets(line, sizeof(line), fp)) {
p = ltrim(line);
if (strlen(p) == 0)
continue;
if (*p != '#')
continue;
p++;
if (strlen(p) && *p == '!')
continue;
p = ltrim(p);
if (strlen(p) && p[strlen(p) - 1] == '\n')
p[strlen(p) - 1] = '\0';
if (!strncmp(p, "description:", strlen("description:"))) {
p += strlen("description:");
desc->half_liner = strdup(ltrim(p));
continue;
}
if (!strncmp(p, "args:", strlen("args:"))) {
p += strlen("args:");
desc->args = strdup(ltrim(p));
continue;
}
}
fclose(fp);
return 0;
}
static char *get_script_root(struct dirent *script_dirent, const char *suffix)
{
char *script_root, *str;
script_root = strdup(script_dirent->d_name);
if (!script_root)
return NULL;
str = (char *)ends_with(script_root, suffix);
if (!str) {
free(script_root);
return NULL;
}
*str = '\0';
return script_root;
}
perf tools: Use __maybe_used for unused variables perf defines both __used and __unused variables to use for marking unused variables. The variable __used is defined to __attribute__((__unused__)), which contradicts the kernel definition to __attribute__((__used__)) for new gcc versions. On Android, __used is also defined in system headers and this leads to warnings like: warning: '__used__' attribute ignored __unused is not defined in the kernel and is not a standard definition. If __unused is included everywhere instead of __used, this leads to conflicts with glibc headers, since glibc has a variables with this name in its headers. The best approach is to use __maybe_unused, the definition used in the kernel for __attribute__((unused)). In this way there is only one definition in perf sources (instead of 2 definitions that point to the same thing: __used and __unused) and it works on both Linux and Android. This patch simply replaces all instances of __used and __unused with __maybe_unused. Signed-off-by: Irina Tirdea <irina.tirdea@intel.com> Acked-by: Pekka Enberg <penberg@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Namhyung Kim <namhyung.kim@lge.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/1347315303-29906-7-git-send-email-irina.tirdea@intel.com [ committer note: fixed up conflict with a116e05 in builtin-sched.c ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2012-09-10 22:15:03 +00:00
static int list_available_scripts(const struct option *opt __maybe_unused,
const char *s __maybe_unused,
int unset __maybe_unused)
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
{
struct dirent *script_dirent, *lang_dirent;
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
char scripts_path[MAXPATHLEN];
DIR *scripts_dir, *lang_dir;
char script_path[MAXPATHLEN];
char lang_path[MAXPATHLEN];
struct script_desc *desc;
char first_half[BUFSIZ];
char *script_root;
snprintf(scripts_path, MAXPATHLEN, "%s/scripts", get_argv_exec_path());
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
scripts_dir = opendir(scripts_path);
if (!scripts_dir) {
fprintf(stdout,
"open(%s) failed.\n"
"Check \"PERF_EXEC_PATH\" env to set scripts dir.\n",
scripts_path);
exit(-1);
}
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
for_each_lang(scripts_path, scripts_dir, lang_dirent) {
scnprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path,
lang_dirent->d_name);
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
lang_dir = opendir(lang_path);
if (!lang_dir)
continue;
for_each_script(lang_path, lang_dir, script_dirent) {
script_root = get_script_root(script_dirent, REPORT_SUFFIX);
if (script_root) {
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
desc = script_desc__findnew(script_root);
scnprintf(script_path, MAXPATHLEN, "%s/%s",
lang_path, script_dirent->d_name);
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
read_script_info(desc, script_path);
free(script_root);
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
}
}
}
fprintf(stdout, "List of available trace scripts:\n");
list_for_each_entry(desc, &script_descs, node) {
sprintf(first_half, "%s %s", desc->name,
desc->args ? desc->args : "");
fprintf(stdout, " %-36s %s\n", first_half,
desc->half_liner ? desc->half_liner : "");
}
exit(0);
}
/*
* Some scripts specify the required events in their "xxx-record" file,
* this function will check if the events in perf.data match those
* mentioned in the "xxx-record".
*
* Fixme: All existing "xxx-record" are all in good formats "-e event ",
* which is covered well now. And new parsing code should be added to
* cover the future complexing formats like event groups etc.
*/
static int check_ev_match(char *dir_name, char *scriptname,
struct perf_session *session)
{
char filename[MAXPATHLEN], evname[128];
char line[BUFSIZ], *p;
struct perf_evsel *pos;
int match, len;
FILE *fp;
scnprintf(filename, MAXPATHLEN, "%s/bin/%s-record", dir_name, scriptname);
fp = fopen(filename, "r");
if (!fp)
return -1;
while (fgets(line, sizeof(line), fp)) {
p = ltrim(line);
if (*p == '#')
continue;
while (strlen(p)) {
p = strstr(p, "-e");
if (!p)
break;
p += 2;
p = ltrim(p);
len = strcspn(p, " \t");
if (!len)
break;
snprintf(evname, len + 1, "%s", p);
match = 0;
evlist__for_each_entry(session->evlist, pos) {
if (!strcmp(perf_evsel__name(pos), evname)) {
match = 1;
break;
}
}
if (!match) {
fclose(fp);
return -1;
}
}
}
fclose(fp);
return 0;
}
/*
* Return -1 if none is found, otherwise the actual scripts number.
*
* Currently the only user of this function is the script browser, which
* will list all statically runnable scripts, select one, execute it and
* show the output in a perf browser.
*/
int find_scripts(char **scripts_array, char **scripts_path_array)
{
struct dirent *script_dirent, *lang_dirent;
char scripts_path[MAXPATHLEN], lang_path[MAXPATHLEN];
DIR *scripts_dir, *lang_dir;
struct perf_session *session;
struct perf_data data = {
.file = {
.path = input_name,
},
.mode = PERF_DATA_MODE_READ,
};
char *temp;
int i = 0;
session = perf_session__new(&data, false, NULL);
if (!session)
return -1;
snprintf(scripts_path, MAXPATHLEN, "%s/scripts", get_argv_exec_path());
scripts_dir = opendir(scripts_path);
if (!scripts_dir) {
perf_session__delete(session);
return -1;
}
for_each_lang(scripts_path, scripts_dir, lang_dirent) {
scnprintf(lang_path, MAXPATHLEN, "%s/%s", scripts_path,
lang_dirent->d_name);
#ifndef HAVE_LIBPERL_SUPPORT
if (strstr(lang_path, "perl"))
continue;
#endif
#ifndef HAVE_LIBPYTHON_SUPPORT
if (strstr(lang_path, "python"))
continue;
#endif
lang_dir = opendir(lang_path);
if (!lang_dir)
continue;
for_each_script(lang_path, lang_dir, script_dirent) {
/* Skip those real time scripts: xxxtop.p[yl] */
if (strstr(script_dirent->d_name, "top."))
continue;
sprintf(scripts_path_array[i], "%s/%s", lang_path,
script_dirent->d_name);
temp = strchr(script_dirent->d_name, '.');
snprintf(scripts_array[i],
(temp - script_dirent->d_name) + 1,
"%s", script_dirent->d_name);
if (check_ev_match(lang_path,
scripts_array[i], session))
continue;
i++;
}
closedir(lang_dir);
}
closedir(scripts_dir);
perf_session__delete(session);
return i;
}
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
static char *get_script_path(const char *script_root, const char *suffix)
{
struct dirent *script_dirent, *lang_dirent;
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
char scripts_path[MAXPATHLEN];
char script_path[MAXPATHLEN];
DIR *scripts_dir, *lang_dir;
char lang_path[MAXPATHLEN];
char *__script_root;
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
snprintf(scripts_path, MAXPATHLEN, "%s/scripts", get_argv_exec_path());
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
scripts_dir = opendir(scripts_path);
if (!scripts_dir)
return NULL;
for_each_lang(scripts_path, scripts_dir, lang_dirent) {
scnprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path,
lang_dirent->d_name);
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
lang_dir = opendir(lang_path);
if (!lang_dir)
continue;
for_each_script(lang_path, lang_dir, script_dirent) {
__script_root = get_script_root(script_dirent, suffix);
if (__script_root && !strcmp(script_root, __script_root)) {
free(__script_root);
closedir(lang_dir);
closedir(scripts_dir);
scnprintf(script_path, MAXPATHLEN, "%s/%s",
lang_path, script_dirent->d_name);
return strdup(script_path);
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
}
free(__script_root);
}
closedir(lang_dir);
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
}
closedir(scripts_dir);
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
return NULL;
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
static bool is_top_script(const char *script_path)
{
return ends_with(script_path, "top") == NULL ? false : true;
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
}
static int has_required_arg(char *script_path)
{
struct script_desc *desc;
int n_args = 0;
char *p;
desc = script_desc__new(NULL);
if (read_script_info(desc, script_path))
goto out;
if (!desc->args)
goto out;
for (p = desc->args; *p; p++)
if (*p == '<')
n_args++;
out:
script_desc__delete(desc);
return n_args;
}
static int have_cmd(int argc, const char **argv)
{
char **__argv = malloc(sizeof(const char *) * argc);
if (!__argv) {
pr_err("malloc failed\n");
return -1;
}
memcpy(__argv, argv, sizeof(const char *) * argc);
argc = parse_options(argc, (const char **)__argv, record_options,
NULL, PARSE_OPT_STOP_AT_NON_OPTION);
free(__argv);
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
system_wide = (argc == 0);
return 0;
}
static void script__setup_sample_type(struct perf_script *script)
{
struct perf_session *session = script->session;
u64 sample_type = perf_evlist__combined_sample_type(session->evlist);
if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain) {
if ((sample_type & PERF_SAMPLE_REGS_USER) &&
perf unwind: Do not look just at the global callchain_param.record_mode When setting up DWARF callchains on specific events, without using 'record' or 'trace' --call-graph, but instead doing it like: perf trace -e cycles/call-graph=dwarf/ The unwind__prepare_access() call in thread__insert_map() when we process PERF_RECORD_MMAP(2) metadata events were not being performed, precluding us from using per-event DWARF callchains, handling them just when we asked for all events to be DWARF, using "--call-graph dwarf". We do it in the PERF_RECORD_MMAP because we have to look at one of the executable maps to figure out the executable type (64-bit, 32-bit) of the DSO laid out in that mmap. Also to look at the architecture where the perf.data file was recorded. All this probably should be deferred to when we process a sample for some thread that has callchains, so that we do this processing only for the threads with samples, not for all of them. For now, fix using DWARF on specific events. Before: # perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.048 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.048/0.048/0.048/0.000 ms 0.000 probe_libc:inet_pton:(7fe9597bb350)) Problem processing probe_libc:inet_pton callchain, skipping... # After: # perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.060 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.060/0.060/0.060/0.000 ms 0.000 probe_libc:inet_pton:(7fd4aa930350)) __inet_pton (inlined) gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so) __GI_getaddrinfo (inlined) [0xffffaa804e51af3f] (/usr/bin/ping) __libc_start_main (/usr/lib64/libc-2.26.so) [0xffffaa804e51b379] (/usr/bin/ping) # # perf trace --call-graph=dwarf --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.057 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.057/0.057/0.057/0.000 ms 0.000 probe_libc:inet_pton:(7f9363b9e350)) __inet_pton (inlined) gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so) __GI_getaddrinfo (inlined) [0xffffa9e8a14e0f3f] (/usr/bin/ping) __libc_start_main (/usr/lib64/libc-2.26.so) [0xffffa9e8a14e1379] (/usr/bin/ping) # # perf trace --call-graph=fp --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.077 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms 0.000 probe_libc:inet_pton:(7f4947e1c350)) __inet_pton (inlined) gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so) __GI_getaddrinfo (inlined) [0xffffaa716d88ef3f] (/usr/bin/ping) __libc_start_main (/usr/lib64/libc-2.26.so) [0xffffaa716d88f379] (/usr/bin/ping) # # perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=fp/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.078 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.078/0.078/0.078/0.000 ms 0.000 probe_libc:inet_pton:(7fa157696350)) __GI___inet_pton (/usr/lib64/libc-2.26.so) getaddrinfo (/usr/lib64/libc-2.26.so) [0xffffa9ba39c74f40] (/usr/bin/ping) # Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Hendrick Brueckner <brueckner@linux.vnet.ibm.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Thomas Richter <tmricht@linux.vnet.ibm.com> Cc: Wang Nan <wangnan0@huawei.com> Link: https://lkml.kernel.org/r/20180116182650.GE16107@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-01-15 19:48:46 +00:00
