forked from Minki/linux
66f69b2197
10922 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Masami Hiramatsu
|
66f69b2197 |
perf probe: Support DW_AT_const_value constant value
Support DW_AT_const_value for variable assignment instead of location. Note that this requires ftrace supporting immediate value. Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Cc: Steven Rostedt (VMware) <rostedt@goodmis.org> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Link: http://lore.kernel.org/lkml/157406476012.24476.16096289871757175775.stgit@devnote2 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Masami Hiramatsu
|
72363540c0 |
perf probe: Support multiprobe event
Support multiprobe event if the event is based on function and lines and
kernel supports it. In this case, perf probe creates the first probe
with an event, and tries to append following probes on that event, since
those probes must be on the same source code line.
Before this patch;
# perf probe -a vfs_read:18
Added new events:
probe:vfs_read_L18 (on vfs_read:18)
probe:vfs_read_L18_1 (on vfs_read:18)
You can now use it in all perf tools, such as:
perf record -e probe:vfs_read_L18_1 -aR sleep 1
#
After this patch (on multiprobe supported kernel)
# perf probe -a vfs_read:18
Added new events:
probe:vfs_read_L18 (on vfs_read:18)
probe:vfs_read_L18 (on vfs_read:18)
You can now use it in all perf tools, such as:
perf record -e probe:vfs_read_L18 -aR sleep 1
#
Committer testing:
On a kernel that doesn't support multiprobe events, after this patch:
# uname -a
Linux quaco 5.3.8-200.fc30.x86_64 #1 SMP Tue Oct 29 14:46:22 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
# grep append /sys/kernel/debug/tracing/README
be modified by appending '.descending' or '.ascending' to a
can be modified by appending any of the following modifiers
#
# perf probe -a vfs_read:18
Added new events:
probe:vfs_read_L18 (on vfs_read:18)
probe:vfs_read_L18_1 (on vfs_read:18)
You can now use it in all perf tools, such as:
perf record -e probe:vfs_read_L18_1 -aR sleep 1
# perf probe -l
probe:vfs_read_L18 (on vfs_read:18@fs/read_write.c)
probe:vfs_read_L18_1 (on vfs_read:18@fs/read_write.c)
#
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Tom Zanussi <tom.zanussi@linux.intel.com>
Link: http://lore.kernel.org/lkml/157406475010.24476.586290752591512351.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Masami Hiramatsu
|
15354d5469 |
perf probe: Generate event name with line number
Generate event name from function name with line number as
<function>_L<line_number>. Note that this is only for the new event
which is defined by the line number of function (except for line 0).
If there is another event on same line, you have to use
"-f" option. In that case, the new event has "_1" suffix.
e.g.
# perf probe -a kernel_read:2
Added new event:
probe:kernel_read_L2 (on kernel_read:2)
You can now use it in all perf tools, such as:
perf record -e probe:kernel_read_L2 -aR sleep 1
But if we omit the line number or 0th line, it will
have no suffix.
# perf probe -a kernel_read:0
Added new event:
probe:kernel_read (on kernel_read)
You can now use it in all perf tools, such as:
perf record -e probe:kernel_read -aR sleep 1
probe:kernel_read (on kernel_read@linux-5.0.0/fs/read_write.c)
probe:kernel_read_L2 (on kernel_read:2@linux-5.0.0/fs/read_write.c)
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Tom Zanussi <tom.zanussi@linux.intel.com>
Link: http://lore.kernel.org/lkml/157406474026.24476.2828897745502059569.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Masami Hiramatsu
|
499144c83d |
perf probe: Do not show non representive lines by perf-probe -L
Since perf probe -L shows non representive lines, it can be mislead
users where user can put probes. This prevents to show such non
representive lines so that user can understand which lines user can
probe.
# perf probe -L kernel_read
<kernel_read@/build/linux-pvZVvI/linux-5.0.0/fs/read_write.c:0>
0 ssize_t kernel_read(struct file *file, void *buf, size_t count, loff_t *pos)
{
2 mm_segment_t old_fs;
ssize_t result;
old_fs = get_fs();
6 set_fs(get_ds());
/* The cast to a user pointer is valid due to the set_fs() */
8 result = vfs_read(file, (void __user *)buf, count, pos);
9 set_fs(old_fs);
10 return result;
}
EXPORT_SYMBOL(kernel_read);
Committer testing:
Before:
# perf probe -L kernel_read
<kernel_read@/usr/src/debug/kernel-5.3.fc30/linux-5.3.8-200.fc30.x86_64/fs/read_write.c:0>
0 ssize_t kernel_read(struct file *file, void *buf, size_t count, loff_t *pos)
1 {
2 mm_segment_t old_fs;
3 ssize_t result;
5 old_fs = get_fs();
6 set_fs(KERNEL_DS);
/* The cast to a user pointer is valid due to the set_fs() */
8 result = vfs_read(file, (void __user *)buf, count, pos);
9 set_fs(old_fs);
10 return result;
}
EXPORT_SYMBOL(kernel_read);
#
See the 1, 3, 5 lines? They shouldn't be there, after this patch:
# perf probe -L kernel_read
<kernel_read@/usr/src/debug/kernel-5.3.fc30/linux-5.3.8-200.fc30.x86_64/fs/read_write.c:0>
0 ssize_t kernel_read(struct file *file, void *buf, size_t count, loff_t *pos)
{
2 mm_segment_t old_fs;
ssize_t result;
old_fs = get_fs();
6 set_fs(KERNEL_DS);
/* The cast to a user pointer is valid due to the set_fs() */
8 result = vfs_read(file, (void __user *)buf, count, pos);
9 set_fs(old_fs);
10 return result;
}
EXPORT_SYMBOL(kernel_read);
#
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Tom Zanussi <tom.zanussi@linux.intel.com>
Link: http://lore.kernel.org/lkml/157406473064.24476.2913278267727587314.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Masami Hiramatsu
|
1ae5d88a4e |
perf probe: Verify given line is a representive line
Verify user given probe line is a representive line (which doesn't share
the address with other lines or the line is the least line among the
lines which shares same address), and if not, it shows what is the
representive line.
Without this fix, user can put a probe on the lines which is not a a
representive line. But since this is not a representive line, perf probe
-l shows a representive line number instead of user given line number.
e.g. (put kernel_read:3, but listed as kernel_read:2)
# perf probe -a kernel_read:3
Added new event:
probe:kernel_read (on kernel_read:3)
You can now use it in all perf tools, such as:
perf record -e probe:kernel_read -aR sleep 1
# perf probe -l
probe:kernel_read (on kernel_read:2@linux-5.0.0/fs/read_write.c)
With this fix, perf probe doesn't allow user to put a probe on a
representive line, and tell what is the representive line.
# perf probe -a kernel_read:3
This line is sharing the addrees with other lines.
Please try to probe at kernel_read:2 instead.
Error: Failed to add events.
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Tom Zanussi <tom.zanussi@linux.intel.com>
Link: http://lore.kernel.org/lkml/157406472071.24476.14915451439785001021.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Masami Hiramatsu
|
57f95bf5f8 |
perf probe: Show correct statement line number by perf probe -l
The dwarf_getsrc_die() can return the line which is not a statement nor
the least line number among the lines which shares same address.
This can lead perf probe --list shows incorrect line number for probed
address.
To fix this, this introduces cu_getsrc_die() which returns only a
statement line and which is the least line number (we call it the
representive line for an address), and use it in cu_find_lineinfo().
Also, if the given address is the entry address of a real function,
cu_find_lineinfo() returns the function declared line number instead of
the start line number of the function body.
For example, without this change perf probe -l shows incorrect line as
below.
# perf probe -a kernel_read:2
Added new event:
probe:kernel_read (on kernel_read:2)
You can now use it in all perf tools, such as:
perf record -e probe:kernel_read -aR sleep 1
# perf probe -l
probe:kernel_read (on kernel_read:1@linux-5.0.0/fs/read_write.c)
With this fix, it shows correct line number as below;
# perf probe -l
probe:kernel_read (on kernel_read:2@linux-5.0.0/fs/read_write.c)
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Tom Zanussi <tom.zanussi@linux.intel.com>
Link: http://lore.kernel.org/lkml/157406471067.24476.17463149618465494448.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Adrian Hunter
|
1e5f015442 |
x86/insn: perf tools: Add some instructions to the new instructions test
Add to the "x86 instruction decoder - new instructions" test the following instructions: cldemote tpause umonitor umwait movdiri movdir64b enqcmd enqcmds encls enclu enclv pconfig wbnoinvd For information about the instructions, refer Intel SDM May 2019 (325462-070US) and Intel Architecture Instruction Set Extensions May 2019 (319433-037). Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Reviewed-by: Andi Kleen <ak@linux.intel.com> Cc: Borislav Petkov <bp@alien8.de> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: x86@kernel.org Link: http://lore.kernel.org/lkml/20191115135447.6519-2-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
Arnaldo Carvalho de Melo
|
7624e69465 |
perf map: Move seldom used ->flags field to second cacheline
So we start with:
$ pahole -C map ~/bin/perf
struct map {
union {
struct rb_node rb_node __attribute__((__aligned__(8))); /* 0 24 */
struct list_head node; /* 0 16 */
} __attribute__((__aligned__(8))); /* 0 24 */
u64 start; /* 24 8 */
u64 end; /* 32 8 */
_Bool erange_warned:1; /* 40: 0 1 */
_Bool priv:1; /* 40: 1 1 */
/* XXX 6 bits hole, try to pack */
/* XXX 3 bytes hole, try to pack */
u32 prot; /* 44 4 */
u32 flags; /* 48 4 */
/* XXX 4 bytes hole, try to pack */
u64 pgoff; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
u64 reloc; /* 64 8 */
u32 maj; /* 72 4 */
u32 min; /* 76 4 */
u64 ino; /* 80 8 */
u64 ino_generation; /* 88 8 */
u64 (*map_ip)(struct map *, u64); /* 96 8 */
u64 (*unmap_ip)(struct map *, u64); /* 104 8 */
struct dso * dso; /* 112 8 */
refcount_t refcnt; /* 120 4 */
/* size: 128, cachelines: 2, members: 17 */
/* sum members: 116, holes: 2, sum holes: 7 */
/* sum bitfield members: 2 bits, bit holes: 1, sum bit holes: 6 bits */
/* padding: 4 */
/* forced alignments: 1 */
} __attribute__((__aligned__(8)));
$
and 'flags' is seldom used when printing details about the map or with
the "cacheline" sort order, we can move them it to the second cacheline,
that will allow combining it with 'refcnt', that is only four bytes:
$ pahole -C map ~/bin/perf
struct map {
union {
struct rb_node rb_node __attribute__((__aligned__(8))); /* 0 24 */
struct list_head node; /* 0 16 */
} __attribute__((__aligned__(8))); /* 0 24 */
u64 start; /* 24 8 */
u64 end; /* 32 8 */
_Bool erange_warned:1; /* 40: 0 1 */
_Bool priv:1; /* 40: 1 1 */
/* XXX 6 bits hole, try to pack */
/* XXX 3 bytes hole, try to pack */
u32 prot; /* 44 4 */
u64 pgoff; /* 48 8 */
u64 reloc; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
u32 maj; /* 64 4 */
u32 min; /* 68 4 */
u64 ino; /* 72 8 */
u64 ino_generation; /* 80 8 */
u64 (*map_ip)(struct map *, u64); /* 88 8 */
u64 (*unmap_ip)(struct map *, u64); /* 96 8 */
struct dso * dso; /* 104 8 */
refcount_t refcnt; /* 112 4 */
u32 flags; /* 116 4 */
/* size: 120, cachelines: 2, members: 17 */
/* sum members: 116, holes: 1, sum holes: 3 */
/* sum bitfield members: 2 bits, bit holes: 1, sum bit holes: 6 bits */
/* forced alignments: 1 */
/* last cacheline: 56 bytes */
} __attribute__((__aligned__(8)));
$
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-2cdw3zlw1mkamaf7nqtdlxfi@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
dbc984c961 |
perf map: Use bitmap for booleans
The map->priv and map->erange_warned are seldom used, the first only in
tests/vmlinux-kallsyms.c, the later only when hist_entry__inc_addr_samples()
returns -ERANGE in 'perf top', which are really rare occasions, so make
them a bool bitfield.
This will open up space for other members on the first cacheline.
$ pahole -C map ~/bin/perf
struct map {
union {
struct rb_node rb_node __attribute__((__aligned__(8))); /* 0 24 */
struct list_head node; /* 0 16 */
} __attribute__((__aligned__(8))); /* 0 24 */
u64 start; /* 24 8 */
u64 end; /* 32 8 */
_Bool erange_warned:1; /* 40: 0 1 */
_Bool priv:1; /* 40: 1 1 */
/* XXX 6 bits hole, try to pack */
/* XXX 3 bytes hole, try to pack */
u32 prot; /* 44 4 */
u32 flags; /* 48 4 */
/* XXX 4 bytes hole, try to pack */
u64 pgoff; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
u64 reloc; /* 64 8 */
u32 maj; /* 72 4 */
u32 min; /* 76 4 */
u64 ino; /* 80 8 */
u64 ino_generation; /* 88 8 */
u64 (*map_ip)(struct map *, u64); /* 96 8 */
u64 (*unmap_ip)(struct map *, u64); /* 104 8 */
struct dso * dso; /* 112 8 */
refcount_t refcnt; /* 120 4 */
/* size: 128, cachelines: 2, members: 17 */
/* sum members: 116, holes: 2, sum holes: 7 */
/* sum bitfield members: 2 bits, bit holes: 1, sum bit holes: 6 bits */
/* padding: 4 */
/* forced alignments: 1 */
} __attribute__((__aligned__(8)));
$
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-g5545pcq4ff0wr17tfb1piqt@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Adrian Hunter
|
aceb98261e |
perf callchain: Fix segfault in thread__resolve_callchain_sample()
Do not dereference 'chain' when it is NULL.
$ perf record -e intel_pt//u -e branch-misses:u uname
$ perf report --itrace=l --branch-history
perf: Segmentation fault
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
a7c2b572e2 |
perf map_groups: Auto sort maps by name, if needed
There are still lots of lookups by name, even if just when loading
vmlinux, till that code is studied to figure out if its possible to do
away with those map lookup by names, provide a way to sort it using
libc's qsort/bsearch.
Doing it at the first lookup defers the sorting a bit, and as the code
stands now, is never done for user maps, just for the kernel ones.
# perf probe -l
# perf probe -x ~/bin/perf -L __map_groups__find_by_name
<__map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 static struct map *__map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
struct map **mapp;
4 if (mg->maps_by_name == NULL &&
5 map__groups__sort_by_name_from_rbtree(mg))
6 return NULL;
8 mapp = bsearch(name, mg->maps_by_name, mg->nr_maps, sizeof(*mapp), map__strcmp_name);
9 if (mapp)
10 return *mapp;
11 return NULL;
12 }
struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
{
# perf probe -x ~/bin/perf 'found=__map_groups__find_by_name:10 name:string'
Added new event:
probe_perf:found (on __map_groups__find_by_name:10 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:found -aR sleep 1
#
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
/*
* If we have mg->maps_by_name, then the name isn't in the rbtree,
* as mg->maps_by_name mirrors the rbtree when lookups by name are
* made.
*/
16 map = __map_groups__find_by_name(mg, name);
17 if (map || mg->maps_by_name != NULL)
18 goto out_unlock;
/* Fallback to traversing the rbtree... */
21 maps__for_each_entry(maps, map)
22 if (strcmp(map->dso->short_name, name) == 0) {
23 mg->last_search_by_name = map;
24 goto out_unlock;
}
27 map = NULL;
out_unlock:
30 up_read(&maps->lock);
31 return map;
32 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
# perf probe -x ~/bin/perf 'fallback=map_groups__find_by_name:21 name:string'
Added new events:
probe_perf:fallback (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
probe_perf:fallback_1 (on map_groups__find_by_name:21 in /home/acme/bin/perf with name:string)
You can now use it in all perf tools, such as:
perf record -e probe_perf:fallback_1 -aR sleep 1
#
# perf probe -l
probe_perf:fallback (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:fallback_1 (on map_groups__find_by_name:21@util/symbol.c in /home/acme/bin/perf with name_string)
probe_perf:found (on __map_groups__find_by_name:10@util/symbol.c in /home/acme/bin/perf with name_string)
#
# perf stat -e probe_perf:*
Now run 'perf top' in another term and then, after a while, stop 'perf stat':
Furthermore, if we ask for interval printing, we can see that that is done just
at the start of the workload:
# perf stat -I1000 -e probe_perf:*
# time counts unit events
1.000319513 0 probe_perf:found
1.000319513 0 probe_perf:fallback_1
1.000319513 0 probe_perf:fallback
2.001868092 23,251 probe_perf:found
2.001868092 0 probe_perf:fallback_1
2.001868092 0 probe_perf:fallback
3.002901597 0 probe_perf:found
3.002901597 0 probe_perf:fallback_1
3.002901597 0 probe_perf:fallback
4.003358591 0 probe_perf:found
4.003358591 0 probe_perf:fallback_1
4.003358591 0 probe_perf:fallback
^C
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-c5lmbyr14x448rcfii7y6t3k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
a94ab91a54 |
perf machine: No need to check if kernel module maps pre-exist
We'only populating maps for kernel modules either from perf.data file PERF_RECORD_MMAP records or when parsing /proc/modules, so there is no need to first look if we already have those module maps in the list, that would mean the kernel has duplicate entries. So ditch one use of looking up maps by name. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-gnzjg2hhuz6jnrw91m35059y@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
6e0a9b3dfa |
perf record: No need to process the synthesized MMAP events twice
At the end of a 'perf record' session, by default, we'll process all samples and populate the threads, maps, etc so as to find out which of the DSOs got samples, to reduce the size of the build-id table we'll add to the perf.data headers. But we don't need to process the PERF_RECORD_MMAP events synthesized for the kernel modules, as we have those already via perf_session__create_kernel_maps(), so add mmap/mmap2 handlers that first look at event->header.misc to see if the event is for a user map, bailing out if not. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-mofoxvcx2dryppcw3o689jdd@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
f068435d9b |
perf map: No need to adjust the long name of modules
At some point in the past we needed to make sure we would get the long name of modules and not just what we get from /proc/modules, but that need, as described in the cset that introduced the adjustment function: Fixes: |
||
|
Arnaldo Carvalho de Melo
|
1ae14516cb |
perf map_groups: Add a front end cache for map lookups by name
Lets see if it helps:
First look at the probeable lines for the function that does lookups by
name in a map_groups struct:
# perf probe -x ~/bin/perf -L map_groups__find_by_name
<map_groups__find_by_name@/home/acme/git/perf/tools/perf/util/symbol.c:0>
0 struct map *map_groups__find_by_name(struct map_groups *mg, const char *name)
1 {
2 struct maps *maps = &mg->maps;
struct map *map;
5 down_read(&maps->lock);
7 if (mg->last_search_by_name && strcmp(mg->last_search_by_name->dso->short_name, name) == 0) {
8 map = mg->last_search_by_name;
9 goto out_unlock;
}
12 maps__for_each_entry(maps, map)
13 if (strcmp(map->dso->short_name, name) == 0) {
14 mg->last_search_by_name = map;
15 goto out_unlock;
}
18 map = NULL;
out_unlock:
21 up_read(&maps->lock);
22 return map;
23 }
int dso__load_vmlinux(struct dso *dso, struct map *map,
const char *vmlinux, bool vmlinux_allocated)
#
Now add a probe to the place where we reuse the last search:
# perf probe -x ~/bin/perf map_groups__find_by_name:8
Added new event:
probe_perf:map_groups__find_by_name (on map_groups__find_by_name:8 in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:map_groups__find_by_name -aR sleep 1
#
Now lets do a system wide 'perf stat' counting those events:
# perf stat -e probe_perf:*
Leave it running and lets do a 'perf top', then, after a while, stop the
'perf stat':
# perf stat -e probe_perf:*
^C
Performance counter stats for 'system wide':
3,603 probe_perf:map_groups__find_by_name
44.565253139 seconds time elapsed
#
yeah, good to have.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-tcz37g3nxv3tvxw3q90vga3p@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
c5c584d2db |
perf maps: Do not use an rbtree to sort by map name
This is only used for the kernel maps, shave 24 bytes out 'struct map'
and just traverse the existing per ip rbtree to look for maps by name,
use a front end cache to reuse the last search if its the same name.
