linux/kernel
Steven Rostedt b2a866f934 ftrace: function tracer with irqs disabled
Impact: disable interrupts during trace entry creation (as opposed to preempt)

To help with performance, I set the ftracer to not disable interrupts,
and only to disable preemption. If an interrupt occurred, it would not
be traced, because the function tracer protects itself from recursion.
This may be faster, but the trace output might miss some traces.

This patch makes the fuction trace disable interrupts, but it also
adds a runtime feature to disable preemption instead. It does this by
having two different tracer functions. When the function tracer is
enabled, it will check to see which version is requested (irqs disabled
or preemption disabled). Then it will use the corresponding function
as the tracer.

Irq disabling is the default behaviour, but if the user wants better
performance, with the chance of missing traces, then they can choose
the preempt disabled version.

Running hackbench 3 times with the irqs disabled and 3 times with
the preempt disabled function tracer yielded:

tracing type       times            entries recorded
------------      --------          ----------------
irq disabled      43.393            166433066
                  43.282            166172618
                  43.298            166256704

preempt disabled  38.969            159871710
                  38.943            159972935
                  39.325            161056510

Average:

   irqs disabled:  43.324           166287462
preempt disabled:  39.079           160300385

 preempt is 10.8 percent faster than irqs disabled.

I wrote a patch to count function trace recursion and reran hackbench.

With irq disabled: 1,150 times the function tracer did not trace due to
  recursion.
with preempt disabled: 5,117,718 times.

The thousand times with irq disabled could be due to NMIs, or simply a case
where it called a function that was not protected by notrace.

But we also see that a large amount of the trace is lost with the
preempt version.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-11-04 10:09:50 +01:00
..
irq irq: make variable static 2008-10-22 07:37:17 +02:00
power PM_TEST_SUSPEND should depend on RTC_CLASS, not RTC_LIB 2008-11-01 12:40:38 -07:00
time Merge branch 'timers/range-hrtimers' into v28-range-hrtimers-for-linus-v2 2008-10-22 09:48:06 +02:00
trace ftrace: function tracer with irqs disabled 2008-11-04 10:09:50 +01:00
.gitignore
acct.c tty: Fix abusers of current->sighand->tty 2008-10-13 09:51:42 -07:00
audit_tree.c [PATCH] get rid of nameidata in audit_tree 2008-10-23 05:12:53 -04:00
audit.c [PATCH] Fix the bug of using AUDIT_STATUS_RATE_LIMIT when set fail, no error output. 2008-08-01 12:15:16 -04:00
audit.h
auditfilter.c Re: [PATCH] the loginuid field should be output in all AUDIT_CONFIG_CHANGE audit messages 2008-08-01 12:15:03 -04:00
auditsc.c tty: Fix abusers of current->sighand->tty 2008-10-13 09:51:42 -07:00
backtracetest.c backtrace: replace timer with tasklet + completions 2008-06-27 18:09:16 +02:00
bounds.c
capability.c security: Fix setting of PF_SUPERPRIV by __capable() 2008-08-14 22:59:43 +10:00
cgroup_debug.c cgroups: fix probable race with put_css_set[_taskexit] and find_css_set 2008-10-20 08:52:38 -07:00
cgroup_freezer.c freezer_cg: simplify freezer_change_state() 2008-10-30 11:38:45 -07:00
cgroup.c cgroup: remove unused variable 2008-10-26 09:38:17 -07:00
compat.c Merge branches 'timers/clocksource', 'timers/hrtimers', 'timers/nohz', 'timers/ntp', 'timers/posixtimers' and 'timers/debug' into v28-timers-for-linus 2008-10-20 13:14:06 +02:00
configs.c kernel/configs.c: remove useless comments 2008-10-20 08:52:34 -07:00
cpu.c Merge branches 'sched/devel', 'sched/cpu-hotplug', 'sched/cpusets' and 'sched/urgent' into sched/core 2008-10-08 11:31:02 +02:00
cpuset.c cpuset: use seq_*mask_* to print masks 2008-10-20 08:52:39 -07:00
delayacct.c per-task-delay-accounting: update taskstats for memory reclaim delay 2008-07-25 10:53:47 -07:00
dma-coherent.c dma-coherent: export dma_[alloc|release]_from_coherent methods 2008-08-22 08:34:53 +02:00
dma.c kernel/dma.c: remove a CVS keyword 2008-10-16 11:21:30 -07:00
exec_domain.c proc: move /proc/execdomains to kernel/exec_domain.c 2008-10-23 14:30:41 +04:00
exit.c Merge branch 'tracing-v28-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2008-10-20 13:35:07 -07:00
extable.c
fork.c Merge branch 'timers/range-hrtimers' into v28-range-hrtimers-for-linus-v2 2008-10-22 09:48:06 +02:00
freezer.c freezer_cg: use thaw_process() in unfreeze_cgroup() 2008-10-30 11:38:45 -07:00
futex_compat.c
futex.c hrtimer: make the futex() system call use the per process slack value 2008-09-11 07:17:00 -07:00
hrtimer.c Merge branch 'timers/range-hrtimers' into v28-range-hrtimers-for-linus-v2 2008-10-22 09:48:06 +02:00
itimer.c timers: fix itimer/many thread hang 2008-09-14 16:25:35 +02:00
kallsyms.c kernel/kallsyms.c: fix double return 2008-10-16 11:21:32 -07:00
Kconfig.freezer container freezer: implement freezer cgroup subsystem 2008-10-20 08:52:34 -07:00
Kconfig.hz sched: fix SCHED_HRTICK dependency 2008-07-28 14:37:38 +02:00
Kconfig.preempt
kexec.c kexec: fix crash_save_vmcoreinfo_init build problem 2008-10-20 15:28:50 -07:00
kfifo.c
kgdb.c kgdb: call touch_softlockup_watchdog on resume 2008-10-06 13:50:59 -05:00
kmod.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus 2008-10-16 12:38:34 -07:00
kprobes.c make kprobes.c:kretprobe_table_lock() static 2008-10-16 11:21:52 -07:00
ksysfs.c profiling: dynamically enable readprofile at runtime 2008-10-16 11:21:31 -07:00
kthread.c Merge branch 'tracing-v28-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2008-10-20 13:35:07 -07:00
latencytop.c
lockdep_internals.h lockdep: build fix 2008-08-13 12:55:10 +02:00
lockdep_proc.c lockstat: fix numerical output rounding error 2008-08-26 10:37:46 +02:00
lockdep.c lockdep: fix irqs on/off ip tracing 2008-10-28 11:19:07 +01:00
Makefile Merge branch 'tracing/ftrace' into tracing/urgent 2008-10-22 09:08:14 +02:00
marker.c markers: break the redundant loop in kernel/marker.c 2008-10-27 13:02:23 +01:00
module.c Merge branch 'proc' of git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/proc 2008-10-23 12:04:37 -07:00
mutex-debug.c mutex-debug: check mutex magic before owner 2008-05-16 16:53:35 +02:00
mutex-debug.h
mutex.c locking: fix mutex @key parameter kernel-doc notation 2008-07-28 18:12:36 +02:00
mutex.h
notifier.c ftrace: ignore functions that cannot be kprobe-ed 2008-10-14 10:34:22 +02:00
ns_cgroup.c cgroup_clone: use pid of newly created task for new cgroup 2008-07-25 10:53:37 -07:00
nsproxy.c removed unused #include <linux/version.h>'s 2008-08-23 12:14:12 -07:00
panic.c Make panic= and panic_on_oops into core_params 2008-10-22 10:00:25 +11:00
params.c Fix compile warning in kernel/params.c 2008-10-23 12:09:00 -07:00
pid_namespace.c pid_ns: (BUG 11391) change ->child_reaper when init->group_leader exits 2008-09-02 19:21:38 -07:00
pid.c pidns: remove now unused find_pid function. 2008-07-25 10:53:45 -07:00
pm_qos_params.c pm_qos_requirement might sleep 2008-09-02 19:21:40 -07:00
posix-cpu-timers.c timers: fix itimer/many thread hang, v2 2008-09-23 13:38:44 +02:00
posix-timers.c Merge branch 'timers/range-hrtimers' into v28-range-hrtimers-for-linus-v2 2008-10-22 09:48:06 +02:00
printk.c printk: remove unused code from kernel/printk.c 2008-10-23 21:54:29 +02:00
profile.c kernel/profile: fix profile_init() section mismatch 2008-10-30 11:38:46 -07:00
ptrace.c make ptrace_untrace() static 2008-10-20 08:52:39 -07:00
rcuclassic.c rcu: RCU-based detection of stalled CPUs for Classic RCU, fix 2008-10-03 10:41:00 +02:00
rcupdate.c rcupdate: fix bug of rcu_barrier*() 2008-10-21 15:59:53 +02:00
rcupreempt_trace.c rcu: trace fix possible mem-leak 2008-08-15 17:54:40 +02:00
rcupreempt.c byteorder: remove direct includes of linux/byteorder/swab[b].h 2008-10-20 08:52:40 -07:00
rcutorture.c byteorder: remove direct includes of linux/byteorder/swab[b].h 2008-10-20 12:51:53 -07:00
relay.c relay: fix "full buffer with exactly full last subbuffer" accounting problem 2008-08-05 14:33:46 -07:00
res_counter.c cgroup files: convert res_counter_write() to be a cgroups write_string() handler 2008-07-25 10:53:36 -07:00
resource.c reserve_region_with_split: Fix GFP_KERNEL usage under spinlock 2008-11-01 09:53:58 -07:00
rtmutex_common.h
rtmutex-debug.c
rtmutex-debug.h
rtmutex-tester.c sysdev: Pass the attribute to the low level sysdev show/store function 2008-07-21 21:55:02 -07:00
rtmutex.c hrtimer: convert kernel/* to the new hrtimer apis 2008-09-05 21:35:13 -07:00
rtmutex.h
rwsem.c
sched_clock.c sched_clock: prevent scd->clock from moving backwards 2008-10-10 11:17:04 +02:00
sched_cpupri.c sched: use a 2-d bitmap for searching lowest-pri CPU 2008-06-06 15:19:28 +02:00
sched_cpupri.h sched: fix the cpuprio count really 2008-06-06 15:19:44 +02:00
sched_debug.c sched: change sched_debug's mode to 0444 2008-10-30 11:37:57 +01:00
sched_fair.c sched: virtual time buddy preemption 2008-10-24 12:51:03 +02:00
sched_features.h sched: disable the hrtick for now 2008-10-20 14:27:43 +02:00
sched_idletask.c sched: add CONFIG_SMP consistency 2008-10-22 10:01:52 +02:00
sched_rt.c Merge commit 'v2.6.28-rc1' into sched/urgent 2008-10-24 12:48:46 +02:00
sched_stats.h Merge branch 'proc' of git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/proc 2008-10-23 12:04:37 -07:00
sched.c sched: virtual time buddy preemption 2008-10-24 12:51:03 +02:00
seccomp.c
semaphore.c semaphore: __down_common: use signal_pending_state() 2008-08-05 14:33:47 -07:00
signal.c 'kill sig -1' must only apply to caller's namespace 2008-10-30 11:38:46 -07:00
smp.c smp: have smp_call_function_single() detect invalid CPUs 2008-08-25 17:45:48 -07:00
softirq.c Merge branches 'timers/clocksource', 'timers/hrtimers', 'timers/nohz', 'timers/ntp', 'timers/posixtimers' and 'timers/debug' into v28-timers-for-linus 2008-10-20 13:14:06 +02:00
softlockup.c Make the taint flags reliable 2008-10-16 11:21:31 -07:00
spinlock.c lockdep: spin_lock_nest_lock(), checkpatch fixes 2008-08-13 13:56:51 +02:00
srcu.c
stacktrace.c stacktrace: fix modular build, export print_stack_trace and save_stack_trace 2008-06-30 09:20:55 +02:00
stop_machine.c Revert "Call init_workqueues before pre smp initcalls." 2008-10-25 19:53:38 -07:00
sys_ni.c Configure out AIO support 2008-10-16 11:21:51 -07:00
sys.c Merge branch 'timers/range-hrtimers' into v28-range-hrtimers-for-linus-v2 2008-10-22 09:48:06 +02:00
sysctl_check.c sysctl: check for bogus modes 2008-07-25 10:53:45 -07:00
sysctl.c Merge branch 'linus' into tracing/ftrace 2008-10-31 00:38:21 +01:00
taskstats.c taskstats: remove initialization of static per-cpu variable 2008-07-25 10:53:47 -07:00
test_kprobes.c
time.c select: add a timespec_add_safe() function 2008-09-05 21:34:57 -07:00
timeconst.pl Make constants in kernel/timeconst.h fixed 64 bits 2008-05-02 16:18:42 -07:00
timer.c Merge branches 'timers/clocksource', 'timers/hrtimers', 'timers/nohz', 'timers/ntp', 'timers/posixtimers' and 'timers/debug' into v28-timers-for-linus 2008-10-20 13:14:06 +02:00
tracepoint.c tracepoint: introduce *_noupdate APIs. 2008-11-03 10:28:52 +01:00
tsacct.c task IO accounting: move all IO statistics in struct task_io_accounting 2008-07-27 16:12:28 -07:00
uid16.c
user_namespace.c removed unused #include <linux/version.h>'s 2008-08-23 12:14:12 -07:00
user.c sched: rt-bandwidth for user grouping interface 2008-08-19 13:10:09 +02:00
utsname_sysctl.c sysctl: simplify ->strategy 2008-10-16 11:21:47 -07:00
utsname.c removed unused #include <linux/version.h>'s 2008-08-23 12:14:12 -07:00
wait.c wait: kill is_sync_wait() 2008-10-16 11:21:31 -07:00
workqueue.c workqueue: introduce create_rt_workqueue 2008-10-22 10:00:25 +11:00