linux-stable.git/tools/perf/Documentation, branch v4.2.3

perf stat: Introduce --per-thread option

2015-06-26T15:05:04+00:00

Currently all the -p option PID arguments tasks values get aggregated
and printed as single values.

Adding --per-tasks option to print values per task.

  $ perf stat  -e cycles,instructions --per-thread -p 30190,30242
  ^C
   Performance counter stats for process id '30190,30242':

               cat-30190                     0      cycles
               yes-30242         3,842,525,421      cycles
               cat-30190                     0      instructions
               yes-30242        10,370,817,010      instructions

         1.143155657 seconds time elapsed

Also works under interval mode:

  $ perf stat  -e cycles,instructions --per-thread -p 30190,30242 -I 1000
  #           time             comm-pid                  counts unit events
       1.000073435              cat-30190                89,058      cycles
       1.000073435              yes-30242         3,360,786,902      cycles                     (100.00%)
       1.000073435              cat-30190                14,066      instructions
       1.000073435              yes-30242         9,069,937,462      instructions
       2.000204830              cat-30190                     0      cycles
       2.000204830              yes-30242         3,351,667,626      cycles
       2.000204830              cat-30190                     0      instructions
       2.000204830              yes-30242         9,045,796,885      instructions
  ^C     2.771286639              cat-30190                     0      cycles
       2.771286639              yes-30242         2,593,884,166      cycles
       2.771286639              cat-30190                     0      instructions
       2.771286639              yes-30242         7,001,171,191      instructions

It works only with -t and -p options, otherwise following error is
printed:

  $ perf stat  -e cycles --per-thread  -I 1000 ls
  The --per-thread option is only available when monitoring via -p -t options.
      -p, --pid        stat events on existing process id
      -t, --tid        stat events on existing thread id

Signed-off-by: Jiri Olsa 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Adrian Hunter 
Cc: Andi Kleen 
Cc: David Ahern 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1435310967-14570-23-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Configurable per thread proc map processing time out

2015-06-19T21:27:13+00:00

The time out to limit the individual proc map processing was hard code
to 500ms. This patch introduce a new option --proc-map-timeout to make
the time limit configurable.

Signed-off-by: Kan Liang 
Cc: Andi Kleen 
Cc: David Ahern 
Cc: Ying Huang 
Link: http://lkml.kernel.org/r/1434549071-25611-2-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf record: Amend option summaries

2015-06-10T15:00:27+00:00

Because there's too many options and I cannot read, I frequently get
confused between -c and -P, and try to do things like:

  perf record -P 50000 -- foo

Which does not work; try and make the option description slightly longer
and hopefully less confusing.

Signed-off-by: Peter Zijlstra (Intel) 
Link: http://lkml.kernel.org/r/20150610144850.GP19282@twins.programming.kicks-ass.net
[ Do those changes on the man page as well ]
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Document relation of per-thread event count feature

2015-05-12T12:59:48+00:00

The 'perf record -s' and 'perf report -T' should be used together to see
per-thread event counts.  Document the relation of these commands.

Signed-off-by: Namhyung Kim 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1431184784-30525-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf probe: Add --no-inlines option to avoid searching inline functions

2015-05-08T19:26:44+00:00

Add --no-inlines(--inlines) option to avoid searching inline functions.

Searching all functions which matches glob pattern can take a long time
and find a lot of inline functions.

With this option perf-probe searches target on the non-inlined
functions.

Signed-off-by: Masami Hiramatsu 
Cc: Ananth N Mavinakayanahalli 
Cc: David Ahern 
Cc: Hemant Kumar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20150508010333.24812.86568.stgit@localhost.localdomain
Signed-off-by: Arnaldo Carvalho de Melo

perf bench futex: Support parallel waker threads

2015-05-08T19:23:50+00:00

The futex-wake benchmark only measures wakeups done within a single
process. While this has value in its own, it does not really generate
any hb->lock contention.

A new benchmark 'wake-parallel' is added, by extending the futex-wake
code such that we can measure parallel waker threads. The program output
shows the avg per-thread latency in order to complete its share of
wakeups:

Run summary [PID 13474]: blocking on 512 threads (at [private] futex 0xa88668), 8 threads waking up 64 at a time.

[Run 1]: Avg per-thread latency (waking 64/512 threads) in 0.6230 ms (+-15.31%)
[Run 2]: Avg per-thread latency (waking 64/512 threads) in 0.5175 ms (+-29.95%)
[Run 3]: Avg per-thread latency (waking 64/512 threads) in 0.7578 ms (+-18.03%)
[Run 4]: Avg per-thread latency (waking 64/512 threads) in 0.8944 ms (+-12.54%)
[Run 5]: Avg per-thread latency (waking 64/512 threads) in 1.1204 ms (+-23.85%)
Avg per-thread latency (waking 64/512 threads) in 0.7826 ms (+-9.91%)

Naturally, different combinations of numbers of blocking and waker
threads will exhibit different information.

Signed-off-by: Davidlohr Bueso 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Davidlohr Bueso 
Link: http://lkml.kernel.org/r/1431110280-20231-1-git-send-email-dave@stgolabs.net
Signed-off-by: Arnaldo Carvalho de Melo

perf probe: Support $params special probe argument

2015-05-08T19:05:03+00:00

$params is similar to $vars but matches only function parameters not
local variables.

Thus, this is useful for tracing function parameter changing or tracing
function call with parameters.