After this 'struct map' is down to just two cachelines:
$ pahole -C map ~/bin/perf
struct map {
union {
struct rb_node rb_node __attribute__((__aligned__(8))); /* 0 24 */
struct list_head node; /* 0 16 */
} __attribute__((__aligned__(8))); /* 0 24 */
u64 start; /* 24 8 */
u64 end; /* 32 8 */
_Bool erange_warned; /* 40 1 */
/* XXX 3 bytes hole, try to pack */
u32 priv; /* 44 4 */
u32 prot; /* 48 4 */
u32 flags; /* 52 4 */
u64 pgoff; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
u64 reloc; /* 64 8 */
u32 maj; /* 72 4 */
u32 min; /* 76 4 */
u64 ino; /* 80 8 */
u64 ino_generation; /* 88 8 */
u64 (*map_ip)(struct map *, u64); /* 96 8 */
u64 (*unmap_ip)(struct map *, u64); /* 104 8 */
struct dso * dso; /* 112 8 */
refcount_t refcnt; /* 120 4 */
/* size: 128, cachelines: 2, members: 17 */
/* sum members: 121, holes: 1, sum holes: 3 */
/* padding: 4 */
/* forced alignments: 1 */
} __attribute__((__aligned__(8)));
$
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-bvr8fqfgzxtgnhnwt5sssx5g@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
bcb8af5c46 |
perf maps: Purge the entries from maps->names in __maps__purge()
No need to iterate via the ->names rbtree, as all the entries there
as in maps->entries as well, reuse __maps__purge() for that.
Doing it this way we can kill maps__for_each_entry_by_name(),
maps__for_each_entry_by_name_safe(), maps__{first,next}_by_name().
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-ps0nrio8pydyo23rr2s696ue@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Adrian Hunter
|
af833988c0 |
perf scripts python: exported-sql-viewer.py: Fix use of TRUE with SQLite
Prior to version 3.23 SQLite does not support TRUE or FALSE, so always
use 1 and 0 for SQLite.
Fixes:
|
||
James Clark
|
da3ef7f6cd |
perf vendor events power9: Fix commas so PMU event files are valid JSON
No functional change.
Remove extra commas in the power9 JSON files so that the files
can be parsed and validated by other utilities such as Python
that fail to parse invalid JSON.
Before:
$ diffstat -l -p1 /wb/1.patch | while read filename ; do echo $filename ; cat $filename | json_verify ; done
tools/perf/pmu-events/arch/powerpc/power9/cache.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x300
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/floating-point.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x141
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/frontend.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x250
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/marked.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x301
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/memory.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x300
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/other.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x308
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/pipeline.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x4D0
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/pmc.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x200
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power9/translation.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x1E"
(right here) ------^
JSON is invalid
$
After:
$ diffstat -l -p1 /wb/1.patch | while read filename ; do echo $filename ; cat $filename | json_verify ; done
tools/perf/pmu-events/arch/powerpc/power9/cache.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/floating-point.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/frontend.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/marked.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/memory.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/other.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/pipeline.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/pmc.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power9/translation.json
JSON is valid
$
Signed-off-by: James Clark <james.clark@arm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kevin Mooney <kevin.mooney@arm.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: nd@arm.com
Link: http://lore.kernel.org/lkml/20191112160342.26470-3-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
James Clark
|
835e5bd909 |
perf vendor events power8: Fix commas so PMU event files are valid JSON
No functional change.
Remove extra commas in the power8 JSON files so that the files
can be parsed and validated by other utilities such as Python
that fail to parse invalid JSON.
Committer testing:
Before:
$ diffstat -l -p1 /wb/1.patch | while read filename ; do echo $filename ; cat $filename | json_verify ; done
tools/perf/pmu-events/arch/powerpc/power8/cache.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x4c0
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/floating-point.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x200
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/frontend.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x250
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/marked.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x351
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/memory.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x100
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/other.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x1f0
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/pipeline.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x100
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/pmc.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x200
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/powerpc/power8/translation.json
parse error: invalid object key (must be a string)
[ {, "EventCode": "0x4c0
(right here) ------^
JSON is invalid
$
After:
$ diffstat -l -p1 /wb/1.patch | while read filename ; do echo $filename ; cat $filename | json_verify ; done
tools/perf/pmu-events/arch/powerpc/power8/cache.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/floating-point.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/frontend.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/marked.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/memory.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/other.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/pipeline.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/pmc.json
JSON is valid
tools/perf/pmu-events/arch/powerpc/power8/translation.json
JSON is valid
$
Signed-off-by: James Clark <james.clark@arm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kevin Mooney <kevin.mooney@arm.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: nd@arm.com
Link: http://lore.kernel.org/lkml/20191112160342.26470-2-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
James Clark
|
a44e4f3ab1 |
perf vendor events arm64: Fix commas so PMU event files are valid JSON
No functional change.
Add and remove extra commas in the arm64 JSON files so that the files
can be parsed and validated by other utilities such as Python that fail
to parse invalid JSON.
Committer testing:
Before:
$ diffstat -l -p1 /wb/1.patch | while read filename ; do echo $filename ; cat $filename | json_verify ; done
tools/perf/pmu-events/arch/arm64/ampere/emag/branch.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/bus.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/cache.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/clock.json
parse error: unallowed token at this point in JSON text
[ { "PublicDescrip
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/exception.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/instruction.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/intrinsic.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/memory.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/ampere/emag/pipeline.json
parse error: unallowed token at this point in JSON text
[ { "PublicDescrip
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/arm/cortex-a53/branch.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent": "BR
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/arm/cortex-a53/bus.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent":
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/arm/cortex-a53/other.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent":
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/arm/cortex-a57-a72/core-imp-def.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/armv8-recommended.json
parse error: after array element, I expect ',' or ']'
[ { "PublicDescrip
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/cavium/thunderx2/core-imp-def.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/core-imp-def.json
parse error: invalid object key (must be a string)
[ { "ArchStdEvent"
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/uncore-ddrc.json
parse error: invalid object key (must be a string)
[ { "EventCode": "0x00
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/uncore-hha.json
parse error: invalid object key (must be a string)
[ { "EventCode": "0x00
(right here) ------^
JSON is invalid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/uncore-l3c.json
parse error: invalid object key (must be a string)
[ { "EventCode": "0x00
(right here) ------^
JSON is invalid
$
After:
$ diffstat -l -p1 /wb/1.patch | while read filename ; do echo $filename ; cat $filename | json_verify ; done
tools/perf/pmu-events/arch/arm64/ampere/emag/branch.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/bus.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/cache.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/clock.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/exception.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/instruction.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/intrinsic.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/memory.json
JSON is valid
tools/perf/pmu-events/arch/arm64/ampere/emag/pipeline.json
JSON is valid
tools/perf/pmu-events/arch/arm64/arm/cortex-a53/branch.json
JSON is valid
tools/perf/pmu-events/arch/arm64/arm/cortex-a53/bus.json
JSON is valid
tools/perf/pmu-events/arch/arm64/arm/cortex-a53/other.json
JSON is valid
tools/perf/pmu-events/arch/arm64/arm/cortex-a57-a72/core-imp-def.json
JSON is valid
tools/perf/pmu-events/arch/arm64/armv8-recommended.json
JSON is valid
tools/perf/pmu-events/arch/arm64/cavium/thunderx2/core-imp-def.json
JSON is valid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/core-imp-def.json
JSON is valid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/uncore-ddrc.json
JSON is valid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/uncore-hha.json
JSON is valid
tools/perf/pmu-events/arch/arm64/hisilicon/hip08/uncore-l3c.json
JSON is valid
$
Signed-off-by: James Clark <james.clark@arm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kevin Mooney <kevin.mooney@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: nd@arm.com
Link: http://lore.kernel.org/lkml/20191112160342.26470-1-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Ian Rogers
|
e1e9b78d39 |
perf parse: Use YYABORT to clear stack after failure, plugging leaks
Using return rather than YYABORT means that the stack isn't cleared up following a failure. The change to YYABORT means the return value is 1 rather than -1, but the callers just check for a result of 0 (success). Add missing free of a list when an error occurs in event_pmu. Signed-off-by: Ian Rogers <irogers@google.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lore.kernel.org/lkml/20191109075840.181231-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
Ravi Bangoria
|
ccd26741f5 |
perf tool: Provide an option to print perf_event_open args and return value
Perf record with verbose=2 already prints this information along with
whole lot of other traces which requires lot of scrolling. Introduce
an option to print only perf_event_open() arguments and return value.
Sample o/p:
$ perf --debug perf-event-open=1 record -- ls > /dev/null
------------------------------------------------------------
perf_event_attr:
size 112
{ sample_period, sample_freq } 4000
sample_type IP|TID|TIME|PERIOD
read_format ID
disabled 1
inherit 1
exclude_kernel 1
mmap 1
comm 1
freq 1
enable_on_exec 1
task 1
precise_ip 3
sample_id_all 1
exclude_guest 1
mmap2 1
comm_exec 1
ksymbol 1
bpf_event 1
------------------------------------------------------------
sys_perf_event_open: pid 4308 cpu 0 group_fd -1 flags 0x8 = 4
sys_perf_event_open: pid 4308 cpu 1 group_fd -1 flags 0x8 = 5
sys_perf_event_open: pid 4308 cpu 2 group_fd -1 flags 0x8 = 6
sys_perf_event_open: pid 4308 cpu 3 group_fd -1 flags 0x8 = 8
sys_perf_event_open: pid 4308 cpu 4 group_fd -1 flags 0x8 = 9
sys_perf_event_open: pid 4308 cpu 5 group_fd -1 flags 0x8 = 10
sys_perf_event_open: pid 4308 cpu 6 group_fd -1 flags 0x8 = 11
sys_perf_event_open: pid 4308 cpu 7 group_fd -1 flags 0x8 = 12
------------------------------------------------------------
perf_event_attr:
type 1
size 112
config 0x9
watermark 1
sample_id_all 1
bpf_event 1
{ wakeup_events, wakeup_watermark } 1
------------------------------------------------------------
sys_perf_event_open: pid -1 cpu 0 group_fd -1 flags 0x8
sys_perf_event_open failed, error -13
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.002 MB perf.data (9 samples) ]
Committer notes:
Just like the 'verbose' variable this new 'debug_peo_args' needs to be
added to util/python.c, since we don't link the debug.o file in the
python binding, which ended up making 'perf test python' fail with:
# perf test -v python
18: 'import perf' in python :
--- start ---
test child forked, pid 19237
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: /tmp/build/perf/python/perf.so: undefined symbol: debug_peo_args
test child finished with -1
---- end ----
'import perf' in python: FAILED!
#
After adding that new variable to util/python.c:
# perf test -v python
18: 'import perf' in python :
--- start ---
test child forked, pid 22364
test child finished with 0
---- end ----
'import perf' in python: Ok
#
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: http://lore.kernel.org/lkml/20191108094128.28769-1-ravi.bangoria@linux.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
7b018e2987 |
perf map: Remove ->groups from 'struct map'
With this 'struct map' uses a bit over 3 cachelines: $ pahole -C map ~/bin/perf <SNIP> /* --- cacheline 2 boundary (128 bytes) --- */ u64 (*unmap_ip)(struct map *, u64); /* 128 8 */ struct dso * dso; /* 136 8 */ refcount_t refcnt; /* 144 4 */ /* size: 152, cachelines: 3, members: 18 */ /* sum members: 145, holes: 1, sum holes: 3 */ /* padding: 4 */ /* forced alignments: 2 */ /* last cacheline: 24 bytes */ } __attribute__((__aligned__(8))); $ We probably can move map->map/unmap_ip() moved to 'struct map_groups', that will shave more 16 bytes, getting this almost to two cachelines. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-ymlv3nzpofv2fugnjnizkrwy@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
3f662fc08d |
perf map: Combine maps__fixup_overlappings with its only use
In the process we can kill some of the struct map->groups usage, trying to get rid of this per-full struct map fields getting in the way of sharing a map across father/parent processes. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-e50eqtqw3za24vmbjnqmmcs6@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
94e44b9ca5 |
perf annotate: Stop using map->groups, use map_symbol->mg instead
These were the last uses of map->groups, next cset will nuke it. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-n3g0foos7l7uxq9nar0zo0vj@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
08f6680e62 |
perf tools: Add a 'struct map_groups' pointer to 'struct map_symbol'
And fill it whenever we setup a a 'struct map_symbol', now we need to use it, next cset. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-fzwfcnddenz1o7uj1fzw3g46@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
93fcce96c7 |
perf symbols: Use kmaps(map)->machine when we know its a kernel map
And then stop using map->groups to achieve that.
To test that that branch is being taken, probe the function that is only
called from there and then run something like 'perf top' in another
xterm:
# perf probe -x ~/bin/perf machine__map_x86_64_entry_trampolines
Added new event:
probe_perf:machine__map_x86_64_entry_trampolines (on machine__map_x86_64_entry_trampolines in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:machine__map_x86_64_entry_trampolines -aR sleep 1
# perf trace -e probe_perf:*
0.000 bash/10614 probe_perf:machine__map_x86_64_entry_trampolines(__probe_ip: 5224944)
^C#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-lgrrzdxo2p9liq2keivcg887@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
d46a4cdf49 |
pref tools: Make 'struct addr_map_symbol' contain 'struct map_symbol'
So that we pass that substructure around and with it consolidate lots of functions that receive a (map, symbol) pair and now can receive just a 'struct map_symbol' pointer. This further paves the way to add 'struct map_groups' to 'struct map_symbol' so that we can have all we need for annotation so that we can ditch 'struct map'->groups, i.e. have the map_groups pointer in a more central place, avoiding the pointer in the 'struct map' that have tons of instances. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-fs90ttd9q12l7989fo7pw81q@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
5f0fef8ac3 |
perf callchain: Use 'struct map_symbol' in 'struct callchain_cursor_node'
To ease passing around map+symbol, just like done for other parts of the tree recently. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
c1529738f5 |
perf unwind: Use 'struct map_symbol' in 'struct unwind_entry'
To help in passing that info around to callchain routines that, for the same reason, are moving to use 'struct map_symbol'. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-epsiibeprpxa8qpwji47uskc@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
2975489458 |
perf annotate: Pass a 'map_symbol' in places receiving a pair of 'map' and 'symbol' pointers
We are already passing things like: symbol__annotate(ms->sym, ms->map, ...) So shorten the signature of such functions to receive the 'map_symbol' pointer. This also paves the way to having the 'struct map_groups' pointer in the 'struct map_symbol' so that we can get rid of 'struct map'->groups. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-23yx8v1t41nzpkpi7rdrozww@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
d3a022cbdc |
perf tools: Add map_groups to 'struct addr_location'
From there we can get al->mg->machine, so replace that field with the more useful 'struct map_groups' that for now we're obtaining from al->map->groups, and that is one thing getting into the way of maps being fully shareable. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-4qdducrm32tgrjupcp0kjh1e@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
9d355b381b |
perf map_groups: Pass the object to map_groups__find_ams()
We were just passing a map to look for and reuse its map->groups member, but the idea is that this is going away, as a map can be in multiple rb_trees when being reused via a map_node, so do as all the other map_groups methods and pass as its first arg the object being operated on. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-nmi2pbggqloogwl6vxrvex5a@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
f2baa060cd |
perf symbols: Stop using map->groups, we can use kmaps instead
To test that that function is being called I just added a probe on that
place, enabled it via 'perf trace' asking for at most 16 levels of
backtraces, system wide, and then ran 'perf top' on another xterm,
voilà:
# perf probe -x ~/bin/perf dso__process_kernel_symbol
Added new event:
probe_perf:dso__process_kernel_symbol (on dso__process_kernel_symbol in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:dso__process_kernel_symbol -aR sleep 1
# perf trace -e probe_perf:dso__process_kernel_symbol/max-stack=16/ --max-events=2
# perf trace -e probe_perf:dso__process_kernel_symbol/max-stack=16/ --max-events=2
0.000 :17345/17345 probe_perf:dso__process_kernel_symbol(__probe_ip: 5680224)
dso__process_kernel_symbol (/home/acme/bin/perf)
dso__load_vmlinux (/home/acme/bin/perf)
dso__load_vmlinux_path (/home/acme/bin/perf)
dso__load (/home/acme/bin/perf)
map__load (/home/acme/bin/perf)
thread__find_map (/home/acme/bin/perf)
machine__resolve (/home/acme/bin/perf)
deliver_event (/home/acme/bin/perf)
__ordered_events__flush.part.0 (/home/acme/bin/perf)
process_thread (/home/acme/bin/perf)
start_thread (/usr/lib64/libpthread-2.29.so)
0.064 :17345/17345 probe_perf:dso__process_kernel_symbol(__probe_ip: 5680224)
dso__process_kernel_symbol (/home/acme/bin/perf)
dso__load_vmlinux (/home/acme/bin/perf)
dso__load_vmlinux_path (/home/acme/bin/perf)
dso__load (/home/acme/bin/perf)
map__load (/home/acme/bin/perf)
thread__find_map (/home/acme/bin/perf)
machine__resolve (/home/acme/bin/perf)
deliver_event (/home/acme/bin/perf)
__ordered_events__flush.part.0 (/home/acme/bin/perf)
process_thread (/home/acme/bin/perf)
start_thread (/usr/lib64/libpthread-2.29.so)
#
# perf stat -e probe_perf:dso__process_kernel_symbol
^C
Performance counter stats for 'system wide':
107,308 probe_perf:dso__process_kernel_symbol
8.215399813 seconds time elapsed
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-5fy66x5hr5ct9pmw84jkiwvm@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
de90d513b2 |
perf map: Use map->dso->kernel + map__kmaps() in map__kmaps()
Its equivalent to using map->groups to obtain the machine struct. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-bdbazuj4ggrmzxdviaqdrdwh@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
Ingo Molnar
|
56b2147f34 |
perf/core improvements and fixes:
perf report:
Jin Yao:
- Introduce --total-cycles, for basic block profiling, further using data
obtained from LBR, an example should suffice:
# perf record -b
^C[ perf record: Woken up 595 times to write data ]
[ perf record: Captured and wrote 156.672 MB perf.data (196873 samples) ]
# perf evlist -v
cycles: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CPU|PERIOD|BRANCH_STACK, read_format: ID, 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, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY
# perf report --total-cycles --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
# Total Lost Samples: 0
#
# Samples: 6M of event 'cycles'
# Event count (approx.): 6299936
#
# Sampled Sampled Avg Avg
# Cycles% Cycles Cycles% Cycles [Program Block Range] Shared Object
# ....... ...... ....... ..... .................................... ................
#
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
0.72% 544.5K 0.03% 230 [entry_64.S:657 -> entry_64.S:662] [kernel.vmlinux]
0.56% 541.8K 0.09% 672 [compiler.h:199 -> common.c:300] [kernel.vmlinux]
0.39% 293.2K 0.01% 104 [list_debug.c:43 -> list_debug.c:61] [kernel.vmlinux]
0.36% 278.6K 0.03% 272 [entry_64.S:1289 -> entry_64.S:1308] [kernel.vmlinux]
perf record:
Adrian Hunter:
- Allow storing perf.data in a directory together with a copy of /proc/kcore.
Jiwei Sun:
- Add support for limit perf output file size, i.e.:
# perf record --all-cpus -F 10000 --max-size=4M sleep 10h
[ perf record: perf size limit reached (4097 KB), stopping session ]
[ perf record: Woken up 6 times to write data ]
[ perf record: Captured and wrote 4.048 MB perf.data (54094 samples) ]
Terminated
# ls -lah perf.data
-rw-------. 1 root root 4.1M Nov 7 15:27 perf.data
#
perf stat:
Jiri Olsa:
- Add --per-node agregation support:
In live mode:
# perf stat -a -I 1000 -e cycles --per-node
# time node cpus counts unit events
1.000542550 N0 20 6,202,097 cycles
1.000542550 N1 20 639,559 cycles
2.002040063 N0 20 7,412,495 cycles
2.002040063 N1 20 2,185,577 cycles
3.003451699 N0 20 6,508,917 cycles
3.003451699 N1 20 765,607 cycles
...
Or in the record/report stat session:
# perf stat record -a -I 1000 -e cycles
# time counts unit events
1.000536937 10,008,468 cycles
2.002090152 9,578,539 cycles
3.003625233 7,647,869 cycles
4.005135036 7,032,086 cycles
^C 4.340902364 3,923,893 cycles
# perf stat report --per-node
# time node cpus counts unit events
1.000536937 N0 20 9,355,086 cycles
1.000536937 N1 20 653,382 cycles
2.002090152 N0 20 7,712,838 cycles
2.002090152 N1 20 1,865,701 cycles
...
perf probe:
Masami Hiramatsu:
Various fixes related to recent additions to the DWARF format:
- Fix to find range-only function instance
- Walk function lines in lexical blocks
- Fix to show function entry line as probe-able
- Fix wrong address verification
- Fix to probe a function which has no entry pc
- Fix to probe an inline function which has no entry pc
- Fix to list probe event with correct line number
- Fix to show inlined function callsite without entry_pc
- Fix to show ranges of variables in functions without entry_pc
- Return a better scope DIE if there is no best scope
- Skip end-of-sequence and non statement lines
- Filter out instances except for inlined subroutine and subprogram
- Fix to show calling lines of inlined functions
- Skip overlapped location on searching variables
perf inject:
Adrian Hunter:
- Do not strip evsels with --strip, as they are needed for create_gcov
(see the autofdo example in tools/perf/Documentation/intel-pt.txt).
Intel PT:
Adrian Hunter:
- Intel PT uses an auxtrace_cache to store the results of code-walking, to avoid
repeated decoding. Add an auxtrace_cache__remove to handle text poke events.
core:
Andi Kleen:
- Always preserve errno while cleaning up perf_event_open failures.
llvm:
Arnaldo Carvalho de Melo:
- No need to tell that the request for saving a .o file for BPF events, as
expressed in ~/.perfconfig was satisfied, make that a debug message.
perf vendor events:
Intel:
Haiyan Song:
- Update CascadelakeX events to v1.05.
- Update all the Intel JSON metrics from TMAM 3.6.
Treewide:
Ian Rogers:
- Improve error paths, plugging leaks found using LLVM tools
such as libFuzzer.
jevents:
Yunfeng Ye:
- Fix resource leak in process_mapfile() and main()
perf kvm:
Igor Lubashev:
- Use evlist layer api when possible.
libsubcmd:
James Clark:
- Move EXTRA_FLAGS to the end to allow overriding existing flags.
- Use -O0 with DEBUG=1
perf diff:
Jin Yao:
- Don't use hack to skip column length calculation
CoreSight ETM:
Leo yan:
- Fix definition of macro TO_CS_QUEUE_NR
ARM64:
John Garry:
- Do not try to include libelf header files when its feature detection
failed, fixing the cross build for ARM64.
perf tests:
Leo Yan:
- Fix out of bounds memory access in the backward ring buffer test.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCXcRowQAKCRCyPKLppCJ+
JxHcAQCTtl9N3zkNjLWif1i6AGKNU9TzYpup+jDR5J83ggLqgQD+O931nR9wXUOe
9bDUr45cNw3ZkRbc1558hKPWIsceJgU=
=Rko+
-----END PGP SIGNATURE-----
Merge tag 'perf-core-for-mingo-5.5-20191107' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
perf report:
Jin Yao:
- Introduce --total-cycles, for basic block profiling, further using data
obtained from LBR, an example should suffice:
# perf record -b
^C[ perf record: Woken up 595 times to write data ]
[ perf record: Captured and wrote 156.672 MB perf.data (196873 samples) ]
# perf evlist -v
cycles: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CPU|PERIOD|BRANCH_STACK, read_format: ID, 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, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY
# perf report --total-cycles --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
# Total Lost Samples: 0
#
# Samples: 6M of event 'cycles'
# Event count (approx.): 6299936
#
# Sampled Sampled Avg Avg
# Cycles% Cycles Cycles% Cycles [Program Block Range] Shared Object
# ....... ...... ....... ..... .................................... ................
#
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
0.72% 544.5K 0.03% 230 [entry_64.S:657 -> entry_64.S:662] [kernel.vmlinux]
0.56% 541.8K 0.09% 672 [compiler.h:199 -> common.c:300] [kernel.vmlinux]
0.39% 293.2K 0.01% 104 [list_debug.c:43 -> list_debug.c:61] [kernel.vmlinux]
0.36% 278.6K 0.03% 272 [entry_64.S:1289 -> entry_64.S:1308] [kernel.vmlinux]
perf record:
Adrian Hunter:
- Allow storing perf.data in a directory together with a copy of /proc/kcore.
Jiwei Sun:
- Add support for limit perf output file size, i.e.:
# perf record --all-cpus -F 10000 --max-size=4M sleep 10h
[ perf record: perf size limit reached (4097 KB), stopping session ]
[ perf record: Woken up 6 times to write data ]
[ perf record: Captured and wrote 4.048 MB perf.data (54094 samples) ]
Terminated
# ls -lah perf.data
-rw-------. 1 root root 4.1M Nov 7 15:27 perf.data
#
perf stat:
Jiri Olsa:
- Add --per-node agregation support:
In live mode:
# perf stat -a -I 1000 -e cycles --per-node
# time node cpus counts unit events
1.000542550 N0 20 6,202,097 cycles
1.000542550 N1 20 639,559 cycles
2.002040063 N0 20 7,412,495 cycles
2.002040063 N1 20 2,185,577 cycles
3.003451699 N0 20 6,508,917 cycles
3.003451699 N1 20 765,607 cycles
...
