linux.git/tools/perf/util, branch v3.3-rc5

perf evsel: Fix an issue where perf report fails to show the proper percentage

2012-02-06T20:59:38+00:00

This patch fixes an issue where perf report shows nan% for certain
perf.data files. The below is from a report for a do_fork probe:

   -nan%           sshd  [kernel.kallsyms]  [k] do_fork
   -nan%    packagekitd  [kernel.kallsyms]  [k] do_fork
   -nan%    dbus-daemon  [kernel.kallsyms]  [k] do_fork
   -nan%           bash  [kernel.kallsyms]  [k] do_fork

A git bisect shows commit f3bda2c as the cause. However, looking back
through the git history, I saw commit 640c03c which seems to have
removed the required initialization for perf_sample->period. The problem
only started showing after commit f3bda2c. The below patch re-introduces
the initialization and it fixes the problem for me.

With the below patch, for the same perf.data:

  73.08%             bash  [kernel.kallsyms]  [k] do_fork
   8.97%      11-dhclient  [kernel.kallsyms]  [k] do_fork
   6.41%             sshd  [kernel.kallsyms]  [k] do_fork
   3.85%        20-chrony  [kernel.kallsyms]  [k] do_fork
   2.56%         sendmail  [kernel.kallsyms]  [k] do_fork

This patch applies over current linux-tip commit 9949284.

Problem introduced in:

$ git describe 640c03c
v2.6.37-rc3-83-g640c03c

Cc: Ananth N Mavinakayanahalli 
Cc: Ingo Molnar 
Cc: Robert Richter 
Cc: Srikar Dronamraju 
Cc: stable@kernel.org
Link: http://lkml.kernel.org/r/20120203170113.5190.25558.stgit@localhost6.localdomain6
Signed-off-by: Naveen N. Rao 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix prefix matching for kernel maps

2012-02-06T20:57:39+00:00

In some perf ancient versions we used '[kernel.kallsyms._text]' as the
name for the kernel map.

This got changed with commit:
  perf: 'perf kvm' tool for monitoring guest performance from host
  commit a1645ce12adb6c9cc9e19d7695466204e3f017fe
  Author: Zhang, Yanmin 

and we started to use following name '[kernel.kallsyms]_text'.

This name change is important for the report code dealing with ancient
perf data. When processing the kernel map event, we need to recognize
the old naming (dont match the last ']') and initialize the kernel map
correctly.

The subsequent call to maps__set_kallsyms_ref_reloc_sym deals with the
superfluous ']' to get correct symbol name.

Cc: Corey Ashford 
Cc: Ingo Molnar 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1328461865-6127-1-git-send-email-jolsa@redhat.com
Signed-off-by: Jiri Olsa 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix strlen() bug in perf_event__synthesize_event_type()

2012-01-31T00:29:12+00:00

The event_type record has a max length for the event name.

It's called MAX_EVENT_NAME.

The name may be truncated to fit the max length. But the header.size still
reflects the original name length. If that length is > MAX_EVENT_NAME, then the
header.size field is bogus. Fix this by using the length of the name after the
potential truncation.

Cc: David Ahern 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20120120094912.GA4882@quad
Signed-off-by: Stephane Eranian 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix broken build by defining _GNU_SOURCE in Makefile

2012-01-31T00:19:54+00:00

When building on my Debian/mips system, util/util.c fails to build
because commit 1aed2671738785e8f5aea663a6fda91aa7ef59b5 (perf kvm: Do
guest-only counting by default) indirectly includes stdio.h before the
feature selection in util.h is done.  This prevents _GNU_SOURCE in
util.h from enabling the declaration of getline(), from now second
inclusion of stdio.h, and the build is broken.

There is another breakage in util/evsel.c caused by include ordering,
but I didn't fully track down the commit that caused it.

The root cause of all this is an inconsistent definition of _GNU_SOURCE,
so I move the definition into the Makefile so that it is passed to all
invocations of the compiler and used uniformly for all system header
files.  All other #define and #undef of _GNU_SOURCE are removed as they
cause conflicts with the definition passed to the compiler.

All the features.h definitions (_LARGEFILE64_SOURCE _FILE_OFFSET_BITS=64
and _GNU_SOURCE) are needed by the python glue code too, so they are
moved to BASIC_CFLAGS, and the misleading comments about BASIC_CFLAGS
are removed.