Testing it:

 # perf probe tcp_sendmsg '$params'
 Added new event:
  probe:tcp_sendmsg    (on tcp_sendmsg with $params)

 You can now use it in all perf tools, such as:

	perf record -e probe:tcp_sendmsg -aR sleep 1

 # perf probe -l
  probe:tcp_sendmsg    (on tcp_sendmsg@acme/git/linux/net/ipv4/tcp.c with iocb sk msg size)
 # perf record -a -e probe:*
 press some random letters to generate TCP (sshd) traffic...

 ^C[ perf record: Woken up 1 times to write data ]
 [ perf record: Captured and wrote 0.223 MB perf.data (6 samples) ]

 # perf script
   sshd 6385 [2] 3.907529: probe:tcp_sendmsg: iocb=0xffff8800ac4cfe70 sk=0xffff88042196c140 msg=0xffff8800ac4cfda8 size=0x24
   sshd 6385 [2] 4.138973: probe:tcp_sendmsg: iocb=0xffff8800ac4cfe70 sk=0xffff88042196c140 msg=0xffff8800ac4cfda8 size=0x24
   sshd 6385 [2] 4.378966: probe:tcp_sendmsg: iocb=0xffff8800ac4cfe70 sk=0xffff88042196c140 msg=0xffff8800ac4cfda8 size=0x24
   sshd 6385 [2] 4.603681: probe:tcp_sendmsg: iocb=0xffff8800ac4cfe70 sk=0xffff88042196c140 msg=0xffff8800ac4cfda8 size=0x24
   sshd 6385 [2] 4.818455: probe:tcp_sendmsg: iocb=0xffff8800ac4cfe70 sk=0xffff88042196c140 msg=0xffff8800ac4cfda8 size=0x24
   sshd 6385 [2] 5.043603: probe:tcp_sendmsg: iocb=0xffff8800ac4cfe70 sk=0xffff88042196c140 msg=0xffff8800ac4cfda8 size=0x24
 # cat /sys/kernel/debug/tracing/events/probe/tcp_sendmsg/format
 name: tcp_sendmsg
 ID: 1927
 format:
   field:unsigned short common_type;	offset:0;	size:2;	signed:0;
   field:unsigned char common_flags;	offset:2;	size:1;	signed:0;
   field:unsigned char common_preempt_count;	offset:3;	size:1;	signed:0;
   field:int common_pid;	offset:4;	size:4;	signed:1;

   field:unsigned long __probe_ip;	offset:8;	size:8;	signed:0;
   field:u64 iocb;	offset:16;	size:8;	signed:0;
   field:u64 sk;	offset:24;	size:8;	signed:0;
   field:u64 msg;	offset:32;	size:8;	signed:0;
   field:u64 size;	offset:40;	size:8;	signed:0;

 print fmt: "(%lx) iocb=0x%Lx sk=0x%Lx msg=0x%Lx size=0x%Lx", REC->__probe_ip, REC->iocb, REC->sk, REC->msg, REC->size
 #

 Do some system wide tracing of this probe + write syscalls:

 # perf trace -e write --ev probe:* --filter-pids 6385
  462.612 (0.010 ms): bash/19153 write(fd: 1, buf: 0x7f7556c78000, count: 29               ) = 29
  462.701 (0.027 ms): sshd/19152 write(fd: 3, buf: 0x7f78dd12e160, count: 68           ) ...
  462.701 (        ): probe:tcp_sendmsg:(ffffffff8163db30) iocb=0xffff8803ebec7e70 sk=0xffff88042196ab80 msg=0xffff8803ebec7da8 size=0x44)
  462.710 (0.035 ms): sshd/19152  ... [continued]: write()) = 68
  462.787 (0.009 ms): bash/19153 write(fd: 2, buf: 0x7f7556c77000, count: 22               ) = 22
  462.865 (0.002 ms): sshd/19152 write(fd: 3, buf: 0x7f78dd12e160, count: 68           ) ...
  462.865 (        ): probe:tcp_sendmsg:(ffffffff8163db30) iocb=0xffff8803ebec7e70 sk=0xffff88042196ab80 msg=0xffff8803ebec7da8 size=0x44)
  462.873 (0.010 ms): sshd/19152  ... [continued]: write()) = 68

Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Ananth N Mavinakayanahalli 
Cc: David Ahern 
Cc: Hemant Kumar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20150506124653.4961.59806.stgit@localhost.localdomain
[ Add some examples to the changelog message showing how to use it ]
Signed-off-by: Arnaldo Carvalho de Melo

perf probe: Accept filter argument for --funcs

2015-05-05T21:13:04+00:00

This allows the user to pass the filter pattern directly to the --funcs
option as below:

  ----
  # ./perf probe -F *kmalloc
  __kmalloc
  devm_kmalloc
  mempool_kmalloc
  sg_kmalloc
  sock_kmalloc
  ----

We previously needed to use the --filter option for that.

Signed-off-by: Masami Hiramatsu 
Tested-by: Arnaldo Carvalho de Melo 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20150505022950.23399.22435.stgit@localhost.localdomain
Signed-off-by: Arnaldo Carvalho de Melo

perf record: Add AUX area tracing Snapshot Mode support

2015-05-05T21:13:01+00:00

Add a new option and support for Instruction Tracing Snapshot Mode.
When the new option is selected, no AUX area tracing data is captured
until a signal (SIGUSR2) is received.

Signed-off-by: Adrian Hunter 
Acked-by: Jiri Olsa 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1430404667-10593-10-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf auxtrace: Add option to synthesize events for transactions

2015-05-05T21:12:55+00:00

Add AUX area tracing option 'x' to synthesize events for transactions.
This will be used by Intel PT to synthesize an event record for each TSX
start, commit or abort.

Signed-off-by: Adrian Hunter 
Acked-by: Jiri Olsa 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1430404667-10593-6-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo