Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 160
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 26 9月, 2012 2 次提交
  3. 07 9月, 2012 1 次提交
  5. 15 8月, 2012 3 次提交
    • A
      perf evlist: Introduce evsel list accessors · 0c21f736
      由 Arnaldo Carvalho de Melo 提交于 
      To replace the longer list_entry constructs for things that are widely
used:

	perf_evlist__{first,last}(evlist)
	perf_evsel__next(evsel)
Acked-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NNamhyung Kim <namhyung@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-ng7azq26wg1jd801qqpcozwp@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      0c21f736
    • A
      perf evlist: Rename __group method to __set_leader · 63dab225
      由 Arnaldo Carvalho de Melo 提交于 
      Just like was done for parse_events__set_leader.

Also we need to have the list_entry set_leader method in evlist.c so that we
don't grow another dep in the python binding:

 # ~acme/git/linux/tools/perf/python/twatch.py
 Traceback (most recent call last):
   File "/home/acme/git/linux/tools/perf/python/twatch.py", line 16, in <module>
     import perf
 ImportError: /home/acme/git/build/perf/python/perf.so: undefined symbol: parse_events__set_leader

And also remove a pr_debug from evsel.c so that we avoid this one too:

 # ~acme/git/linux/tools/perf/python/twatch.py
 Traceback (most recent call last):
   File "/home/acme/git/linux/tools/perf/python/twatch.py", line 16, in <module>
     import perf
 ImportError: /home/acme/git/build/perf/python/perf.so: undefined symbol: eprintf
Acked-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NNamhyung Kim <namhyung@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-0hk9dazg9pora9jylkqngovm@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      63dab225
    • J
      perf tools: Enable grouping logic for parsed events · 6a4bb04c
      由 Jiri Olsa 提交于 
      This patch adds a functionality that allows to create event groups
based on the way they are specified on the command line. Adding
functionality to the '{}' group syntax introduced in earlier patch.

The current '--group/-g' option behaviour remains intact. If you
specify it for record/stat/top command, all the specified events
become members of a single group with the first event as a group
leader.

With the new '{}' group syntax you can create group like:
  # perf record -e '{cycles,faults}' ls

resulting in single event group containing 'cycles' and 'faults'
events, with cycles event as group leader.

All groups are created with regards to threads and cpus. Thus
recording an event group within a 2 threads on server with
4 CPUs will create 8 separate groups.

Examples (first event in brackets is group leader):

  # 1 group (cpu-clock,task-clock)
  perf record --group -e cpu-clock,task-clock ls
  perf record -e '{cpu-clock,task-clock}' ls

  # 2 groups (cpu-clock,task-clock) (minor-faults,major-faults)
  perf record -e '{cpu-clock,task-clock},{minor-faults,major-faults}' ls

  # 1 group (cpu-clock,task-clock,minor-faults,major-faults)
  perf record --group -e cpu-clock,task-clock -e minor-faults,major-faults ls
  perf record -e '{cpu-clock,task-clock,minor-faults,major-faults}' ls

  # 2 groups (cpu-clock,task-clock) (minor-faults,major-faults)
  perf record -e '{cpu-clock,task-clock} -e '{minor-faults,major-faults}' \
   -e instructions ls

  # 1 group
  # (cpu-clock,task-clock,minor-faults,major-faults,instructions)
  perf record --group -e cpu-clock,task-clock \
   -e minor-faults,major-faults -e instructions ls perf record -e
'{cpu-clock,task-clock,minor-faults,major-faults,instructions}' ls

It's possible to use standard event modifier for a group, which spans
over all events in the group and updates each event modifier settings,
for example:

  # perf record -r '{faults:k,cache-references}:p'

resulting in ':kp' modifier being used for 'faults' and ':p' modifier
being used for 'cache-references' event.
Reviewed-by: NNamhyung Kim <namhyung@kernel.org>
Signed-off-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ulrich Drepper <drepper@gmail.com>
Link: http://lkml.kernel.org/n/tip-ho42u0wcr8mn1otkalqi13qp@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      6a4bb04c
  7. 02 8月, 2012 2 次提交
  9. 25 7月, 2012 1 次提交
    • F
      perf tools: Fix trace events storms due to weight demux · 0983cc0d
      由 Frederic Weisbecker 提交于 
      Trace events have a period (weight) of 1 by default. This can be
overriden on events definition by using the __perf_count() macro.

For example, the sched_stat_runtime() is weighted with the runtime of
the task that fired the event.

By default, perf handles such weighted event by dividing it into
individual events carrying a weight of 1. For example if
sched_stat_runtime is fired and the task has run 5000000 nsecs, perf
divides it into 5000000 events in the buffer.

This behaviour makes weighted events unusable because they quickly
fullfill the buffers and we lose most events.

The commit 5d81e5cf ("events: Don't
divide events if it has field period") solves this problem by sending
only one event when PERF_SAMPLE_PERIOD flag is set. The weight is
carried in the sample itself such that we don't need to demultiplex it
anymore.

This patch provides the last missing piece to use this feature by
setting PERF_SAMPLE_PERIOD from perf tools when we deal with trace
events.

Before:
	$ ./perf record -e sched:* -a sleep 1
	[ perf record: Woken up 3 times to write data ]
	[ perf record: Captured and wrote 1.619 MB perf.data (~70749 samples) ]
	Warning:
	Processed 16909 events and lost 1 chunks!

	Check IO/CPU overload!

	$ ./perf script
	perf  1894 [003]   824.898327: sched_migrate_task: comm=perf pid=1898 prio=120 orig_cpu=2 dest_cpu=0
	perf  1894 [003]   824.898335: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	perf  1894 [003]   824.898336: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	perf  1894 [003]   824.898337: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	perf  1894 [003]   824.898338: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	perf  1894 [003]   824.898339: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	perf  1894 [003]   824.898340: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	perf  1894 [003]   824.898341: sched_stat_sleep: comm=perf pid=1898 delay=113179500 [ns]
	[...]

After:
	$ ./perf record -e sched:* -a sleep 1
	[ perf record: Woken up 1 times to write data ]
	[ perf record: Captured and wrote 0.074 MB perf.data (~3228 samples) ]

	$ ./perf script

	perf  1461 [000]   554.286957: sched_migrate_task: comm=perf pid=1465 prio=120 orig_cpu=3 dest_cpu=1
	perf  1461 [000]   554.286964: sched_stat_sleep: comm=perf pid=1465 delay=133047190 [ns]
	perf  1461 [000]   554.286967: sched_wakeup: comm=perf pid=1465 prio=120 success=1 target_cpu=001
	swapper     0 [001]   554.286976: sched_stat_wait: comm=perf pid=1465 delay=0 [ns]
	swapper     0 [001]   554.286983: sched_switch: prev_comm=swapper/1 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=perf
	[...]
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1342631456-7233-1-git-send-email-fweisbec@gmail.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      0983cc0d
  11. 28 6月, 2012 1 次提交
    • A
      perf tools: Stop using a global trace events description list · da378962
      由 Arnaldo Carvalho de Melo 提交于 
      The pevent thing is per perf.data file, so I made it stop being static
and become a perf_session member, so tools processing perf.data files
use perf_session and _there_ we read the trace events description into
session->pevent and then change everywhere to stop using that single
global pevent variable and use the per session one.

Note that it _doesn't_ fall backs to trace__event_id, as we're not
interested at all in what is present in the
/sys/kernel/debug/tracing/events in the workstation doing the analysis,
just in what is in the perf.data file.

This patch also introduces perf_session__set_tracepoints_handlers that
is the perf perf.data/session way to associate handlers to tracepoint
events by resolving their IDs using the events descriptions stored in a
perf.data file. Make 'perf sched' use it.
Reported-by: NDmitry Antipov <dmitry.antipov@linaro.org>
Tested-by: NDmitry Antipov <dmitry.antipov@linaro.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linaro-dev@lists.linaro.org
Cc: patches@linaro.org
Link: http://lkml.kernel.org/r/20120625232016.GA28525@infradead.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      da378962
  13. 31 5月, 2012 2 次提交
    • N
      perf evlist: Pass third argument to ioctl explicitly · 55da8005
      由 Namhyung Kim 提交于 
      The ioctl on perf event fd wants 3 arguments but we only passed 2. As
the only user of the functions is perf record and it calls them for
every event (regardless of group setting), just pass 0 for now.
Signed-off-by: NNamhyung Kim <namhyung.kim@lge.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338443506-25009-3-git-send-email-namhyung.kim@lge.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      55da8005
    • A
      perf stat: Initialize default events wrt exclude_{guest,host} · 79695e1b
      由 Arnaldo Carvalho de Melo 提交于 
      When no event is specified the tools use perf_evlist__add_default(), that will
call event_attr_init to initialize the KVM exclusion bits.

When the change was made to the tools so that by default guest samples would be
excluded, the changes were made just to the parsing routines and to
perf_evlist__add_default(), not to perf_evlist__add_attrs, that is used so far
just by perf stat to add multiple events, according to the level of detail
specified.

Recently the tools were changed to reconstruct the event name from all the
details in perf_event_attr, not just from .type and .config, but taking into
account all the feature bits (.exclude_{guest,host,user,kernel,etc},
.precise_ip, etc).

That is when we noticed that the default for perf stat wasn't the one for the
rest of the tools, i.e. the .exclude_guest bit wasn't being set.

I.e. the default, that doesn't call event_attr_init was showing the :HG
modifier:

  $ perf stat usleep 1

   Performance counter stats for 'usleep 1':

            0.942119 task-clock                #    0.454 CPUs utilized
                   1 context-switches          #    0.001 M/sec
                   0 CPU-migrations            #    0.000 K/sec
                 126 page-faults               #    0.134 M/sec
             693,193 cycles:HG                 #    0.736 GHz                     [40.11%]
             407,461 stalled-cycles-frontend:HG #   58.78% frontend cycles idle    [72.29%]
             365,403 stalled-cycles-backend:HG #   52.71% backend  cycles idle
             465,982 instructions:HG           #    0.67  insns per cycle
                                               #    0.87  stalled cycles per insn
              89,760 branches:HG               #   95.275 M/sec
               6,178 branch-misses:HG          #    6.88% of all branches

         0.002077228 seconds time elapsed

While if one explicitely specifies the same events, which will make the parsing code
to be called and thus event_attr_init is called:

  $ perf stat -e task-clock,context-switches,migrations,page-faults,cycles,stalled-cycles-frontend,stalled-cycles-backend,instructions,branches,branch-misses usleep 1

   Performance counter stats for 'usleep 1':

            1.040349 task-clock                #    0.500 CPUs utilized
                   2 context-switches          #    0.002 M/sec
                   0 CPU-migrations            #    0.000 K/sec
                 127 page-faults               #    0.122 M/sec
             587,966 cycles                    #    0.565 GHz                     [13.18%]
             459,167 stalled-cycles-frontend   #   78.09% frontend cycles idle
             390,249 stalled-cycles-backend    #   66.37% backend  cycles idle
             504,006 instructions              #    0.86  insns per cycle
                                               #    0.91  stalled cycles per insn
              96,455 branches                  #   92.714 M/sec
               6,522 branch-misses             #    6.76% of all branches         [96.12%]

         0.002078681 seconds time elapsed

Fix it by introducing a perf_evlist__add_default_attrs method that will call
evlist_attr_init in all the perf_event_attr entries before adding the events.
Reported-by: NIngo Molnar <mingo@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-4eysr236r0pgiyum9epwxw7s@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      79695e1b
  15. 17 5月, 2012 1 次提交
  17. 16 5月, 2012 2 次提交
  19. 08 5月, 2012 2 次提交
  21. 03 5月, 2012 3 次提交
  23. 17 3月, 2012 1 次提交
  25. 05 3月, 2012 1 次提交
  27. 01 3月, 2012 1 次提交
  29. 14 2月, 2012 1 次提交
  31. 02 2月, 2012 1 次提交
  33. 25 1月, 2012 1 次提交
  35. 07 1月, 2012 1 次提交
  37. 24 12月, 2011 1 次提交
  39. 20 12月, 2011 1 次提交
  41. 29 11月, 2011 2 次提交
    • A
      perf evlist: Always do automatic allocation of pollfd and mmap structures · 806fb630
      由 Arnaldo Carvalho de Melo 提交于 
      At first tools were required to do that, but while writing the python
bindings to simplify the API I made them auto-allocate when needed.

This just makes record, stat and top use that auto allocation,
simplifying them a bit.

Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-iokhcvkzzijr3keioubx8hlq@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      806fb630
    • A
      perf tools: Save some loops using perf_evlist__id2evsel · ee29be62
      由 Arnaldo Carvalho de Melo 提交于 
      Since we already ask for PERF_SAMPLE_ID and use it to quickly find the
associated evsel, add handler func + data to struct perf_evsel to avoid
using chains of if(strcmp(event_name)) and also to avoid all the linear
list searches via trace_event_find.

To demonstrate the technique convert 'perf sched' to it:

 # perf sched record sleep 5m

And then:

 Performance counter stats for '/tmp/oldperf sched lat':

        646.929438 task-clock                #    0.999 CPUs utilized
                 9 context-switches          #    0.000 M/sec
                 0 CPU-migrations            #    0.000 M/sec
            20,901 page-faults               #    0.032 M/sec
     1,290,144,450 cycles                    #    1.994 GHz
   <not supported> stalled-cycles-frontend
   <not supported> stalled-cycles-backend
     1,606,158,439 instructions              #    1.24  insns per cycle
       339,088,395 branches                  #  524.151 M/sec
         4,550,735 branch-misses             #    1.34% of all branches

       0.647524759 seconds time elapsed

Versus:

 Performance counter stats for 'perf sched lat':

        473.564691 task-clock                #    0.999 CPUs utilized
                 9 context-switches          #    0.000 M/sec
                 0 CPU-migrations            #    0.000 M/sec
            20,903 page-faults               #    0.044 M/sec
       944,367,984 cycles                    #    1.994 GHz
   <not supported> stalled-cycles-frontend
   <not supported> stalled-cycles-backend
     1,442,385,571 instructions              #    1.53  insns per cycle
       308,383,106 branches                  #  651.195 M/sec
         4,481,784 branch-misses             #    1.45% of all branches

       0.474215751 seconds time elapsed

[root@emilia ~]#

Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-1kbzpl74lwi6lavpqke2u2p3@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      ee29be62
  43. 28 11月, 2011 6 次提交
  45. 26 10月, 2011 1 次提交
  47. 07 10月, 2011 1 次提交
  49. 24 9月, 2011 1 次提交