Or in the record/report stat session:
# perf stat record -a -I 1000 -e cycles
# time counts unit events
1.000536937 10,008,468 cycles
2.002090152 9,578,539 cycles
3.003625233 7,647,869 cycles
4.005135036 7,032,086 cycles
^C 4.340902364 3,923,893 cycles
# perf stat report --per-node
# time node cpus counts unit events
1.000536937 N0 20 9,355,086 cycles
1.000536937 N1 20 653,382 cycles
2.002090152 N0 20 7,712,838 cycles
2.002090152 N1 20 1,865,701 cycles
...
perf probe:
Masami Hiramatsu:
Various fixes related to recent additions to the DWARF format:
- Fix to find range-only function instance
- Walk function lines in lexical blocks
- Fix to show function entry line as probe-able
- Fix wrong address verification
- Fix to probe a function which has no entry pc
- Fix to probe an inline function which has no entry pc
- Fix to list probe event with correct line number
- Fix to show inlined function callsite without entry_pc
- Fix to show ranges of variables in functions without entry_pc
- Return a better scope DIE if there is no best scope
- Skip end-of-sequence and non statement lines
- Filter out instances except for inlined subroutine and subprogram
- Fix to show calling lines of inlined functions
- Skip overlapped location on searching variables
perf inject:
Adrian Hunter:
- Do not strip evsels with --strip, as they are needed for create_gcov
(see the autofdo example in tools/perf/Documentation/intel-pt.txt).
Intel PT:
Adrian Hunter:
- Intel PT uses an auxtrace_cache to store the results of code-walking, to avoid
repeated decoding. Add an auxtrace_cache__remove to handle text poke events.
core:
Andi Kleen:
- Always preserve errno while cleaning up perf_event_open failures.
llvm:
Arnaldo Carvalho de Melo:
- No need to tell that the request for saving a .o file for BPF events, as
expressed in ~/.perfconfig was satisfied, make that a debug message.
perf vendor events:
Intel:
Haiyan Song:
- Update CascadelakeX events to v1.05.
- Update all the Intel JSON metrics from TMAM 3.6.
Treewide:
Ian Rogers:
- Improve error paths, plugging leaks found using LLVM tools
such as libFuzzer.
jevents:
Yunfeng Ye:
- Fix resource leak in process_mapfile() and main()
perf kvm:
Igor Lubashev:
- Use evlist layer api when possible.
libsubcmd:
James Clark:
- Move EXTRA_FLAGS to the end to allow overriding existing flags.
- Use -O0 with DEBUG=1
perf diff:
Jin Yao:
- Don't use hack to skip column length calculation
CoreSight ETM:
Leo yan:
- Fix definition of macro TO_CS_QUEUE_NR
ARM64:
John Garry:
- Do not try to include libelf header files when its feature detection
failed, fixing the cross build for ARM64.
perf tests:
Leo Yan:
- Fix out of bounds memory access in the backward ring buffer test.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
||
|
Ingo Molnar
|
1ca7feb590 |
Linux 5.4-rc7
-----BEGIN PGP SIGNATURE----- iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAl3IqJQeHHRvcnZhbGRz QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGOiUH+gOEDwid5OODaFAd CggXugdFIlBZefKqGVNW5sjgX8pxFWHXuEMC8iNb6QXtQZdFrI6LFf9hhUDmzQtm 6y1LPxxEiTZjObMEsBNylb7tyzgujFHcAlp0Zro3w/HLCqmYTSP3FF46i2u6KZfL XhkpM4X7R7qxlfpdhlfESv/ElRGocZe6SwXfC7pcPo5flFcmkdu9ijqhNd/6CZ/h Nf9rTsD/wEDVUelFbgVN+LJzlaB0tsyc4Zbof07n8OsFZjhdEOop8gfM/kTBLcyY 6bh66SfDScdsNnC/l8csbPjSZRx+i+nQs67DyhGNnsSAFgHBZdC4Tb/2mDCwhCLR dUvuYZc= =1N6F -----END PGP SIGNATURE----- Merge tag 'v5.4-rc7' into perf/core, to pick up fixes Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
Linus Torvalds
|
b584a17628 |
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf tooling fixes from Thomas Gleixner: - Fix the time sorting algorithm which was broken due to truncation of big numbers - Fix the python script generator fail caused by a broken tracepoint array iterator * 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: perf tools: Fix time sorting perf tools: Remove unused trace_find_next_event() perf scripting engines: Iterate on tep event arrays directly |
||
Jin Yao
|
7fa46cbf20 |
perf report: Sort by sampled cycles percent per block for tui
Previous patch has implemented a new option "--total-cycles". But only
stdio mode is supported.
This patch supports the tui mode and support '--percent-limit'.
For example,
perf record -b ./div
perf report --total-cycles --percent-limit 1
# Samples: 2753248 of event 'cycles'
Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
26.04% 2.8M 0.40% 18 [div.c:42 -> div.c:39] div
15.17% 1.2M 0.16% 7 [random_r.c:357 -> random_r.c:380] libc-2.27.so
5.11% 402.0K 0.04% 2 [div.c:27 -> div.c:28] div
4.87% 381.6K 0.04% 2 [random.c:288 -> random.c:291] libc-2.27.so
4.53% 381.0K 0.04% 2 [div.c:40 -> div.c:40] div
3.85% 300.9K 0.02% 1 [div.c:22 -> div.c:25] div
3.08% 241.1K 0.02% 1 [rand.c:26 -> rand.c:27] libc-2.27.so
3.06% 240.0K 0.02% 1 [random.c:291 -> random.c:291] libc-2.27.so
2.78% 215.7K 0.02% 1 [random.c:298 -> random.c:298] libc-2.27.so
2.52% 198.3K 0.02% 1 [random.c:293 -> random.c:293] libc-2.27.so
2.36% 184.8K 0.02% 1 [rand.c:28 -> rand.c:28] libc-2.27.so
2.33% 180.5K 0.02% 1 [random.c:295 -> random.c:295] libc-2.27.so
2.28% 176.7K 0.02% 1 [random.c:295 -> random.c:295] libc-2.27.so
2.20% 168.8K 0.02% 1 [rand@plt+0 -> rand@plt+0] div
1.98% 158.2K 0.02% 1 [random_r.c:388 -> random_r.c:388] libc-2.27.so
1.57% 123.3K 0.02% 1 [div.c:42 -> div.c:44] div
1.44% 116.0K 0.42% 19 [random_r.c:357 -> random_r.c:394] libc-2.27.so
--------------------------------------------------
v7:
---
1. Since we have used use_browser in report__browse_block_hists
to support stdio mode, now we also add supporting for tui.
2. Move block tui browser code from ui/browsers/hists.c
to block-info.c.
v6:
---
Create report__tui_browse_block_hists in block-info.c
(codes are moved from builtin-report.c).
v5:
---
Fix a crash issue when running perf report without
'--total-cycles'. The issue is because the internal flag
is renamed from 'total_cycles' to 'total_cycles_mode' in
previous patch but this patch still uses 'total_cycles'
to check if the '--total-cycles' option is enabled, which
causes the code to be inconsistent.
v4:
---
Since the block collection is moved out of printing in
previous patch, this patch is updated accordingly for
tui supporting.
v3:
---
Minor change since the function name is changed:
block_total_cycles_percent -> block_info__total_cycles_percent
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
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: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20191107074719.26139-8-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Jin Yao
|
0b49f83657 |
perf report: Support --percent-limit for --total-cycles
We have already supported the '--total-cycles' option in previous patch.
It's also useful to show entries only above a threshold percent.
This patch enables '--percent-limit' for not showing entries
under that percent.
For example:
perf report --total-cycles --stdio --percent-limit 1
# To display the perf.data header info, please use --header/--header-only options.
#
#
# Total Lost Samples: 0
#
# Samples: 2M of event 'cycles'
# Event count (approx.): 2753248
#
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
# ............... .............. ........... .......... ................................................................. ....................
#
26.04% 2.8M 0.40% 18 [div.c:42 -> div.c:39] div
15.17% 1.2M 0.16% 7 [random_r.c:357 -> random_r.c:380] libc-2.27.so
5.11% 402.0K 0.04% 2 [div.c:27 -> div.c:28] div
4.87% 381.6K 0.04% 2 [random.c:288 -> random.c:291] libc-2.27.so
4.53% 381.0K 0.04% 2 [div.c:40 -> div.c:40] div
3.85% 300.9K 0.02% 1 [div.c:22 -> div.c:25] div
3.08% 241.1K 0.02% 1 [rand.c:26 -> rand.c:27] libc-2.27.so
3.06% 240.0K 0.02% 1 [random.c:291 -> random.c:291] libc-2.27.so
2.78% 215.7K 0.02% 1 [random.c:298 -> random.c:298] libc-2.27.so
2.52% 198.3K 0.02% 1 [random.c:293 -> random.c:293] libc-2.27.so
2.36% 184.8K 0.02% 1 [rand.c:28 -> rand.c:28] libc-2.27.so
2.33% 180.5K 0.02% 1 [random.c:295 -> random.c:295] libc-2.27.so
2.28% 176.7K 0.02% 1 [random.c:295 -> random.c:295] libc-2.27.so
2.20% 168.8K 0.02% 1 [rand@plt+0 -> rand@plt+0] div
1.98% 158.2K 0.02% 1 [random_r.c:388 -> random_r.c:388] libc-2.27.so
1.57% 123.3K 0.02% 1 [div.c:42 -> div.c:44] div
1.44% 116.0K 0.42% 19 [random_r.c:357 -> random_r.c:394] libc-2.27.so
Committer testing:
From second exapmple onwards slightly edited for brevity:
# perf report --total-cycles --percent-limit 2 --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
#
# Total Lost Samples: 0
#
# Samples: 6M of event 'cycles'
# Event count (approx.): 6299936
#
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
# ............... .............. ........... .......... ...................................................................... ....................
#
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
#
# (Tip: Create an archive with symtabs to analyse on other machine: perf archive)
#
# perf report --total-cycles --percent-limit 1 --stdio
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
1.75% 1.3M 8.34% 65.5K [memset-vec-unaligned-erms.S:147 -> memset-vec-unaligned-erms.S:151] libc-2.29.so
#
# perf report --total-cycles --percent-limit 0.7 --stdio
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
1.75% 1.3M 8.34% 65.5K [memset-vec-unaligned-erms.S:147 -> memset-vec-unaligned-erms.S:151] libc-2.29.so
0.72% 544.5K 0.03% 230 [entry_64.S:657 -> entry_64.S:662] [kernel.vmlinux]
#
-------------------------------------------
It only shows the entries which 'Sampled Cycles%' > 1%.
v7:
---
No functional change. Only fix the conflict issue because
previous patches are changed.
v6:
---
No functional change. Only fix the conflict issue because
previous patches are changed.
v5:
---
No functional change. Only fix the conflict issue because
previous patches are changed.
v4:
---
No functional change. Only fix the build issue because
previous patches are changed.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
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: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20191107074719.26139-7-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Jin Yao
|
6f7164fa23 |
perf report: Sort by sampled cycles percent per block for stdio
It would be useful to support sorting for all blocks by the sampled
cycles percent per block. This is useful to concentrate on the globally
hottest blocks.
This patch implements a new option "--total-cycles" which sorts all
blocks by 'Sampled Cycles%'. The 'Sampled Cycles%' is the percent:
percent = block sampled cycles aggregation / total sampled cycles
Note that, this patch only supports "--stdio" mode.
For example,
# perf record -b ./div
# perf report --total-cycles --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
# Total Lost Samples: 0
#
# Samples: 2M of event 'cycles'
# Event count (approx.): 2753248
#
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
# ............... .............. ........... .......... ................................................ .................
#
26.04% 2.8M 0.40% 18 [div.c:42 -> div.c:39] div
15.17% 1.2M 0.16% 7 [random_r.c:357 -> random_r.c:380] libc-2.27.so
5.11% 402.0K 0.04% 2 [div.c:27 -> div.c:28] div
4.87% 381.6K 0.04% 2 [random.c:288 -> random.c:291] libc-2.27.so
4.53% 381.0K 0.04% 2 [div.c:40 -> div.c:40] div
3.85% 300.9K 0.02% 1 [div.c:22 -> div.c:25] div
3.08% 241.1K 0.02% 1 [rand.c:26 -> rand.c:27] libc-2.27.so
3.06% 240.0K 0.02% 1 [random.c:291 -> random.c:291] libc-2.27.so
2.78% 215.7K 0.02% 1 [random.c:298 -> random.c:298] libc-2.27.so
2.52% 198.3K 0.02% 1 [random.c:293 -> random.c:293] libc-2.27.so
2.36% 184.8K 0.02% 1 [rand.c:28 -> rand.c:28] libc-2.27.so
2.33% 180.5K 0.02% 1 [random.c:295 -> random.c:295] libc-2.27.so
2.28% 176.7K 0.02% 1 [random.c:295 -> random.c:295] libc-2.27.so
2.20% 168.8K 0.02% 1 [rand@plt+0 -> rand@plt+0] div
1.98% 158.2K 0.02% 1 [random_r.c:388 -> random_r.c:388] libc-2.27.so
1.57% 123.3K 0.02% 1 [div.c:42 -> div.c:44] div
1.44% 116.0K 0.42% 19 [random_r.c:357 -> random_r.c:394] libc-2.27.so
0.25% 182.5K 0.02% 1 [random_r.c:388 -> random_r.c:391] libc-2.27.so
0.00% 48 1.07% 48 [x86_pmu_enable+284 -> x86_pmu_enable+298] [kernel.kallsyms]
0.00% 74 1.64% 74 [vm_mmap_pgoff+0 -> vm_mmap_pgoff+92] [kernel.kallsyms]
0.00% 73 1.62% 73 [vm_mmap+0 -> vm_mmap+48] [kernel.kallsyms]
0.00% 63 0.69% 31 [up_write+0 -> up_write+34] [kernel.kallsyms]
0.00% 13 0.29% 13 [setup_arg_pages+396 -> setup_arg_pages+413] [kernel.kallsyms]
0.00% 3 0.07% 3 [setup_arg_pages+418 -> setup_arg_pages+450] [kernel.kallsyms]
0.00% 616 6.84% 308 [security_mmap_file+0 -> security_mmap_file+72] [kernel.kallsyms]
0.00% 23 0.51% 23 [security_mmap_file+77 -> security_mmap_file+87] [kernel.kallsyms]
0.00% 4 0.02% 1 [sched_clock+0 -> sched_clock+4] [kernel.kallsyms]
0.00% 4 0.02% 1 [sched_clock+9 -> sched_clock+12] [kernel.kallsyms]
0.00% 1 0.02% 1 [rcu_nmi_exit+0 -> rcu_nmi_exit+9] [kernel.kallsyms]
Committer testing:
This should provide material for hours of endless joy, both from looking
for suspicious things in the implementation of this patch, such as the
top one:
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
As well from things that look legit:
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
0.16% 123.0K 0.60% 4.7K [nospec-branch.h:265 -> nospec-branch.h:278] [kernel.vmlinux]
:-)
Very short system wide taken branches session:
# perf record -h -b
Usage: perf record [<options>] [<command>]
or: perf record [<options>] -- <command> [<options>]
-b, --branch-any sample any taken branches
#
# perf record -b
^C[ perf record: Woken up 595 times to write data ]
[ perf record: Captured and wrote 156.672 MB perf.data (196873 samples) ]
#
# perf evlist -v
cycles: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CPU|PERIOD|BRANCH_STACK, read_format: ID, 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, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY
#
# perf report --total-cycles --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
# Total Lost Samples: 0
#
# Samples: 6M of event 'cycles'
# Event count (approx.): 6299936
#
# Sampled Cycles% Sampled Cycles Avg Cycles% Avg Cycles [Program Block Range] Shared Object
# ............... .............. ........... .......... ...................................................................... ....................
#
2.17% 1.7M 0.08% 607 [compiler.h:199 -> common.c:221] [kernel.vmlinux]
1.75% 1.3M 8.34% 65.5K [memset-vec-unaligned-erms.S:147 -> memset-vec-unaligned-erms.S:151] libc-2.29.so
0.72% 544.5K 0.03% 230 [entry_64.S:657 -> entry_64.S:662] [kernel.vmlinux]
0.56% 541.8K 0.09% 672 [compiler.h:199 -> common.c:300] [kernel.vmlinux]
0.39% 293.2K 0.01% 104 [list_debug.c:43 -> list_debug.c:61] [kernel.vmlinux]
0.36% 278.6K 0.03% 272 [entry_64.S:1289 -> entry_64.S:1308] [kernel.vmlinux]
0.30% 260.8K 0.07% 564 [clear_page_64.S:47 -> clear_page_64.S:50] [kernel.vmlinux]
0.28% 215.3K 0.05% 369 [traps.c:623 -> traps.c:628] [kernel.vmlinux]
0.23% 178.1K 0.04% 278 [entry_64.S:271 -> entry_64.S:275] [kernel.vmlinux]
0.20% 152.6K 0.09% 706 [paravirt.c:177 -> paravirt.c:179] [kernel.vmlinux]
0.20% 155.8K 0.05% 373 [entry_64.S:153 -> entry_64.S:175] [kernel.vmlinux]
0.18% 136.6K 0.03% 222 [msr.h:105 -> msr.h:166] [kernel.vmlinux]
0.16% 123.0K 0.60% 4.7K [nospec-branch.h:265 -> nospec-branch.h:278] [kernel.vmlinux]
0.16% 118.3K 0.01% 44 [entry_64.S:632 -> entry_64.S:657] [kernel.vmlinux]
0.14% 104.5K 0.00% 28 [rwsem.c:1541 -> rwsem.c:1544] [kernel.vmlinux]
0.13% 99.2K 0.01% 53 [spinlock.c:150 -> spinlock.c:152] [kernel.vmlinux]
0.13% 95.5K 0.00% 35 [swap.c:456 -> swap.c:471] [kernel.vmlinux]
0.12% 96.2K 0.05% 407 [copy_user_64.S:175 -> copy_user_64.S:209] [kernel.vmlinux]
0.11% 85.9K 0.00% 31 [swap.c:400 -> page-flags.h:188] [kernel.vmlinux]
0.10% 73.0K 0.01% 52 [paravirt.h:763 -> list.h:131] [kernel.vmlinux]
0.07% 56.2K 0.03% 214 [filemap.c:1524 -> filemap.c:1557] [kernel.vmlinux]
0.07% 54.2K 0.02% 145 [memory.c:1032 -> memory.c:1049] [kernel.vmlinux]
0.07% 50.3K 0.00% 39 [mmzone.c:49 -> mmzone.c:69] [kernel.vmlinux]
0.06% 48.3K 0.01% 40 [paravirt.h:768 -> page_alloc.c:3304] [kernel.vmlinux]
0.06% 46.7K 0.02% 155 [memory.c:1032 -> memory.c:1056] [kernel.vmlinux]
0.06% 46.9K 0.01% 103 [swap.c:867 -> swap.c:902] [kernel.vmlinux]
0.06% 47.8K 0.00% 34 [entry_64.S:1201 -> entry_64.S:1202] [kernel.vmlinux]
-----------------------------------------------------------
v7:
---
Use use_browser in report__browse_block_hists for supporting
stdio and potential tui mode.
v6:
---
Create report__browse_block_hists in block-info.c (codes are
moved from builtin-report.c). It's called from
perf_evlist__tty_browse_hists.
v5:
---
1. Move all block functions to block-info.c
2. Move the code of setting ms in block hist_entry to
other patch.
v4:
---
1. Use new option '--total-cycles' to replace
'-s total_cycles' in v3.
2. Move block info collection out of block info
printing.
v3:
---
1. Use common function block_info__process_sym to
process the blocks per symbol.
2. Remove the nasty hack for skipping calculation
of column length
3. Some minor cleanup
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
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: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20191107074719.26139-6-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Jin Yao
|
b65a7d372b |
perf hist: Support block formats with compare/sort/display
This patch provides helper routines to support new columns for block
info output.
The new columns are:
Sampled Cycles%
Sampled Cycles
Avg Cycles%
Avg Cycles
[Program Block Range]
Shared Object
v5:
---
1. Move more block related functions from builtin-report.c to
block-info.c
2. Set ms (map+sym) in block hist_entry. Because this info
is needed for reporting the block range (i.e. source line)
Committer notes:
Remove unused set_fmt() function, some build were not completing with:
util/block-info.c:396:20: error: unused function 'set_fmt' [-Werror,-Wunused-function]
static inline void set_fmt(struct block_fmt *block_fmt,
^
1 error generated.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20191107074719.26139-5-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Jin Yao
|
7841f40aed |
perf hist: Count the total cycles of all samples
We can get the per sample cycles by hist__account_cycles(). It's also useful to know the total cycles of all samples in order to get the cycles coverage for a single program block in further. For example: coverage = per block sampled cycles / total sampled cycles This patch creates a new argument 'total_cycles' in hist__account_cycles(), which will be added with the cycles of each sample. Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Reviewed-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jin Yao <yao.jin@intel.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lore.kernel.org/lkml/20191107074719.26139-4-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Jin Yao
|
6041441870 |
perf block: Cleanup and refactor block info functions
We have already implemented some block-info related functions.
Now it's time to do some cleanup, refactoring and move the
functions and structures to new block-info.h/block-info.c.
v4:
---
Move code for skipping column length calculation to patch:
'perf diff: Don't use hack to skip column length calculation'
v3:
---
1. Rename the patch title
2. Rename from block.h/block.c to block-info.h/block-info.c
3. Move more common part to block-info, such as
block_info__process_sym.
4. Remove the nasty hack for skipping calculation of column
length
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20191107074719.26139-3-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Jin Yao
|
0bdf181fe0 |
perf diff: Don't use hack to skip column length calculation
Previously we use a nasty hack to skip the hists__calc_col_len for block since this function is not very suitable for block column length calculation. This patch removes the hack code and add a check at the entry of hists__calc_col_len to skip for block case. Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Reviewed-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jin Yao <yao.jin@intel.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lore.kernel.org/lkml/20191107074719.26139-2-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
Leo Yan
|
af8490eb2b |
perf tests: Fix out of bounds memory access
The test case 'Read backward ring buffer' failed on 32-bit architectures
which were found by LKFT perf testing. The test failed on arm32 x15
device, qemu_arm32, qemu_i386, and found intermittent failure on i386;
the failure log is as below:
50: Read backward ring buffer :
--- start ---
test child forked, pid 510
Using CPUID GenuineIntel-6-9E-9
mmap size 1052672B
mmap size 8192B
Finished reading overwrite ring buffer: rewind
free(): invalid next size (fast)
test child interrupted
---- end ----
Read backward ring buffer: FAILED!
The log hints there have issue for memory usage, thus free() reports
error 'invalid next size' and directly exit for the case. Finally, this
issue is root caused as out of bounds memory access for the data array
'evsel->id'.
The backward ring buffer test invokes do_test() twice. 'evsel->id' is
allocated at the first call with the flow:
test__backward_ring_buffer()
`-> do_test()
`-> evlist__mmap()
`-> evlist__mmap_ex()
`-> perf_evsel__alloc_id()
So 'evsel->id' is allocated with one item, and it will be used in
function perf_evlist__id_add():
evsel->id[0] = id
evsel->ids = 1
At the second call for do_test(), it skips to initialize 'evsel->id'
and reuses the array which is allocated in the first call. But
'evsel->ids' contains the stale value. Thus:
evsel->id[1] = id -> out of bound access
evsel->ids = 2
To fix this issue, we will use evlist__open() and evlist__close() pair
functions to prepare and cleanup context for evlist; so 'evsel->id' and
'evsel->ids' can be initialized properly when invoke do_test() and avoid
the out of bounds memory access.
Fixes:
|
||
Jiwei Sun
|
6d57581659 |
perf record: Add support for limit perf output file size
The patch adds a new option to limit the output file size, then based on
it, we can create a wrapper of the perf command that uses the option to
avoid exhausting the disk space by the unconscious user.
In order to make the perf.data parsable, we just limit the sample data
size, since the perf.data consists of many headers and sample data and
other data, the actual size of the recorded file will bigger than the
setting value.
Testing it:
# ./perf record -a -g --max-size=10M
Couldn't synthesize bpf events.
[ perf record: perf size limit reached (10249 KB), stopping session ]
[ perf record: Woken up 32 times to write data ]
[ perf record: Captured and wrote 10.133 MB perf.data (71964 samples) ]
# ls -lh perf.data
-rw------- 1 root root 11M Oct 22 14:32 perf.data
# ./perf record -a -g --max-size=10K
[ perf record: perf size limit reached (10 KB), stopping session ]
Couldn't synthesize bpf events.
[ perf record: Woken up 0 times to write data ]
[ perf record: Captured and wrote 1.546 MB perf.data (69 samples) ]
# ls -l perf.data
-rw------- 1 root root 1626952 Oct 22 14:36 perf.data
Committer notes:
Fixed the build in multiple distros by using PRIu64 to print u64 struct
members, fixing this:
builtin-record.c: In function 'record__write':
builtin-record.c:150:5: error: format '%lu' expects argument of type 'long unsigned int', but argument 3 has type 'u64' [-Werror=format=]
rec->bytes_written >> 10);
^
CC /tmp/build/pe
Signed-off-by: Jiwei Sun <jiwei.sun@windriver.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Richard Danter <richard.danter@windriver.com>
Link: http://lore.kernel.org/lkml/20191022080901.3841-1-jiwei.sun@windriver.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Masami Hiramatsu
|
dee36a2abb |
perf probe: Skip overlapped location on searching variables
Since debuginfo__find_probes() callback function can be called with the location which already passed, the callback function must filter out such overlapped locations. add_probe_trace_event() has already done it by commit |
||
|
Masami Hiramatsu
|
86c0bf8539 |
perf probe: Fix to show calling lines of inlined functions
Fix to show calling lines of inlined functions (where an inline function
is called).
die_walk_lines() filtered out the lines inside inlined functions based
on the address. However this also filtered out the lines which call
those inlined functions from the target function.
To solve this issue, check the call_file and call_line attributes and do
not filter out if it matches to the line information.
Without this fix, perf probe -L doesn't show some lines correctly.
(don't see the lines after 17)
# perf probe -L vfs_read
<vfs_read@/home/mhiramat/ksrc/linux/fs/read_write.c:0>
0 ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
1 {
2 ssize_t ret;
4 if (!(file->f_mode & FMODE_READ))
return -EBADF;
6 if (!(file->f_mode & FMODE_CAN_READ))
return -EINVAL;
8 if (unlikely(!access_ok(buf, count)))
return -EFAULT;
11 ret = rw_verify_area(READ, file, pos, count);
12 if (!ret) {
13 if (count > MAX_RW_COUNT)
count = MAX_RW_COUNT;
15 ret = __vfs_read(file, buf, count, pos);
16 if (ret > 0) {
fsnotify_access(file);
add_rchar(current, ret);
}
With this fix:
# perf probe -L vfs_read
<vfs_read@/home/mhiramat/ksrc/linux/fs/read_write.c:0>
0 ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
1 {
2 ssize_t ret;
4 if (!(file->f_mode & FMODE_READ))
return -EBADF;
6 if (!(file->f_mode & FMODE_CAN_READ))
return -EINVAL;
8 if (unlikely(!access_ok(buf, count)))
return -EFAULT;
11 ret = rw_verify_area(READ, file, pos, count);
12 if (!ret) {
13 if (count > MAX_RW_COUNT)
count = MAX_RW_COUNT;
15 ret = __vfs_read(file, buf, count, pos);
16 if (ret > 0) {
17 fsnotify_access(file);
18 add_rchar(current, ret);
}
20 inc_syscr(current);
}
Fixes:
|