(sample_type & PERF_SAMPLE_STACK_USER)) {
callchain_param.record_mode = CALLCHAIN_DWARF;
perf unwind: Do not look just at the global callchain_param.record_mode When setting up DWARF callchains on specific events, without using 'record' or 'trace' --call-graph, but instead doing it like: perf trace -e cycles/call-graph=dwarf/ The unwind__prepare_access() call in thread__insert_map() when we process PERF_RECORD_MMAP(2) metadata events were not being performed, precluding us from using per-event DWARF callchains, handling them just when we asked for all events to be DWARF, using "--call-graph dwarf". We do it in the PERF_RECORD_MMAP because we have to look at one of the executable maps to figure out the executable type (64-bit, 32-bit) of the DSO laid out in that mmap. Also to look at the architecture where the perf.data file was recorded. All this probably should be deferred to when we process a sample for some thread that has callchains, so that we do this processing only for the threads with samples, not for all of them. For now, fix using DWARF on specific events. Before: # perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.048 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.048/0.048/0.048/0.000 ms 0.000 probe_libc:inet_pton:(7fe9597bb350)) Problem processing probe_libc:inet_pton callchain, skipping... # After: # perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.060 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.060/0.060/0.060/0.000 ms 0.000 probe_libc:inet_pton:(7fd4aa930350)) __inet_pton (inlined) gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so) __GI_getaddrinfo (inlined) [0xffffaa804e51af3f] (/usr/bin/ping) __libc_start_main (/usr/lib64/libc-2.26.so) [0xffffaa804e51b379] (/usr/bin/ping) # # perf trace --call-graph=dwarf --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.057 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.057/0.057/0.057/0.000 ms 0.000 probe_libc:inet_pton:(7f9363b9e350)) __inet_pton (inlined) gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so) __GI_getaddrinfo (inlined) [0xffffa9e8a14e0f3f] (/usr/bin/ping) __libc_start_main (/usr/lib64/libc-2.26.so) [0xffffa9e8a14e1379] (/usr/bin/ping) # # perf trace --call-graph=fp --no-syscalls -e probe_libc:inet_pton/call-graph=dwarf/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.077 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.077/0.077/0.077/0.000 ms 0.000 probe_libc:inet_pton:(7f4947e1c350)) __inet_pton (inlined) gaih_inet.constprop.7 (/usr/lib64/libc-2.26.so) __GI_getaddrinfo (inlined) [0xffffaa716d88ef3f] (/usr/bin/ping) __libc_start_main (/usr/lib64/libc-2.26.so) [0xffffaa716d88f379] (/usr/bin/ping) # # perf trace --no-syscalls -e probe_libc:inet_pton/call-graph=fp/ ping -6 -c 1 ::1 PING ::1(::1) 56 data bytes 64 bytes from ::1: icmp_seq=1 ttl=64 time=0.078 ms --- ::1 ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.078/0.078/0.078/0.000 ms 0.000 probe_libc:inet_pton:(7fa157696350)) __GI___inet_pton (/usr/lib64/libc-2.26.so) getaddrinfo (/usr/lib64/libc-2.26.so) [0xffffa9ba39c74f40] (/usr/bin/ping) # Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Hendrick Brueckner <brueckner@linux.vnet.ibm.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Thomas Richter <tmricht@linux.vnet.ibm.com> Cc: Wang Nan <wangnan0@huawei.com> Link: https://lkml.kernel.org/r/20180116182650.GE16107@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-01-15 19:48:46 +00:00
dwarf_callchain_users = true;
} else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
callchain_param.record_mode = CALLCHAIN_LBR;
else
callchain_param.record_mode = CALLCHAIN_FP;
}
}
static int process_stat_round_event(struct perf_session *session,
union perf_event *event)
{
struct stat_round_event *round = &event->stat_round;
struct perf_evsel *counter;
evlist__for_each_entry(session->evlist, counter) {
perf_stat_process_counter(&stat_config, counter);
process_stat(counter, round->time);
}
process_stat_interval(round->time);
return 0;
}
static int process_stat_config_event(struct perf_session *session __maybe_unused,
union perf_event *event)
{
perf_event__read_stat_config(&stat_config, &event->stat_config);
return 0;
}
static int set_maps(struct perf_script *script)
{
struct perf_evlist *evlist = script->session->evlist;
if (!script->cpus || !script->threads)
return 0;
if (WARN_ONCE(script->allocated, "stats double allocation\n"))
return -EINVAL;
perf_evlist__set_maps(evlist, script->cpus, script->threads);
if (perf_evlist__alloc_stats(evlist, true))
return -ENOMEM;
script->allocated = true;
return 0;
}
static
int process_thread_map_event(struct perf_session *session,
union perf_event *event)
{
struct perf_tool *tool = session->tool;
struct perf_script *script = container_of(tool, struct perf_script, tool);
if (script->threads) {
pr_warning("Extra thread map event, ignoring.\n");
return 0;
}
script->threads = thread_map__new_event(&event->thread_map);
if (!script->threads)
return -ENOMEM;
return set_maps(script);
}
static
int process_cpu_map_event(struct perf_session *session,
union perf_event *event)
{
struct perf_tool *tool = session->tool;
struct perf_script *script = container_of(tool, struct perf_script, tool);
if (script->cpus) {
pr_warning("Extra cpu map event, ignoring.\n");
return 0;
}
script->cpus = cpu_map__new_data(&event->cpu_map.data);
if (!script->cpus)
return -ENOMEM;
return set_maps(script);
}
static int process_feature_event(struct perf_session *session,
union perf_event *event)
{
if (event->feat.feat_id < HEADER_LAST_FEATURE)
return perf_event__process_feature(session, event);
return 0;
}
#ifdef HAVE_AUXTRACE_SUPPORT
static int perf_script__process_auxtrace_info(struct perf_session *session,
union perf_event *event)
{
struct perf_tool *tool = session->tool;
int ret = perf_event__process_auxtrace_info(session, event);
if (ret == 0) {
struct perf_script *script = container_of(tool, struct perf_script, tool);
ret = perf_script__setup_per_event_dump(script);
}
return ret;
}
#else
#define perf_script__process_auxtrace_info 0
#endif
perf script: Add --insn-trace for instruction decoding Add a --insn-trace short hand option for decoding and disassembling instruction streams for intel_pt. This automatically pipes the output into the xed disassembler to generate disassembled instructions. This just makes this use model much nicer to use. Before % perf record -e intel_pt// ... % perf script --itrace=i0ns --ns -F +insn,-event,-period | xed -F insn: -A -64 swapper 0 [000] 17276.429606186: ffffffff81010486 pt_config ([kernel.kallsyms]) nopl %eax, (%rax,%rax,1) swapper 0 [000] 17276.429606186: ffffffff8101048b pt_config ([kernel.kallsyms]) add $0x10, %rsp swapper 0 [000] 17276.429606186: ffffffff8101048f pt_config ([kernel.kallsyms]) popq %rbx swapper 0 [000] 17276.429606186: ffffffff81010490 pt_config ([kernel.kallsyms]) popq %rbp swapper 0 [000] 17276.429606186: ffffffff81010491 pt_config ([kernel.kallsyms]) popq %r12 swapper 0 [000] 17276.429606186: ffffffff81010493 pt_config ([kernel.kallsyms]) popq %r13 swapper 0 [000] 17276.429606186: ffffffff81010495 pt_config ([kernel.kallsyms]) popq %r14 swapper 0 [000] 17276.429606186: ffffffff81010497 pt_config ([kernel.kallsyms]) popq %r15 swapper 0 [000] 17276.429606186: ffffffff81010499 pt_config ([kernel.kallsyms]) retq swapper 0 [000] 17276.429606186: ffffffff8101063e pt_event_add ([kernel.kallsyms]) cmpl $0x1, 0x1b0(%rbx) swapper 0 [000] 17276.429606186: ffffffff81010645 pt_event_add ([kernel.kallsyms]) mov $0xffffffea, %eax swapper 0 [000] 17276.429606186: ffffffff8101064a pt_event_add ([kernel.kallsyms]) mov $0x0, %edx swapper 0 [000] 17276.429606186: ffffffff8101064f pt_event_add ([kernel.kallsyms]) popq %rbx swapper 0 [000] 17276.429606186: ffffffff81010650 pt_event_add ([kernel.kallsyms]) cmovnz %edx, %eax swapper 0 [000] 17276.429606186: ffffffff81010653 pt_event_add ([kernel.kallsyms]) jmp 0xffffffff81010635 swapper 0 [000] 17276.429606186: ffffffff81010635 pt_event_add ([kernel.kallsyms]) retq swapper 0 [000] 17276.429606186: ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) test %eax, %eax Now: % perf record -e intel_pt// ... % perf script --insn-trace --xed ... same output ... XED needs to be installed with: $ git clone https://github.com/intelxed/mbuild.git mbuild $ git clone https://github.com/intelxed/xed $ cd xed $ ./mfile.py $ ./mfile.py examples $ sudo ./mfile.py --prefix=/usr/local install $ sudo cp obj/examples/xed /usr/local/bin $ xed | head -3 ERROR: required argument(s) were missing Copyright (C) 2017, Intel Corporation. All rights reserved. XED version: [v10.0-328-g7d62c8c49b7b] $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20180920180540.14039-2-andi@firstfloor.org [ Fixed up whitespace damage, added the 'mfile.py examples + cp obj/examples/xed ... ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:36 +00:00
static int parse_insn_trace(const struct option *opt __maybe_unused,
const char *str __maybe_unused,
int unset __maybe_unused)
{
parse_output_fields(NULL, "+insn,-event,-period", 0);
itrace_parse_synth_opts(opt, "i0ns", 0);
nanosecs = true;
return 0;
}
static int parse_xed(const struct option *opt __maybe_unused,
const char *str __maybe_unused,
int unset __maybe_unused)
{
force_pager("xed -F insn: -A -64 | less");
return 0;
}
tools script: Add --call-trace and --call-ret-trace Add short cut options to print PT call trace and call-ret-trace, for calls and call and returns. Roughly corresponds to ftrace function tracer and function graph tracer. Just makes these common use cases nicer to use. % perf record -a -e intel_pt// sleep 1 % perf script --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_filter_match perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage % perf script --call-ret-trace perf 900 [000] 194167.205652203: tr strt ([unknown]) pt_config perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) pt_config perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) pt_event_add perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) perf_pmu_nop_void perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) perf_pmu_nop_int perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) event_filter_match perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) event_filter_match perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) perf_pmu_nop_txn perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) perf_event_set_state.part.71 Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:38 +00:00
static int parse_call_trace(const struct option *opt __maybe_unused,
const char *str __maybe_unused,
int unset __maybe_unused)
{
parse_output_fields(NULL, "-ip,-addr,-event,-period,+callindent", 0);
itrace_parse_synth_opts(opt, "cewp", 0);
nanosecs = true;
return 0;
}
static int parse_callret_trace(const struct option *opt __maybe_unused,
const char *str __maybe_unused,
int unset __maybe_unused)
{
parse_output_fields(NULL, "-ip,-addr,-event,-period,+callindent,+flags", 0);
itrace_parse_synth_opts(opt, "crewp", 0);
nanosecs = true;
return 0;
}
int cmd_script(int argc, const char **argv)
{
bool show_full_info = false;
bool header = false;
bool header_only = false;
bool script_started = false;
char *rec_script_path = NULL;
char *rep_script_path = NULL;
struct perf_session *session;
struct itrace_synth_opts itrace_synth_opts = {
.set = false,
.default_no_sample = true,
};
char *script_path = NULL;
const char **__argv;
int i, j, err = 0;
struct perf_script script = {
.tool = {
.sample = process_sample_event,
.mmap = perf_event__process_mmap,
.mmap2 = perf_event__process_mmap2,
.comm = perf_event__process_comm,
perf tools: Add PERF_RECORD_NAMESPACES to include namespaces related info Introduce a new option to record PERF_RECORD_NAMESPACES events emitted by the kernel when fork, clone, setns or unshare are invoked. And update perf-record documentation with the new option to record namespace events. Committer notes: Combined it with a later patch to allow printing it via 'perf report -D' and be able to test the feature introduced in this patch. Had to move here also perf_ns__name(), that was introduced in another later patch. Also used PRIu64 and PRIx64 to fix the build in some enfironments wrt: util/event.c:1129:39: error: format '%lx' expects argument of type 'long unsigned int', but argument 6 has type 'long long unsigned int' [-Werror=format=] ret += fprintf(fp, "%u/%s: %lu/0x%lx%s", idx ^ Testing it: # perf record --namespaces -a ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.083 MB perf.data (423 samples) ] # # perf report -D <SNIP> 3 2028902078892 0x115140 [0xa0]: PERF_RECORD_NAMESPACES 14783/14783 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] 0x1151e0 [0x30]: event: 9 . . ... raw event: size 48 bytes . 0000: 09 00 00 00 02 00 30 00 c4 71 82 68 0c 7f 00 00 ......0..q.h.... . 0010: a9 39 00 00 a9 39 00 00 94 28 fe 63 d8 01 00 00 .9...9...(.c.... . 0020: 03 00 00 00 00 00 00 00 ce c4 02 00 00 00 00 00 ................ <SNIP> NAMESPACES events: 1 <SNIP> # Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com> Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Eric Biederman <ebiederm@xmission.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sargun Dhillon <sargun@sargun.me> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/148891930386.25309.18412039920746995488.stgit@hbathini.in.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-03-07 20:41:43 +00:00
.namespaces = perf_event__process_namespaces,
.exit = perf_event__process_exit,
.fork = perf_event__process_fork,
perf script: Set up output options for in-stream attributes Attributes (struct perf_event_attr) are recorded separately in the perf.data file. perf script uses them to set up output options. However attributes can also be in the event stream, for example when the input is a pipe (i.e. live mode). This patch makes perf script process in-stream attributes in the same way as on-file attributes. Here is an example: Before this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4220 4220 [-01] 2933367.838906: cycles: :4220 4220 [-01] 2933367.838910: cycles: :4220 4220 [-01] 2933367.838912: cycles: :4220 4220 [-01] 2933367.838914: cycles: :4220 4220 [-01] 2933367.838916: cycles: :4220 4220 [-01] 2933367.838918: cycles: uname 4220 [-01] 2933367.838938: cycles: uname 4220 [-01] 2933367.839207: cycles: After this patch: $ perf record uname | perf script Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.015 MB (null) (~655 samples) ] :4582 4582 2933425.707724: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707728: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707730: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707732: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707734: cycles: ffffffff81043ffa native_write_msr_safe ([kernel.kallsyms]) :4582 4582 2933425.707736: cycles: ffffffff81309a24 memcpy ([kernel.kallsyms]) uname 4582 2933425.707760: cycles: ffffffff8109c1c7 enqueue_task_fair ([kernel.kallsyms]) uname 4582 2933425.707978: cycles: ffffffff81308457 clear_page_c ([kernel.kallsyms]) Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1383313899-15987-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2013-11-01 13:51:30 +00:00
.attr = process_attr,
perf script: Process event update events Andreas reported following command produces no output: # cat test.py #!/usr/bin/env python def stat__krava(cpu, thread, time, val, ena, run): print "event %s cpu %d, thread %d, time %d, val %d, ena %d, run %d" % \ ("krava", cpu, thread, time, val, ena, run) # perf stat -a -I 1000 -e cycles,"cpu/config=0x6530160,name=krava/" record | perf script -s test.py ^C # The reason is that 'perf script' does not process event update events and will never get the event name update thus the python callback is never called. The fix is just to add already existing callback we use in 'perf stat report'. Committer note: After the patch: # perf stat -a -I 1000 -e cycles,"cpu/config=0x6530160,name=krava/" record | perf script -s test.py event krava cpu -1, thread -1, time 1000239179, val 1789051, ena 4000690920, run 4000690920 event krava cpu -1, thread -1, time 2000479061, val 2391338, ena 4000879596, run 4000879596 event krava cpu -1, thread -1, time 3000740802, val 1939121, ena 4000977209, run 4000977209 event krava cpu -1, thread -1, time 4001006730, val 2356115, ena 4001000489, run 4001000489 ^C # Reported-by: Andreas Hollmann <hollmann@in.tum.de> Signed-off-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: David Ahern <dsahern@gmail.com> Cc: Milian Wolff <milian.wolff@kdab.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1460013073-18444-3-git-send-email-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-04-07 07:11:13 +00:00
.event_update = perf_event__process_event_update,
.tracing_data = perf_event__process_tracing_data,
.feature = process_feature_event,
.build_id = perf_event__process_build_id,
.id_index = perf_event__process_id_index,
.auxtrace_info = perf_script__process_auxtrace_info,
.auxtrace = perf_event__process_auxtrace,
.auxtrace_error = perf_event__process_auxtrace_error,
.stat = perf_event__process_stat_event,
.stat_round = process_stat_round_event,
.stat_config = process_stat_config_event,
.thread_map = process_thread_map_event,
.cpu_map = process_cpu_map_event,
.ordered_events = true,
.ordering_requires_timestamps = true,
},
};
struct perf_data data = {
perf script: Support using -f to override perf.data file ownership Enable perf script to use perf.data when it is not owned by current user or root. Change the short option name of --fields to -F to avoid confusion with --force. Example: # perf record ls # chown Yunlong.Song:Yunlong.Song perf.data # ls -al perf.data -rw------- 1 Yunlong.Song Yunlong.Song 28360 Apr 2 14:53 perf.data # id uid=0(root) gid=0(root) groups=0(root),64(pkcs11) Before this patch: # perf script File perf.data not owned by current user or root (use -f to override) # perf script -f Error: switch `f' requires a value usage: perf script [<options>] or: perf script [<options>] record <script> [<record-options>] <command> or: perf script [<options>] report <script> [script-args] or: perf script [<options>] <script> [<record-options>] <command> or: perf script [<options>] <top-script> [script-args] -f, --fields <str> comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso,addr,symoff,period As shown above, the -f option does not work at all. And -f is already taken up by --fields, which makes --force confused, so change the short option name of --fields to -F like what other perf commands do (e.g. perf report -F) and use -f as the short option name of --force. After this patch: # perf script File perf.data not owned by current user or root (use -f to override) # perf script -f :41298 41298 2590086.564226: 1 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) :41298 41298 2590086.564244: 1 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) :41298 41298 2590086.564249: 7 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) :41298 41298 2590086.564255: 176 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) ls 41298 2590086.567346: 4059 cycles: ffffffff8105a592 raise_softirq ([kernel.kallsyms]) ls 41298 2590086.567353: 3717 cycles: ffffffff8105a592 raise_softirq ([kernel.kallsyms]) ls 41298 2590086.567358: 63058 cycles: ffffffff8105a592 raise_softirq ([kernel.kallsyms]) ls 41298 2590086.567448: 1706255 cycles: 406ae0 [unknown] (/usr/bin/ls) As shown above, the -f option really works now. Signed-off-by: Yunlong Song <yunlong.song@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1427982439-27388-8-git-send-email-yunlong.song@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-02 13:47:16 +00:00
.mode = PERF_DATA_MODE_READ,
};
const struct option options[] = {
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
"dump raw trace in ASCII"),
perf: Fix endianness argument compatibility with OPT_BOOLEAN() and introduce OPT_INCR() Parsing an option from the command line with OPT_BOOLEAN on a bool data type would not work on a big-endian machine due to the manner in which the boolean was being cast into an int and incremented. For example, running 'perf probe --list' on a PowerPC machine would fail to properly set the list_events bool and would therefore print out the usage information and terminate. This patch makes OPT_BOOLEAN work as expected with a bool datatype. For cases where the original OPT_BOOLEAN was intentionally being used to increment an int each time it was passed in on the command line, this patch introduces OPT_INCR with the old behaviour of OPT_BOOLEAN (the verbose variable is currently the only such example of this). I have reviewed every use of OPT_BOOLEAN to verify that a true C99 bool was passed. Where integers were used, I verified that they were only being used for boolean logic and changed them to bools to ensure that they would not be mistakenly used as ints. The major exception was the verbose variable which now uses OPT_INCR instead of OPT_BOOLEAN. Signed-off-by: Ian Munsie <imunsie@au.ibm.com> Acked-by: David S. Miller <davem@davemloft.net> Cc: <stable@kernel.org> # NOTE: wont apply to .3[34].x cleanly, please backport Cc: Git development list <git@vger.kernel.org> Cc: Ian Munsie <imunsie@au1.ibm.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Eric B Munson <ebmunson@us.ibm.com> Cc: Valdis.Kletnieks@vt.edu Cc: WANG Cong <amwang@redhat.com> Cc: Thiago Farina <tfransosi@gmail.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Jaswinder Singh Rajput <jaswinderrajput@gmail.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Cc: Mike Galbraith <efault@gmx.de> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: Anton Blanchard <anton@samba.org> Cc: John Kacur <jkacur@redhat.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Steven Rostedt <rostedt@goodmis.org> LKML-Reference: <1271147857-11604-1-git-send-email-imunsie@au.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-04-13 08:37:33 +00:00
OPT_INCR('v', "verbose", &verbose,
"be more verbose (show symbol address, etc)"),
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
OPT_BOOLEAN('L', "Latency", &latency_format,
"show latency attributes (irqs/preemption disabled, etc)"),
perf trace/scripting: List available scripts Lists the available perf trace scripts, one per line e.g.: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity To be consistent with the other listing options in perf, the current latency trace option was changed to '-L', and '-l' is now used to access the script listing as: To create the list, it searches each scripts/*/bin directory for files ending with "-report" and reads information found in certain comment lines contained in those shell scripts: - if the comment line starts with "description:", the rest of the line is used as a 'half-line' description. To keep each line in the list to a single line, the description should be limited to 40 characters (the rest of the line contains the script name and args) - if the comment line starts with "args:", the rest of the line names the args the script supports. Required args should be surrounded by <> brackets, optional args by [] brackets. The current scripts in scripts/perl/bin have also been updated with description: and args: comments. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-5-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:38 +00:00
OPT_CALLBACK_NOOPT('l', "list", NULL, NULL, "list available scripts",
list_available_scripts),
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
OPT_CALLBACK('s', "script", NULL, "name",
"script file name (lang:script name, script name, or *)",
parse_scriptname),
OPT_STRING('g', "gen-script", &generate_script_lang, "lang",
"generate perf-script.xx script in specified language"),
OPT_STRING('i', "input", &input_name, "file", "input file name"),
OPT_BOOLEAN('d', "debug-mode", &debug_mode,
"do various checks like samples ordering and lost events"),
OPT_BOOLEAN(0, "header", &header, "Show data header."),
OPT_BOOLEAN(0, "header-only", &header_only, "Show only data header."),
perf script: Add support for dumping symbols Add option to dump symbols found in events. e.g., perf script -f comm,pid,tid,time,trace,sym swapper 0/0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff8100134a cpu_idle ([kernel.kallsyms]) ffffffff81370b39 rest_init ([kernel.kallsyms]) ffffffff81696c23 start_kernel ([kernel.kallsyms].init.text) ffffffff816962af x86_64_start_reservations ([kernel.kallsyms].init.text) ffffffff816963b9 x86_64_start_kernel ([kernel.kallsyms].init.text) sshd 1675/1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff813837aa schedule_hrtimeout_range_clock ([kernel.kallsyms]) ffffffff81383886 schedule_hrtimeout_range ([kernel.kallsyms]) ffffffff8110c4f9 poll_schedule_timeout ([kernel.kallsyms]) ffffffff8110cd20 do_select ([kernel.kallsyms]) ffffffff8110ced8 core_sys_select ([kernel.kallsyms]) ffffffff8110d00d sys_select ([kernel.kallsyms]) ffffffff81002bc2 system_call ([kernel.kallsyms]) 7f1647e56e93 __GI_select (/lib64/libc-2.12.90.so) netstat 1692/1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff81002c3a sysret_careful ([kernel.kallsyms]) 7f7a6cd1b210 __GI___libc_read (/lib64/libc-2.12.90.so) Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-6-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:27 +00:00
OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
"file", "vmlinux pathname"),
OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name,
"file", "kallsyms pathname"),
OPT_BOOLEAN('G', "hide-call-graph", &no_callchain,
"When printing symbols do not display call chain"),
OPT_CALLBACK(0, "symfs", NULL, "directory",
"Look for files with symbols relative to this directory",
symbol__config_symfs),
perf script: Support using -f to override perf.data file ownership Enable perf script to use perf.data when it is not owned by current user or root. Change the short option name of --fields to -F to avoid confusion with --force. Example: # perf record ls # chown Yunlong.Song:Yunlong.Song perf.data # ls -al perf.data -rw------- 1 Yunlong.Song Yunlong.Song 28360 Apr 2 14:53 perf.data # id uid=0(root) gid=0(root) groups=0(root),64(pkcs11) Before this patch: # perf script File perf.data not owned by current user or root (use -f to override) # perf script -f Error: switch `f' requires a value usage: perf script [<options>] or: perf script [<options>] record <script> [<record-options>] <command> or: perf script [<options>] report <script> [script-args] or: perf script [<options>] <script> [<record-options>] <command> or: perf script [<options>] <top-script> [script-args] -f, --fields <str> comma separated output fields prepend with 'type:'. Valid types: hw,sw,trace,raw. Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso,addr,symoff,period As shown above, the -f option does not work at all. And -f is already taken up by --fields, which makes --force confused, so change the short option name of --fields to -F like what other perf commands do (e.g. perf report -F) and use -f as the short option name of --force. After this patch: # perf script File perf.data not owned by current user or root (use -f to override) # perf script -f :41298 41298 2590086.564226: 1 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) :41298 41298 2590086.564244: 1 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) :41298 41298 2590086.564249: 7 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) :41298 41298 2590086.564255: 176 cycles: ffffffff8103efc6 native_write_msr_safe ([kernel.kallsyms]) ls 41298 2590086.567346: 4059 cycles: ffffffff8105a592 raise_softirq ([kernel.kallsyms]) ls 41298 2590086.567353: 3717 cycles: ffffffff8105a592 raise_softirq ([kernel.kallsyms]) ls 41298 2590086.567358: 63058 cycles: ffffffff8105a592 raise_softirq ([kernel.kallsyms]) ls 41298 2590086.567448: 1706255 cycles: 406ae0 [unknown] (/usr/bin/ls) As shown above, the -f option really works now. Signed-off-by: Yunlong Song <yunlong.song@huawei.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1427982439-27388-8-git-send-email-yunlong.song@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-04-02 13:47:16 +00:00
OPT_CALLBACK('F', "fields", NULL, "str",
"comma separated output fields prepend with 'type:'. "
perf script: Allow adding and removing fields With 'perf script' it is common that we just want to add or remove a field. Currently this requires figuring out the long list of default fields and specifying them first, and then adding/removing the new field. This patch adds a new + - syntax to merely add or remove fields, that allows more succint and clearer command lines For example to remove the comm field from PMU samples: Previously $ perf script -F tid,cpu,time,event,sym,ip,dso,period | head -1 swapper 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) with the new syntax perf script -F -comm | head -1 0 [000] 504345.383126: 1 cycles: ffffffff90060c66 native_write_msr ([kernel.kallsyms]) The new syntax cannot be mixed with normal overriding. v2: Fix example in description. Use tid vs pid. No functional changes. v3: Don't skip initialization when user specified explicit type. v4: Rebase. Remove empty line. Committer testing: # perf record -a usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.748 MB perf.data (14 samples) ] Without a explicit field list specified via -F, defaults to: # perf script | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Which is equivalent to: # perf script -F comm,tid,cpu,time,period,event,ip,sym,dso | head -2 perf 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) swapper 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # So if we want to remove the comm, as in your original example, we would have to figure out the default field list and remove ' comm' from it: # perf script -F tid,cpu,time,period,event,ip,sym,dso | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # With your patch this becomes simpler, one can remove fields by prefixing them with '-': # perf script -F -comm | head -2 6338 [000] 18467.058607: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) 0 [001] 18467.058617: 1 cycles: ffffffff89060c36 native_write_msr (/lib/modules/4.11.0-rc8+/build/vmlinux) # Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Milian Wolff <milian.wolff@kdab.com> Link: http://lkml.kernel.org/r/20170602154810.15875-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-06-02 15:48:10 +00:00
"+field to add and -field to remove."
"Valid types: hw,sw,trace,raw,synth. "
"Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso,"
"addr,symoff,srcline,period,iregs,uregs,brstack,"
"brstacksym,flags,bpf-output,brstackinsn,brstackoff,"
"callindent,insn,insnlen,synth,phys_addr,metric,misc",
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
parse_output_fields),
OPT_BOOLEAN('a', "all-cpus", &system_wide,
"system-wide collection from all CPUs"),
OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
"only consider these symbols"),
perf script: Add --insn-trace for instruction decoding Add a --insn-trace short hand option for decoding and disassembling instruction streams for intel_pt. This automatically pipes the output into the xed disassembler to generate disassembled instructions. This just makes this use model much nicer to use. Before % perf record -e intel_pt// ... % perf script --itrace=i0ns --ns -F +insn,-event,-period | xed -F insn: -A -64 swapper 0 [000] 17276.429606186: ffffffff81010486 pt_config ([kernel.kallsyms]) nopl %eax, (%rax,%rax,1) swapper 0 [000] 17276.429606186: ffffffff8101048b pt_config ([kernel.kallsyms]) add $0x10, %rsp swapper 0 [000] 17276.429606186: ffffffff8101048f pt_config ([kernel.kallsyms]) popq %rbx swapper 0 [000] 17276.429606186: ffffffff81010490 pt_config ([kernel.kallsyms]) popq %rbp swapper 0 [000] 17276.429606186: ffffffff81010491 pt_config ([kernel.kallsyms]) popq %r12 swapper 0 [000] 17276.429606186: ffffffff81010493 pt_config ([kernel.kallsyms]) popq %r13 swapper 0 [000] 17276.429606186: ffffffff81010495 pt_config ([kernel.kallsyms]) popq %r14 swapper 0 [000] 17276.429606186: ffffffff81010497 pt_config ([kernel.kallsyms]) popq %r15 swapper 0 [000] 17276.429606186: ffffffff81010499 pt_config ([kernel.kallsyms]) retq swapper 0 [000] 17276.429606186: ffffffff8101063e pt_event_add ([kernel.kallsyms]) cmpl $0x1, 0x1b0(%rbx) swapper 0 [000] 17276.429606186: ffffffff81010645 pt_event_add ([kernel.kallsyms]) mov $0xffffffea, %eax swapper 0 [000] 17276.429606186: ffffffff8101064a pt_event_add ([kernel.kallsyms]) mov $0x0, %edx swapper 0 [000] 17276.429606186: ffffffff8101064f pt_event_add ([kernel.kallsyms]) popq %rbx swapper 0 [000] 17276.429606186: ffffffff81010650 pt_event_add ([kernel.kallsyms]) cmovnz %edx, %eax swapper 0 [000] 17276.429606186: ffffffff81010653 pt_event_add ([kernel.kallsyms]) jmp 0xffffffff81010635 swapper 0 [000] 17276.429606186: ffffffff81010635 pt_event_add ([kernel.kallsyms]) retq swapper 0 [000] 17276.429606186: ffffffff8115e687 event_sched_in.isra.107 ([kernel.kallsyms]) test %eax, %eax Now: % perf record -e intel_pt// ... % perf script --insn-trace --xed ... same output ... XED needs to be installed with: $ git clone https://github.com/intelxed/mbuild.git mbuild $ git clone https://github.com/intelxed/xed $ cd xed $ ./mfile.py $ ./mfile.py examples $ sudo ./mfile.py --prefix=/usr/local install $ sudo cp obj/examples/xed /usr/local/bin $ xed | head -3 ERROR: required argument(s) were missing Copyright (C) 2017, Intel Corporation. All rights reserved. XED version: [v10.0-328-g7d62c8c49b7b] $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20180920180540.14039-2-andi@firstfloor.org [ Fixed up whitespace damage, added the 'mfile.py examples + cp obj/examples/xed ... ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:36 +00:00
OPT_CALLBACK_OPTARG(0, "insn-trace", &itrace_synth_opts, NULL, NULL,
"Decode instructions from itrace", parse_insn_trace),
OPT_CALLBACK_OPTARG(0, "xed", NULL, NULL, NULL,
"Run xed disassembler on output", parse_xed),
tools script: Add --call-trace and --call-ret-trace Add short cut options to print PT call trace and call-ret-trace, for calls and call and returns. Roughly corresponds to ftrace function tracer and function graph tracer. Just makes these common use cases nicer to use. % perf record -a -e intel_pt// sleep 1 % perf script --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_filter_match perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage % perf script --call-ret-trace perf 900 [000] 194167.205652203: tr strt ([unknown]) pt_config perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) pt_config perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) pt_event_add perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) perf_pmu_nop_void perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) perf_pmu_nop_int perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) event_filter_match perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) event_filter_match perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: return ([kernel.kallsyms]) perf_pmu_nop_txn perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: call ([kernel.kallsyms]) perf_event_set_state.part.71 Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-4-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:38 +00:00
OPT_CALLBACK_OPTARG(0, "call-trace", &itrace_synth_opts, NULL, NULL,
"Decode calls from from itrace", parse_call_trace),
OPT_CALLBACK_OPTARG(0, "call-ret-trace", &itrace_synth_opts, NULL, NULL,
"Decode calls and returns from itrace", parse_callret_trace),
perf script: Implement --graph-function Add a ftrace style --graph-function argument to 'perf script' that allows to print itrace function calls only below a given function. This makes it easier to find the code of interest in a large trace. % perf record -e intel_pt//k -a sleep 1 % perf script --graph-function group_sched_in --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) calc_timer_values perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_cpu perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) arch_perf_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __fentry__ perf 900 [000] 194167.205652203: ([kernel.kallsyms]) using_native_sched_clock perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_stable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) group_sched_in swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_set_state.part.71 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_time swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_pmu_disable swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_log_itrace_start swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) calc_timer_values swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_cpu swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) arch_perf_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __fentry__ swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) using_native_sched_clock swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_stable Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:39 +00:00
OPT_STRING(0, "graph-function", &symbol_conf.graph_function, "symbol[,symbol...]",
"Only print symbols and callees with --call-trace/--call-ret-trace"),
perf script: Add option to stop printing callchain Allow user to specify list of symbols which cause the dump of callchains to stop at that symbol. Committer notes: Testing it: # perf record -ag usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 1.177 MB perf.data (33 samples) ] # # # Without it: # # perf script swapper 0 [000] 9693.370039: 1 cycles:ppp: 2072ad x86_pmu_enable (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a29d7 perf_pmu_enable.part.90 (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a713a ctx_resched (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a76c1 __perf_event_enable (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a0390 event_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a1cff remote_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 326978 flush_smp_call_function_queue (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 327413 generic_smp_call_function_single_interrupt (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 249b37 smp_call_function_single_interrupt (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) a04b2c call_function_single_interrupt (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 889427 cpuidle_enter (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 2e534a call_cpuidle (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 2e5730 cpu_startup_entry (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 9f5167 rest_init (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 137ffeb start_kernel ([kernel.vmlinux].init.text) 137f2ca x86_64_start_reservations ([kernel.vmlinux].init.text) 137f419 x86_64_start_kernel ([kernel.vmlinux].init.text) swapper 0 [000] 9693.370044: 1 cycles:ppp: 20ca1b intel_pmu_handle_irq (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 205b0c perf_event_nmi_handler (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 22a14a nmi_handle (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 22a6b3 default_do_nmi (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 22a83c do_nmi (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) a03fb1 end_repeat_nmi (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a29d7 perf_pmu_enable.part.90 (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a713a ctx_resched (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a76c1 __perf_event_enable (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a0390 event_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a1cff remote_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 326978 flush_smp_call_function_queue (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 327413 generic_smp_call_function_single_interrupt (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 249b37 smp_call_function_single_interrupt (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) a04b2c call_function_single_interrupt (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 889427 cpuidle_enter (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 2e534a call_cpuidle (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 2e5730 cpu_startup_entry (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 9f5167 rest_init (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 137ffeb start_kernel ([kernel.vmlinux].init.text) 137f2ca x86_64_start_reservations ([kernel.vmlinux].init.text) # # # Using it to see just what are the calls from the 'remote_function' function: # # perf script --stop-bt remote_function swapper 0 [000] 9693.370039: 1 cycles:ppp: 2072ad x86_pmu_enable (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a29d7 perf_pmu_enable.part.90 (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a713a ctx_resched (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a76c1 __perf_event_enable (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a0390 event_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a1cff remote_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [000] 9693.370044: 1 cycles:ppp: 20ca1b intel_pmu_handle_irq (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 205b0c perf_event_nmi_handler (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 22a14a nmi_handle (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 22a6b3 default_do_nmi (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 22a83c do_nmi (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) a03fb1 end_repeat_nmi (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a29d7 perf_pmu_enable.part.90 (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a713a ctx_resched (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a76c1 __perf_event_enable (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a0390 event_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) 3a1cff remote_function (/usr/lib/debug/lib/modules/4.8.8-300.fc25.x86_64/vmlinux) Signed-off-by: David Ahern <dsa@cumulusnetworks.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480104021-36275-1-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-25 20:00:21 +00:00
OPT_STRING(0, "stop-bt", &symbol_conf.bt_stop_list_str, "symbol[,symbol...]",
"Stop display of callgraph at these symbols"),
OPT_STRING('C', "cpu", &cpu_list, "cpu", "list of cpus to profile"),
OPT_STRING('c', "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
"only display events for these comms"),
OPT_STRING(0, "pid", &symbol_conf.pid_list_str, "pid[,pid...]",
"only consider symbols in these pids"),
OPT_STRING(0, "tid", &symbol_conf.tid_list_str, "tid[,tid...]",
"only consider symbols in these tids"),
OPT_UINTEGER(0, "max-stack", &scripting_max_stack,
"Set the maximum stack depth when parsing the callchain, "
"anything beyond the specified depth will be ignored. "
"Default: kernel.perf_event_max_stack or " __stringify(PERF_MAX_STACK_DEPTH)),
perf tools: Make perf.data more self-descriptive (v8) The goal of this patch is to include more information about the host environment into the perf.data so it is more self-descriptive. Overtime, profiles are captured on various machines and it becomes hard to track what was recorded, on what machine and when. This patch provides a way to solve this by extending the perf.data file with basic information about the host machine. To add those extensions, we leverage the feature bits capabilities of the perf.data format. The change is backward compatible with existing perf.data files. We define the following useful new extensions: - HEADER_HOSTNAME: the hostname - HEADER_OSRELEASE: the kernel release number - HEADER_ARCH: the hw architecture - HEADER_CPUDESC: generic CPU description - HEADER_NRCPUS: number of online/avail cpus - HEADER_CMDLINE: perf command line - HEADER_VERSION: perf version - HEADER_TOPOLOGY: cpu topology - HEADER_EVENT_DESC: full event description (attrs) - HEADER_CPUID: easy-to-parse low level CPU identication The small granularity for the entries is to make it easier to extend without breaking backward compatiblity. Many entries are provided as ASCII strings. Perf report/script have been modified to print the basic information as easy-to-parse ASCII strings. Extended information about CPU and NUMA topology may be requested with the -I option. Thanks to David Ahern for reviewing and testing the many versions of this patch. $ perf report --stdio # ======== # captured on : Mon Sep 26 15:22:14 2011 # hostname : quad # os release : 3.1.0-rc4-tip # perf version : 3.1.0-rc4 # arch : x86_64 # nrcpus online : 4 # nrcpus avail : 4 # cpudesc : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz # cpuid : GenuineIntel,6,15,11 # total memory : 8105360 kB # cmdline : /home/eranian/perfmon/official/tip/build/tools/perf/perf record date # event : name = cycles, type = 0, config = 0x0, config1 = 0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, id = { 29, 30, 31, # HEADER_CPU_TOPOLOGY info available, use -I to display # HEADER_NUMA_TOPOLOGY info available, use -I to display # ======== # ... $ perf report --stdio -I # ======== # captured on : Mon Sep 26 15:22:14 2011 # hostname : quad # os release : 3.1.0-rc4-tip # perf version : 3.1.0-rc4 # arch : x86_64 # nrcpus online : 4 # nrcpus avail : 4 # cpudesc : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz # cpuid : GenuineIntel,6,15,11 # total memory : 8105360 kB # cmdline : /home/eranian/perfmon/official/tip/build/tools/perf/perf record date # event : name = cycles, type = 0, config = 0x0, config1 = 0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, id = { 29, 30, 31, # sibling cores : 0-3 # sibling threads : 0 # sibling threads : 1 # sibling threads : 2 # sibling threads : 3 # node0 meminfo : total = 8320608 kB, free = 7571024 kB # node0 cpu list : 0-3 # ======== # ... Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Robert Richter <robert.richter@amd.com> Cc: Andi Kleen <ak@linux.intel.com> Link: http://lkml.kernel.org/r/20110930134040.GA5575@quad Signed-off-by: Stephane Eranian <eranian@google.com> [ committer notes: Use --show-info in the tools as was in the docs, rename perf_header_fprintf_info to perf_file_section__fprintf_info, fixup conflict with f69b64f7 "perf: Support setting the disassembler style" ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-09-30 13:40:40 +00:00
OPT_BOOLEAN('I', "show-info", &show_full_info,
"display extended information from perf.data file"),
OPT_BOOLEAN('\0', "show-kernel-path", &symbol_conf.show_kernel_path,
"Show the path of [kernel.kallsyms]"),
OPT_BOOLEAN('\0', "show-task-events", &script.show_task_events,
"Show the fork/comm/exit events"),
OPT_BOOLEAN('\0', "show-mmap-events", &script.show_mmap_events,
"Show the mmap events"),
OPT_BOOLEAN('\0', "show-switch-events", &script.show_switch_events,
"Show context switch events (if recorded)"),
perf script: Add script print support for namespace events Introduce a new option to display events of type PERF_RECORD_NAMESPACES and update perf-script documentation accordingly. Shown below is output (trimmed) of perf script command with the newly introduced option, on perf.data generated with perf record command using --namespaces option. $ perf script --show-namespace-events swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 1/1 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 0.000000: PERF_RECORD_NAMESPACES 2/2 - nr_namespaces: 7 [0/net: 3/0xf000001c, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] Commiter notes: Testing it: Investigating that double PERF_RECORD_NAMESPACES for the 19155 pid/tid... Its more than that, there are two PERF_RECORD_COMM as well, and with zeroed timestamps, so probably a synthesizing artifact... # perf script --show-task --show-namespace <SNIP> perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19154/19154 perf 0 [000] 0.000000: PERF_RECORD_FORK(19155:19155):(19154:19154) perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_COMM: perf:19155/19155 perf 0 [000] 0.000000: PERF_RECORD_NAMESPACES 19155/19155 - nr_namespaces: 7 [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc, 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb] swapper 0 [000] 3110.881834: 1 cycles: ffffffffa7060bf6 native_write_msr (/lib/modules/4.11.0-rc1+/build/vmlinux) <SNIP> Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com> Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Eric Biederman <ebiederm@xmission.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sargun Dhillon <sargun@sargun.me> Cc: Steven Rostedt <rostedt@goodmis.org> Link: http://lkml.kernel.org/r/148891932627.25309.1941587059154176221.stgit@hbathini.in.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-03-07 20:42:06 +00:00
OPT_BOOLEAN('\0', "show-namespace-events", &script.show_namespace_events,
"Show namespace events (if recorded)"),
OPT_BOOLEAN('\0', "show-lost-events", &script.show_lost_events,
"Show lost events (if recorded)"),
OPT_BOOLEAN('\0', "show-round-events", &script.show_round_events,
"Show round events (if recorded)"),
OPT_BOOLEAN('\0', "per-event-dump", &script.per_event_dump,
"Dump trace output to files named by the monitored events"),
OPT_BOOLEAN('f', "force", &symbol_conf.force, "don't complain, do it"),
perf script: Add 'brstackinsn' for branch stacks Implement printing instruction sequences as hex dump for branch stacks. This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually. This is good enough for pattern matching. This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems). % perf record -b ... % perf script -F brstackinsn ... read_hpet+67: ffffffff9905b843 insn: 74 ea # PRED ffffffff9905b82f insn: 85 c9 ffffffff9905b831 insn: 74 12 ffffffff9905b833 insn: f3 90 ffffffff9905b835 insn: 48 8b 0f ffffffff9905b838 insn: 48 89 ca ffffffff9905b83b insn: 48 c1 ea 20 ffffffff9905b83f insn: 39 f2 ffffffff9905b841 insn: 89 d0 ffffffff9905b843 insn: 74 ea # PRED Only works when no special branch filters are specified. Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message. The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder. An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder. Committer note: Added hint about how to get suitable perf.data files for use with '-F brstackinsm': $ perf record usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ] $ $ perf script -F brstackinsn Display of branch stack assembler requested, but non all-branch filter set Hint: run 'perf record -b ...' $ Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-02-23 23:46:34 +00:00
OPT_INTEGER(0, "max-blocks", &max_blocks,
"Maximum number of code blocks to dump with brstackinsn"),
OPT_BOOLEAN(0, "ns", &nanosecs,
"Use 9 decimal places when displaying time"),
OPT_CALLBACK_OPTARG(0, "itrace", &itrace_synth_opts, NULL, "opts",
"Instruction Tracing options\n" ITRACE_HELP,
itrace_parse_synth_opts),
OPT_BOOLEAN(0, "full-source-path", &srcline_full_filename,
"Show full source file name path for source lines"),
OPT_BOOLEAN(0, "demangle", &symbol_conf.demangle,
"Enable symbol demangling"),
OPT_BOOLEAN(0, "demangle-kernel", &symbol_conf.demangle_kernel,
"Enable kernel symbol demangling"),
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
OPT_STRING(0, "time", &script.time_str, "str",
"Time span of interest (start,stop)"),
perf script: Add --inline option for debugging The --inline option is to show inlined functions in callchains. For example: $ perf script a.out 5644 11611.467597: 309961 cycles:u: 790 main (/home/namhyung/tmp/perf/a.out) 20511 __libc_start_main (/usr/lib/libc-2.25.so) 8ba _start (/home/namhyung/tmp/perf/a.out) ... $ perf script --inline a.out 5644 11611.467597: 309961 cycles:u: 790 main (/home/namhyung/tmp/perf/a.out) std::__detail::_Adaptor<std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>, double>::operator() std::uniform_real_distribution<double>::operator()<std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul> > std::uniform_real_distribution<double>::operator()<std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul> > main 20511 __libc_start_main (/usr/lib/libc-2.25.so) 8ba _start (/home/namhyung/tmp/perf/a.out) ... Reviewed-and-tested-by: Milian Wolff <milian.wolff@kdab.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jin Yao <yao.jin@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Milian Wolff <milian.wolff@kdab.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: kernel-team@lge.com Link: http://lkml.kernel.org/r/20170524062129.32529-5-namhyung@kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-05-24 06:21:26 +00:00
OPT_BOOLEAN(0, "inline", &symbol_conf.inline_name,
"Show inline function"),
OPT_END()
};
const char * const script_subcommands[] = { "record", "report", NULL };
const char *script_usage[] = {
"perf script [<options>]",
"perf script [<options>] record <script> [<record-options>] <command>",
"perf script [<options>] report <script> [script-args]",
"perf script [<options>] <script> [<record-options>] <command>",
"perf script [<options>] <top-script> [script-args]",
NULL
};
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
perf_set_singlethreaded();
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
setup_scripting();
argc = parse_options_subcommand(argc, argv, options, script_subcommands, script_usage,
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
PARSE_OPT_STOP_AT_NON_OPTION);
data.file.path = input_name;
data.force = symbol_conf.force;
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) {
rec_script_path = get_script_path(argv[1], RECORD_SUFFIX);
if (!rec_script_path)
return cmd_record(argc, argv);
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (argc > 1 && !strncmp(argv[0], "rep", strlen("rep"))) {
rep_script_path = get_script_path(argv[1], REPORT_SUFFIX);
if (!rep_script_path) {
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
fprintf(stderr,
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
"Please specify a valid report script"
"(see 'perf script -l' for listing)\n");
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
return -1;
}
}
if (itrace_synth_opts.callchain &&
itrace_synth_opts.callchain_sz > scripting_max_stack)
scripting_max_stack = itrace_synth_opts.callchain_sz;
/* make sure PERF_EXEC_PATH is set for scripts */
set_argv_exec_path(get_argv_exec_path());
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (argc && !script_name && !rec_script_path && !rep_script_path) {
int live_pipe[2];
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
int rep_args;
pid_t pid;
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
rec_script_path = get_script_path(argv[0], RECORD_SUFFIX);
rep_script_path = get_script_path(argv[0], REPORT_SUFFIX);
if (!rec_script_path && !rep_script_path) {
usage_with_options_msg(script_usage, options,
"Couldn't find script `%s'\n\n See perf"
" script -l for available scripts.\n", argv[0]);
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (is_top_script(argv[0])) {
rep_args = argc - 1;
} else {
int rec_args;
rep_args = has_required_arg(rep_script_path);
rec_args = (argc - 1) - rep_args;
if (rec_args < 0) {
usage_with_options_msg(script_usage, options,
"`%s' script requires options."
"\n\n See perf script -l for available "
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
"scripts and options.\n", argv[0]);
}
}
if (pipe(live_pipe) < 0) {
perror("failed to create pipe");
return -1;
}
pid = fork();
if (pid < 0) {
perror("failed to fork");
return -1;
}
if (!pid) {
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
j = 0;
dup2(live_pipe[1], 1);
close(live_pipe[0]);
if (is_top_script(argv[0])) {
system_wide = true;
} else if (!system_wide) {
if (have_cmd(argc - rep_args, &argv[rep_args]) != 0) {
err = -1;
goto out;
}
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
__argv = malloc((argc + 6) * sizeof(const char *));
if (!__argv) {
pr_err("malloc failed\n");
err = -ENOMEM;
goto out;
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
__argv[j++] = "/bin/sh";
__argv[j++] = rec_script_path;
if (system_wide)
__argv[j++] = "-a";
__argv[j++] = "-q";
__argv[j++] = "-o";
__argv[j++] = "-";
for (i = rep_args + 1; i < argc; i++)
__argv[j++] = argv[i];
__argv[j++] = NULL;
execvp("/bin/sh", (char **)__argv);
free(__argv);
exit(-1);
}
dup2(live_pipe[0], 0);
close(live_pipe[1]);
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
__argv = malloc((argc + 4) * sizeof(const char *));
if (!__argv) {
pr_err("malloc failed\n");
err = -ENOMEM;
goto out;
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
j = 0;
__argv[j++] = "/bin/sh";
__argv[j++] = rep_script_path;
for (i = 1; i < rep_args + 1; i++)
__argv[j++] = argv[i];
__argv[j++] = "-i";
__argv[j++] = "-";
__argv[j++] = NULL;
execvp("/bin/sh", (char **)__argv);
free(__argv);
exit(-1);
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (rec_script_path)
script_path = rec_script_path;
if (rep_script_path)
script_path = rep_script_path;
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
if (script_path) {
j = 0;
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
if (!rec_script_path)
system_wide = false;
else if (!system_wide) {
if (have_cmd(argc - 1, &argv[1]) != 0) {
err = -1;
goto out;
}
}
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
__argv = malloc((argc + 2) * sizeof(const char *));
if (!__argv) {
pr_err("malloc failed\n");
err = -ENOMEM;
goto out;
}
__argv[j++] = "/bin/sh";
__argv[j++] = script_path;
if (system_wide)
__argv[j++] = "-a";
perf trace: live-mode command-line cleanup This patch attempts to make the perf trace command-line for live-mode commands more user-friendly and consistent with other perf commands. The main change it makes is to allow <commands> to be run as part of perf trace live-mode commands, as other perf commands do, instead of the system-wide traces they're currently hard-coded to by the shell scripts. With this patch, the following live-mode trace now works as expected: $ perf trace rw-by-pid ls -al The previous system-wide behavior for this command would still be available by explicitly specifying -a: $ perf trace rw-by-pid -a ls -al and if no <command> is specified, the output is also system-wide: $ perf trace rw-by-pid Because live-mode requires both record and report steps to be invoked, it isn't always possible to know which args to send to the report and which to send to the record steps - mainly this is the case for report scripts with optional args - in those cases it would be necessary to use separate 'perf trace record' and 'perf trace report' steps. For example: $ perf trace syscall-counts ls Here we can't decide whether ls should be passed as a param to the syscall-counts script or whether we should invoke ls as a <command>. In these cases, we just say that we'll ignore optional script params and always interpret the extra arguments as a <command>. If the user instead wants the other interpretation, that can be accomplished by using separate record and report commands explicitly: $ perf trace record syscall-counts $ perf trace report syscall-counts ls So the rules that this patch implements, which seem to make the most intuitive sense for live-mode commands: - for commands with optional args and commands with no args, no args are sent to the report script, all are sent to the record step - for 'top' commands i.e. that end with 'top', <commands> can't be used - all extra args are send to the report script as params - for commands with required args, the n required args are taken to be the first n args after the script name and sent to the report script, and the rest are sent to the record step Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2010-11-10 14:16:51 +00:00
for (i = 2; i < argc; i++)
__argv[j++] = argv[i];
__argv[j++] = NULL;
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
execvp("/bin/sh", (char **)__argv);
free(__argv);
perf trace/scripting: Add 'record' and 'report' options Allow scripts to be recorded/executed by simply specifying the script root name (the script name minus extension) along with 'record' or 'report' to 'perf trace'. The script names shown by 'perf trace -l' can be directly used to run the command-line contained within the corresponding '-record' and '-report' versions of scripts in the scripts/*/bin directories. For example, to record the trace data needed to run the wakeup-latency.pl script, the user can easily find the name of the corresponding script from the script list and invoke it using 'perf trace record', without having to remember the details of how to do the same thing using the lower-level perf trace command-line options: root@tropicana:~# perf trace -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency rw-by-file <comm> r/w activity for a program, by file check-perf-trace useless but exhaustive test script rw-by-pid system-wide r/w activity root@tropicana:~# perf trace record wakeup-latency ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.296 MB perf.data (~12931 samples) ] To run the wakeup-latency.pl script using the captured data, change 'record' to 'report' in the command-line: root@tropicana:~# perf trace report wakeup-latency wakeup_latency stats: total_wakeups: 65 avg_wakeup_latency (ns): 22417 min_wakeup_latency (ns): 3470 max_wakeup_latency (ns): 223311 perf trace Perl script stopped If the script takes options, thay can be simply added to the end of the 'report' invocation: root@tropicana:~# perf trace record rw-by-file ^C[ perf record: Woken up 2 times to write data ] [ perf record: Captured and wrote 0.782 MB perf.data (~34171 samples) ] root@tropicana:~# perf trace report rw-by-file perf file read counts for perf: fd # reads bytes_requested ------ ---------- ----------- 122 1934 1980416 120 1 32 file write counts for perf: fd # writes bytes_written ------ ---------- ----------- 3 4006 280568 perf trace Perl script stopped Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org LKML-Reference: <1260867220-15699-6-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-12-15 08:53:39 +00:00
exit(-1);
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
if (!script_name) {
setup_pager();
use_browser = 0;
}
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
session = perf_session__new(&data, false, &script.tool);
if (session == NULL)
return -1;
if (header || header_only) {
script.tool.show_feat_hdr = SHOW_FEAT_HEADER;
perf_session__fprintf_info(session, stdout, show_full_info);
if (header_only)
goto out_delete;
}
if (show_full_info)
script.tool.show_feat_hdr = SHOW_FEAT_HEADER_FULL_INFO;
perf tools: Check recorded kernel version when finding vmlinux Currently vmlinux_path__init() only tries to find vmlinux file from current directory, /boot and some canonical directories with version number of the running kernel. This can be a problem when reporting old data recorded on a kernel version not running currently. We can use --symfs option for this but it's annoying for user to do it always. As we already have the info in the perf.data file, it can be changed to use it for the search automatically. Before: $ perf report ... # Samples: 4K of event 'cpu-clock' # Event count (approx.): 1067250000 # # Overhead Command Shared Object Symbol # ........ .......... ................. .............................. 71.87% swapper [kernel.kallsyms] [k] recover_probed_instruction After: # Overhead Command Shared Object Symbol # ........ .......... ................. .................... 71.87% swapper [kernel.kallsyms] [k] native_safe_halt This requires to change signature of symbol__init() to receive struct perf_session_env *. Reported-by: Minchan Kim <minchan@kernel.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Namhyung Kim <namhyung.kim@lge.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1407825645-24586-14-git-send-email-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2014-08-12 06:40:45 +00:00
if (symbol__init(&session->header.env) < 0)
goto out_delete;
script.session = session;
script__setup_sample_type(&script);
perf script: Implement --graph-function Add a ftrace style --graph-function argument to 'perf script' that allows to print itrace function calls only below a given function. This makes it easier to find the code of interest in a large trace. % perf record -e intel_pt//k -a sleep 1 % perf script --graph-function group_sched_in --call-trace perf 900 [000] 194167.205652203: ([kernel.kallsyms]) group_sched_in perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) event_sched_in.isra.107 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_set_state.part.71 perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_time perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_disable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_log_itrace_start perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_event_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) calc_timer_values perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_cpu perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax perf 900 [000] 194167.205652203: ([kernel.kallsyms]) arch_perf_update_userpage perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __fentry__ perf 900 [000] 194167.205652203: ([kernel.kallsyms]) using_native_sched_clock perf 900 [000] 194167.205652203: ([kernel.kallsyms]) sched_clock_stable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) perf_pmu_enable perf 900 [000] 194167.205652203: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) group_sched_in swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) event_sched_in.isra.107 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_set_state.part.71 swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_time swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_pmu_disable swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_log_itrace_start swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) perf_event_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) calc_timer_values swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_cpu swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __x86_indirect_thunk_rax swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) arch_perf_update_userpage swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) __fentry__ swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) using_native_sched_clock swapper 0 [001] 194167.205660693: ([kernel.kallsyms]) sched_clock_stable Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Leo Yan <leo.yan@linaro.org> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Link: http://lkml.kernel.org/r/20180920180540.14039-5-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-09-20 18:05:39 +00:00
if ((output[PERF_TYPE_HARDWARE].fields & PERF_OUTPUT_CALLINDENT) ||
symbol_conf.graph_function)
perf script: Add callindent option Based on patches from Andi Kleen. When printing PT instruction traces with perf script it is rather useful to see some indentation for the call tree. This patch adds a new callindent field to perf script that prints spaces for the function call stack depth. We already have code to track the function call stack for PT, that we can reuse with minor modifications. The resulting output is not quite as nice as ftrace yet, but a lot better than what was there before. Note there are some corner cases when the thread stack gets code confused and prints incorrect indentation. Even with that it is fairly useful. When displaying kernel code traces it is recommended to run as root, as otherwise perf doesn't understand the kernel addresses properly, and may not reset the call stack correctly on kernel boundaries. Example output: sudo perf-with-kcore record eg2 -a -e intel_pt// -- sleep 1 sudo perf-with-kcore script eg2 --ns -F callindent,time,comm,pid,sym,ip,addr,flags,cpu --itrace=cre | less ... swapper 0 [000] 5830.389116586: call irq_exit ffffffff8104d620 smp_call_function_single_interrupt+0x30 => ffffffff8107e720 irq_exit swapper 0 [000] 5830.389116586: call idle_cpu ffffffff8107e769 irq_exit+0x49 => ffffffff810a3970 idle_cpu swapper 0 [000] 5830.389116586: return idle_cpu ffffffff810a39b7 idle_cpu+0x47 => ffffffff8107e76e irq_exit swapper 0 [000] 5830.389116586: call tick_nohz_irq_exit ffffffff8107e7bd irq_exit+0x9d => ffffffff810f2fc0 tick_nohz_irq_exit swapper 0 [000] 5830.389116919: call __tick_nohz_idle_enter ffffffff810f2fe0 tick_nohz_irq_exit+0x20 => ffffffff810f28d0 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call ktime_get ffffffff810f28f1 __tick_nohz_idle_enter+0x21 => ffffffff810e9ec0 ktime_get swapper 0 [000] 5830.389116919: call read_tsc ffffffff810e9ef6 ktime_get+0x36 => ffffffff81035070 read_tsc swapper 0 [000] 5830.389116919: return read_tsc ffffffff81035084 read_tsc+0x14 => ffffffff810e9efc ktime_get swapper 0 [000] 5830.389116919: return ktime_get ffffffff810e9f46 ktime_get+0x86 => ffffffff810f28f6 __tick_nohz_idle_enter swapper 0 [000] 5830.389116919: call sched_clock_idle_sleep_event ffffffff810f290b __tick_nohz_idle_enter+0x3b => ffffffff810a7380 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: call sched_clock_cpu ffffffff810a738b sched_clock_idle_sleep_event+0xb => ffffffff810a72e0 sched_clock_cpu swapper 0 [000] 5830.389116919: call sched_clock ffffffff810a734d sched_clock_cpu+0x6d => ffffffff81035750 sched_clock swapper 0 [000] 5830.389116919: call native_sched_clock ffffffff81035754 sched_clock+0x4 => ffffffff81035640 native_sched_clock swapper 0 [000] 5830.389116919: return native_sched_clock ffffffff8103568c native_sched_clock+0x4c => ffffffff81035759 sched_clock swapper 0 [000] 5830.389116919: return sched_clock ffffffff8103575c sched_clock+0xc => ffffffff810a7352 sched_clock_cpu swapper 0 [000] 5830.389116919: return sched_clock_cpu ffffffff810a7356 sched_clock_cpu+0x76 => ffffffff810a7390 sched_clock_idle_sleep_event swapper 0 [000] 5830.389116919: return sched_clock_idle_sleep_event ffffffff810a7391 sched_clock_idle_sleep_event+0x11 => ffffffff810f2910 __tick_nohz_idle_enter ... Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Acked-by: Andi Kleen <ak@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1466689258-28493-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-06-23 13:40:58 +00:00
itrace_synth_opts.thread_stack = true;
session->itrace_synth_opts = &itrace_synth_opts;
if (cpu_list) {
err = perf_session__cpu_bitmap(session, cpu_list, cpu_bitmap);
if (err < 0)
goto out_delete;
itrace_synth_opts.cpu_bitmap = cpu_bitmap;
}
perf script: Add support for H/W and S/W events Custom fields set for each type by prepending field argument with type. For file with multiple event types (e.g., trace and S/W) display of an event type suppressed by setting output fields to "". e.g., perf record -ga -e sched:sched_switch -e cpu-clock -c 10000000 -R -- sleep 1 perf script openssl 11496 [000] 9711.807107: cpu-clock-msecs: ffffffff810c22dc arch_local_irq_restore ([kernel.kallsyms]) ffffffff810c518c __alloc_pages_nodemask ([kernel.kallsyms]) ffffffff810297b2 pte_alloc_one ([kernel.kallsyms]) ffffffff810d8b98 __pte_alloc ([kernel.kallsyms]) ffffffff810daf07 handle_mm_fault ([kernel.kallsyms]) ffffffff8138763a do_page_fault ([kernel.kallsyms]) ffffffff81384a65 page_fault ([kernel.kallsyms]) 7f6130507d70 asn1_check_tlen (/lib64/libcrypto.so.1.0.0c) 0 () openssl 11496 [000] 9711.808042: sched_switch: prev_comm=openssl ... kworker/0:0 4 [000] 9711.808067: sched_switch: prev_comm=kworker/... swapper 0 [001] 9711.808090: sched_switch: prev_comm=kworker/... sshd 11451 [001] 9711.808185: sched_switch: prev_comm=sshd pre... swapper 0 [001] 9711.816155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) openssl 11496 [000] 9711.817104: cpu-clock-msecs: 7f61304ad723 AES_cbc_encrypt (/lib64/libcrypto.so.1.0.0c) 7fff3402f950 () 12f0debc9a785634 () swapper 0 [001] 9711.826155: cpu-clock-msecs: ffffffff81023609 native_safe_halt ([kernel.kallsyms]) ffffffff8100132a cpu_idle ([kernel.kallsyms]) ffffffff8137cf9b start_secondary ([kernel.kallsyms]) To suppress trace events within the file and use default output for S/W events: perf script -f trace: or to suppress S/W events and do default display for trace events: perf script -f sw: Custom field selections: perf script -f sw:comm,tid,time -f trace:time,trace openssl 11496 9711.797162: swapper 0 9711.807071: openssl 11496 9711.807107: 9711.808042: prev_comm=openssl prev_pid=11496 prev_prio=120 prev_state=R ... 9711.808067: prev_comm=kworker/0:0 prev_pid=4 prev_prio=120 prev_state=S ... 9711.808090: prev_comm=kworker/0:0 prev_pid=0 prev_prio=120 prev_state=R ... 9711.808185: prev_comm=sshd prev_pid=11451 prev_prio=120 prev_state=S ==>... swapper 0 9711.816155: openssl 11496 9711.817104: swapper 0 9711.826155: Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-7-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:28 +00:00
if (!no_callchain)
perf script: Add support for dumping symbols Add option to dump symbols found in events. e.g., perf script -f comm,pid,tid,time,trace,sym swapper 0/0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff8100134a cpu_idle ([kernel.kallsyms]) ffffffff81370b39 rest_init ([kernel.kallsyms]) ffffffff81696c23 start_kernel ([kernel.kallsyms].init.text) ffffffff816962af x86_64_start_reservations ([kernel.kallsyms].init.text) ffffffff816963b9 x86_64_start_kernel ([kernel.kallsyms].init.text) sshd 1675/1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff813837aa schedule_hrtimeout_range_clock ([kernel.kallsyms]) ffffffff81383886 schedule_hrtimeout_range ([kernel.kallsyms]) ffffffff8110c4f9 poll_schedule_timeout ([kernel.kallsyms]) ffffffff8110cd20 do_select ([kernel.kallsyms]) ffffffff8110ced8 core_sys_select ([kernel.kallsyms]) ffffffff8110d00d sys_select ([kernel.kallsyms]) ffffffff81002bc2 system_call ([kernel.kallsyms]) 7f1647e56e93 __GI_select (/lib64/libc-2.12.90.so) netstat 1692/1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=... ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms]) ffffffff81382ac5 schedule ([kernel.kallsyms]) ffffffff81002c3a sysret_careful ([kernel.kallsyms]) 7f7a6cd1b210 __GI___libc_read (/lib64/libc-2.12.90.so) Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-6-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:27 +00:00
symbol_conf.use_callchain = true;
else
symbol_conf.use_callchain = false;
if (session->tevent.pevent &&
tep_set_function_resolver(session->tevent.pevent,
machine__resolve_kernel_addr,
&session->machines.host) < 0) {
pr_err("%s: failed to set libtraceevent function resolver\n", __func__);
err = -1;
goto out_delete;
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
if (generate_script_lang) {
struct stat perf_stat;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
int input;
perf script: If type not given fields apply to all event types Allow: perf script -f <fields> to be equivalent to: perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> i.e., the specified fields apply to all event types if the type string is not given. The field (-f) arguments are processed in the order received. A later usage can reset a prior request. e.g., -f trace: -f comm,tid,time,sym The first -f suppresses trace events (field list is ""), but then the second invocation sets the fields to comm,tid,time,sym. In this case a warning is given to the user: "Overriding previous field request for all events." Alternativey, consider the order: -f comm,tid,time,sym -f trace: The first -f sets the fields for all events and the second -f suppresses trace events. The user is given a warning message about the override, and the result of the above is that only S/W and H/W events are displayed with the given fields. For the 'wildcard' option if a user selected field is invalid for an event type, a message is displayed to the user that the option is ignored for that type. For example: perf script -f comm,tid,trace 2>&1 | less 'trace' not valid for hardware events. Ignoring. 'trace' not valid for software events. Ignoring. Alternatively, if the type is given an invalid field is specified it is an error. For example: perf script -v -f sw:comm,tid,trace 2>&1 | less 'trace' not valid for software events. At this point usage is displayed, and perf-script exits. Finally, a user may not set fields to none for all event types. i.e., -f "" is not allowed. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: linux-kernel@vger.kernel.org LPU-Reference: <1300377801-27246-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-17 16:03:21 +00:00
if (output_set_by_user()) {
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
fprintf(stderr,
"custom fields not supported for generated scripts");
err = -EINVAL;
goto out_delete;
perf script: Support custom field selection for output Allow a user to select which fields to print to stdout for event data. Options include comm (command name), tid (thread id), pid (process id), time (perf timestamp), cpu, event (for event name), and trace (for trace data). Default is set to maintain compatibility with current output; this feature does alter output format slightly -- no '-' between command and pid/tid. Thanks to Frederic Weisbecker for detailed suggestions on this approach. Examples (output compressed) 1. trace, default format perf record -ga -e sched:sched_switch perf script swapper 0 [000] 537.037184: sched_switch: prev_comm=swapper prev_pid=0... sshd 1675 [000] 537.037309: sched_switch: prev_comm=sshd prev_pid=1675... netstat 1692 [001] 537.038664: sched_switch: prev_comm=netstat prev_pid=1692... 2. trace, custom format perf record -ga -e sched:sched_switch perf script -f comm,pid,time,trace <--- omitting cpu and event name swapper 0 537.037184: prev_comm=swapper prev_pid=0 prev_prio=120 ... sshd 1675 537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120 ... netstat 1692 537.038664: prev_comm=netstat prev_pid=1692 prev_prio=120 ... Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <1299734608-5223-5-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern <daahern@cisco.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2011-03-10 05:23:26 +00:00
}
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
input = open(data.file.path, O_RDONLY); /* input_name */
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
if (input < 0) {
err = -errno;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
perror("failed to open file");
goto out_delete;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
err = fstat(input, &perf_stat);
if (err < 0) {
perror("failed to stat file");
goto out_delete;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
if (!perf_stat.st_size) {
fprintf(stderr, "zero-sized file, nothing to do!\n");
goto out_delete;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
scripting_ops = script_spec__lookup(generate_script_lang);
if (!scripting_ops) {
fprintf(stderr, "invalid language specifier");
err = -ENOENT;
goto out_delete;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
err = scripting_ops->generate_script(session->tevent.pevent,
"perf-script");
goto out_delete;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
if (script_name) {
err = scripting_ops->start_script(script_name, argc, argv);
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
if (err)
goto out_delete;
pr_debug("perf script started with script %s\n\n", script_name);
script_started = true;
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
}
err = perf_session__check_output_opt(session);
if (err < 0)
goto out_delete;
script.ptime_range = perf_time__range_alloc(script.time_str,
&script.range_size);
if (!script.ptime_range) {
err = -ENOMEM;
goto out_delete;
}
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
/* needs to be parsed after looking up reference time */
if (perf_time__parse_str(script.ptime_range, script.time_str) != 0) {
if (session->evlist->first_sample_time == 0 &&
session->evlist->last_sample_time == 0) {
pr_err("HINT: no first/last sample time found in perf data.\n"
"Please use latest perf binary to execute 'perf record'\n"
"(if '--buildid-all' is enabled, please set '--timestamp-boundary').\n");
err = -EINVAL;
goto out_delete;
}
script.range_num = perf_time__percent_parse_str(
script.ptime_range, script.range_size,
script.time_str,
session->evlist->first_sample_time,
session->evlist->last_sample_time);
if (script.range_num < 0) {
pr_err("Invalid time string\n");
err = -EINVAL;
goto out_delete;
}
} else {
script.range_num = 1;
perf script: Add option to specify time window of interest Add option to allow user to control analysis window. e.g., collect data for some amount of time and analyze a segment of interest within that window. Committer notes: Testing it: # perf evlist -v cycles:ppp: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 # # perf script --hide-call-graph | head -15 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370048: 126 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370049: 2701 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370051: 58823 cycles:ppp: ffffffffb90cd2e0 idle_cpu (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370059: 1 cycles:ppp: ffffffffb91a713a ctx_resched (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370062: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370076: 1 cycles:ppp: ffffffffb91a76c1 __perf_event_enable (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370091: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) perf 5124 [2] 9693.370095: 3 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # # perf script --hide-call-graph --time ,9693.370048 swapper 0 [0] 9693.370039: 1 cycles:ppp: ffffffffb90072ad x86_pmu_enable (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370044: 1 cycles:ppp: ffffffffb900ca1b intel_pmu_handle_irq (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [0] 9693.370046: 7 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) # perf script --hide-call-graph --time 9693.370064,9693.370076 swapper 0 [1] 9693.370064: 13 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370065: 250 cycles:ppp: ffffffffb902fd93 native_sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370067: 5269 cycles:ppp: ffffffffb902fe79 sched_clock (.../4.8.8-300.fc25.x86_64/vmlinux) swapper 0 [1] 9693.370069: 114602 cycles:ppp: ffffffffb90c1c5a atomic_notifier_call_chain (.../4.8.8-300.fc25.x86_64/vmlinux) # Signed-off-by: David Ahern <dsahern@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1480439746-42695-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-29 17:15:43 +00:00
}
err = __cmd_script(&script);
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
flush_scripting();
out_delete:
zfree(&script.ptime_range);
perf_evlist__free_stats(session->evlist);
perf_session__delete(session);
if (script_started)
cleanup_scripting();
perf trace: Add scripting ops Adds an interface, scripting_ops, that when implemented for a particular scripting language enables built-in support for trace stream processing using that language. The interface is designed to enable full-fledged language interpreters to be embedded inside the perf executable and thereby make the full capabilities of the supported languages available for trace processing. See below for details on the interface. This patch also adds a couple command-line options to 'perf trace': The -s option option is used to specify the script to be run. Script names that can be used with -s take the form: [language spec:]scriptname[.ext] Scripting languages register a set of 'language specs' that can be used to specify scripts for the registered languages. The specs can be used either as prefixes or extensions. If [language spec:] is used, the script is taken as a script of the matching language regardless of any extension it might have. If [language spec:] is not used, [.ext] is used to look up the language it corresponds to. Language specs are case insensitive. e.g. Perl scripts can be specified in the following ways: Perl:scriptname pl:scriptname.py # extension ignored PL:scriptname scriptname.pl scriptname.perl The -g [language spec] option gives users an easy starting point for writing scripts in the specified language. Scripting support for a particular language can implement a generate_script() scripting op that outputs an empty (or near-empty) set of handlers for all the events contained in a given perf.data trace file - this option gives users a direct way to access that. Adding support for a scripting language --------------------------------------- The main thing that needs to be done do add support for a new language is to implement the scripting_ops interface: It consists of the following four functions: start_script() stop_script() process_event() generate_script() start_script() is called before any events are processed, and is meant to give the scripting language support an opportunity to set things up to receive events e.g. create and initialize an instance of a language interpreter. stop_script() is called after all events are processed, and is meant to give the scripting language support an opportunity to clean up e.g. destroy the interpreter instance, etc. process_event() is called once for each event and takes as its main parameter a pointer to the binary trace event record to be processed. The implementation is responsible for picking out the binary fields from the event record and sending them to the script handler function associated with that event e.g. a function derived from the event name it's meant to handle e.g. 'sched::sched_switch()'. The 'format' information for trace events can be used to parse the binary data and map it into a form usable by a given scripting language; see the Perl implemention in subsequent patches for one possible way to leverage the existing trace format parsing code in perf and map that info into specific scripting language types. generate_script() should generate a ready-to-run script for the current set of events in the trace, preferably with bodies that print out every field for each event. Again, look at the Perl implementation for clues as to how that can be done. This is an optional, but very useful op. Support for a given language should also add a language-specific setup function and call it from setup_scripting(). The language-specific setup function associates the the scripting ops for that language with one or more 'language specifiers' (see below) using script_spec_register(). When a script name is specified on the command line, the scripting ops associated with the specified language are used to instantiate and use the appropriate interpreter to process the trace stream. In general, it should be relatively easy to add support for a new language, especially if the language implementation supports an interface allowing an interpreter to be 'embedded' inside another program (in this case the containing program will be 'perf trace'). If so, it should be relatively straightforward to translate trace events into invocations of user-defined script functions where e.g. the function name corresponds to the event type and the function parameters correspond to the event fields. The event and field type information exported by the event tracing infrastructure (via the event 'format' files) should be enough to parse and send any piece of trace data to the user script. The easiest way to see how this can be done would be to look at the Perl implementation contained in perf/util/trace-event-perl.c/.h. There are a couple of other things that aren't covered by the scripting_ops or setup interface and are technically optional, but should be implemented if possible. One of these is support for 'flag' and 'symbolic' fields e.g. being able to use more human-readable values such as 'GFP_KERNEL' or HI/BLOCK_IOPOLL/TASKLET in place of raw flag values. See the Perl implementation to see how this can be done. The other thing is support for 'calling back' into the perf executable to access e.g. uncommon fields not passed by default into handler functions, or any metadata the implementation might want to make available to users via the language interface. Again, see the Perl implementation for examples. Signed-off-by: Tom Zanussi <tzanussi@gmail.com> Cc: fweisbec@gmail.com Cc: rostedt@goodmis.org Cc: anton@samba.org Cc: hch@infradead.org LKML-Reference: <1259133352-23685-2-git-send-email-tzanussi@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-25 07:15:46 +00:00
out:
return err;
perf tools: Add perf trace This adds perf trace into the set of perf tools. It is written to fetch the tracepoint samples from perf events and display them, according to the events information given by the debugfs files through the util/trace* tools. It is a rough first shot and doesn't yet handle the cpu, timestamps fields and some other things. Example: perf record -f -e workqueue:workqueue_execution:record -F 1 -a perf trace kblockd/0-236 [000] 0.000000: workqueue_execution: thread=:236 func=cfq_kick_queue+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/0-360 [000] 0.000000: workqueue_execution: thread=:360 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 kondemand/1-361 [000] 0.000000: workqueue_execution: thread=:361 func=do_dbs_timer+0x0 Todo: - A lot of things! Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Mike Galbraith <efault@gmx.de> Cc: "Luis Claudio R. Goncalves" <lclaudio@uudg.org> Cc: Clark Williams <williams@redhat.com> Cc: Jon Masters <jonathan@jonmasters.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: Christoph Hellwig <hch@infradead.org> Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Zhaolei <zhaolei@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: "Frank Ch. Eigler" <fche@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Jiaying Zhang <jiayingz@google.com> Cc: Anton Blanchard <anton@samba.org> LKML-Reference: <1250518688-7207-4-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-17 14:18:08 +00:00
}