This gives me a clean build on x86_64 (fc12) and mips (Debian).

Cc: David Daney 
Cc: Ingo Molnar 
Cc: Joerg Roedel 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1326836461-11952-1-git-send-email-ddaney.cavm@gmail.com
Signed-off-by: David Daney 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix compile error on x86_64 Ubuntu

2012-01-08T15:34:55+00:00

The ctype.h include is not needed here and it breaks build on some systems (at
least 64bit Ubuntu 10.04) like below. Just get rid of it.

    CC util/trace-event-info.o
cc1: warnings being treated as errors
util/trace-event-info.c: In function ‘record_file’:
util/trace-event-info.c:192: error: implicit declaration of function ‘pwrite’
util/trace-event-info.c:192: error: nested extern declaration of ‘pwrite’
make: *** [util/trace-event-info.o] Error 1

Cc: Ingo Molnar 
Cc: Joerg Roedel 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1326035430-7621-1-git-send-email-namhyung@gmail.com
Signed-off-by: Namhyung Kim 
Signed-off-by: Arnaldo Carvalho de Melo

perf report: Fix --stdio output alignment when --showcpuutilization used

2012-01-08T15:32:51+00:00

Current perf report output is broken if --showcpuutilization is used.
Combination with -n and/or --show-total-period make things worse.
This patch fixes it as follows:

before:
    48.25%    48.25%     0.00%    sleep  [kernel.kallsyms]  [k] trace_hardirqs_off
    34.99%    34.99%     0.00%    sleep  [kernel.kallsyms]  [k] __find_get_block_slow
    15.99%    15.99%     0.00%    sleep  [kernel.kallsyms]  [k] lock_release_holdtime
     0.77%     0.77%     0.00%    sleep  [kernel.kallsyms]  [k] native_write_msr_safe

after:
    48.25%    48.25%     0.00%    sleep  [kernel.kallsyms]  [k] trace_hardirqs_off
    34.99%    34.99%     0.00%    sleep  [kernel.kallsyms]  [k] __find_get_block_slow
    15.99%    15.99%     0.00%    sleep  [kernel.kallsyms]  [k] lock_release_holdtime
     0.77%     0.77%     0.00%    sleep  [kernel.kallsyms]  [k] native_write_msr_safe

Cc: Ingo Molnar 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1325957132-10600-8-git-send-email-namhyung@gmail.com
Signed-off-by: Namhyung Kim 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Add support for guest/host-only profiling

2012-01-06T17:49:49+00:00

To restrict a counter to either host or guest mode this patch introduces
two new event modifiers: G and H.

With G the counter is configured in guest-only mode and with H in
host-only mode.

Cc: Gleb Natapov 
Cc: Ingo Molnar 
Cc: Joerg Roedel 
Cc: Peter Zijlstra 
Signed-off-by: Gleb Natapov 
Signed-off-by: Joerg Roedel 
Link: http://lkml.kernel.org/n/tip-or5aj3rghy9ngyg882z6kln9@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf kvm: Do guest-only counting by default

2012-01-06T17:47:37+00:00

Make use of exclude_guest and exlude_host in perf-kvm to do only
guest-only counting by default.

Cc: Gleb Natapov 
Cc: Ingo Molnar 
Cc: Joerg Roedel 
Cc: Peter Zijlstra 
Signed-off-by: Gleb Natapov 
Signed-off-by: Joerg Roedel 
[ committer note: Moved perf_{guest,host} & event_attr_init to util.c ]
[                 so as not to drag more stuff to the python binding]
Signed-off-by: Arnaldo Carvalho de Melo

perf hists: Stop using 'self' for struct hist_entry

2012-01-06T17:42:52+00:00

Stop using this python/OOP convention, doesn't really helps. Will do
more from time to time till we get it cleaned up in all of /perf.

Suggested-by: Thomas Gleixner 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/n/tip-me4dyj6s5snh7jr8wb9gzt82@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf hists: Rename total_session to total_period

2012-01-06T17:42:08+00:00

Nowadays we do it per evsel, not per session (that may have multiple
evsels), so rename it to avoid confusion.

Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/n/tip-azsgomr5h4dmaudoogw48w49@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo