lkml.org 
[lkml]   [2008]   [Sep]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [RFC PATCH 1/3] Unified trace buffer

* Ingo Molnar <mingo@elte.hu> wrote:

> * Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> > On Thu, 25 Sep 2008, Linus Torvalds wrote:
> > >
> > > You also ignore the early trace issues, and have apparently not used
> > > it for FTRACE. You also ignore the fact that without TSC, it goes
> > > into the same "crap mode" that is appropriate for the scheduler, but
> > > totally useless for tracing.
> >
> > Oh, and I didn't notice (because Steven pointed out "notrace" and I
> > didn't see any of them), that in order to get things to work you had
> > just added
> >
> > CFLAGS_REMOVE_lockdep.o = -pg
> > CFLAGS_REMOVE_lockdep_proc.o = -pg
> > CFLAGS_REMOVE_mutex-debug.o = -pg
> > CFLAGS_REMOVE_rtmutex-debug.o = -pg
> > CFLAGS_REMOVE_cgroup-debug.o = -pg
> > CFLAGS_REMOVE_sched_clock.o = -pg
> > CFLAGS_REMOVE_sched.o = -mno-spe -pg
> >
> > all ovr the place, which was part of my argument against this crap in
> > the first place.
>
> Really, we traced all these files for ages. I can restore it if it's
> worthwile - but lockdep totally kills the usability of function
> traces, it inserts thousands of uninteresting events over and over
> again.

to prove it, i just applied this patch:

Index: linux/kernel/Makefile
===================================================================
--- linux.orig/kernel/Makefile
+++ linux/kernel/Makefile
@@ -21,7 +21,6 @@ CFLAGS_REMOVE_mutex-debug.o = -pg
CFLAGS_REMOVE_rtmutex-debug.o = -pg
CFLAGS_REMOVE_cgroup-debug.o = -pg
CFLAGS_REMOVE_sched_clock.o = -pg
-CFLAGS_REMOVE_sched.o = -mno-spe -pg
endif

obj-$(CONFIG_PROFILING) += profile.o
and sched.o was fully traced again. For example schedule() to idle is 38
function calls:

$ cd /debug/tracing
$ echo ftrace > current_tracer
$ cat trace
# tracer: ftrace
#
# TASK-PID CPU# TIMESTAMP FUNCTION
# | | | | |
[...]
ssh-2734 [001] 52.291772: schedule <-schedule_timeout
ssh-2734 [001] 52.291772: hrtick_clear <-schedule
ssh-2734 [001] 52.291774: _spin_lock <-schedule
ssh-2734 [001] 52.291775: deactivate_task <-schedule
ssh-2734 [001] 52.291776: dequeue_task <-deactivate_task
ssh-2734 [001] 52.291777: dequeue_task_fair <-dequeue_task
ssh-2734 [001] 52.291778: update_curr <-dequeue_task_fair
ssh-2734 [001] 52.291779: calc_delta_mine <-update_curr
ssh-2734 [001] 52.291780: hrtick_start_fair <-dequeue_task_fair
ssh-2734 [001] 52.291782: find_busiest_group <-schedule
ssh-2734 [001] 52.291783: idle_cpu <-find_busiest_group
ssh-2734 [001] 52.291784: target_load <-find_busiest_group
ssh-2734 [001] 52.291785: weighted_cpuload <-target_load
ssh-2734 [001] 52.291786: weighted_cpuload <-find_busiest_group
ssh-2734 [001] 52.291787: cpu_avg_load_per_task <-find_busiest_group
ssh-2734 [001] 52.291788: source_load <-find_busiest_group
ssh-2734 [001] 52.291789: weighted_cpuload <-source_load
ssh-2734 [001] 52.291790: weighted_cpuload <-find_busiest_group
ssh-2734 [001] 52.291791: cpu_avg_load_per_task <-find_busiest_group
ssh-2734 [001] 52.291792: msecs_to_jiffies <-schedule
ssh-2734 [001] 52.291792: msecs_to_jiffies <-schedule
ssh-2734 [001] 52.291793: put_prev_task_fair <-schedule
ssh-2734 [001] 52.291795: pick_next_task_fair <-schedule
ssh-2734 [001] 52.291796: pick_next_task_rt <-schedule
ssh-2734 [001] 52.291796: pick_next_task_fair <-schedule
ssh-2734 [001] 52.291797: pick_next_task_idle <-schedule
ssh-2734 [001] 52.291798: _spin_trylock <-tracing_record_cmdline
ssh-2734 [001] 52.291800: _spin_unlock <-tracing_record_cmdline
ssh-2734 [001] 52.291802: __switch_to <-thread_return
<idle>-0 [001] 52.291804: finish_task_switch <-thread_return
<idle>-0 [001] 52.291805: _spin_unlock_irq <-finish_task_switch
<idle>-0 [001] 52.291806: tick_nohz_stop_sched_tick <-cpu_idle
<idle>-0 [001] 52.291807: ktime_get <-tick_nohz_stop_sched_tick
<idle>-0 [001] 52.291807: ktime_get_ts <-ktime_get
<idle>-0 [001] 52.291808: getnstimeofday <-ktime_get_ts
<idle>-0 [001] 52.291808: acpi_pm_read <-getnstimeofday
<idle>-0 [001] 52.291810: set_normalized_timespec <-ktime_get_ts
<idle>-0 [001] 52.291811: get_next_timer_interrupt <-tick_nohz_stop_sched_tick
it worked just fine in my first blind attempt.

Ingo


\
 
 \ /
  Last update: 2008-09-25 23:45    [W:0.564 / U:0.132 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site