lkml.org 
[lkml]   [2014]   [Sep]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [GIT PULL 00/13] perf/core improvements and fixes

* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:

> Hi Ingo,
>
> Please consider pulling,
>
> - Arnaldo
>
> The following changes since commit 39b5a56ec0be5effe9b7d0f18cb27724bf2e5d47:
>
> Merge branch 'rfc/perf' into perf/core, because it's ready for inclusion (2014-08-24 22:35:42 +0200)
>
> are available in the git repository at:
>
>
> git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo
>
> for you to fetch changes up to fa3480cb0a96c8af9ded1a8c836c91bd5704bda1:
>
> perf tools: Add perf_pmu__scan_file() (2014-09-08 17:16:06 -0300)
>
> ----------------------------------------------------------------
> perf/core improvements and fixes:
>
> User visible:
>
> o Add +field argument support for --sort option (Jiri Olsa)
>
> Infrastructure:
>
> o More intel PT prep work, from Adrian Hunter, including:
>
> - Let a user specify a PMU event without any config terms
> - Add perf-with-kcore script
> - Build programs to copy 32-bit compatibility VDSOs
> - Add support for 32-bit compatibility VDSOs
> - Let default config be defined for a PMU
> - Add perf_pmu__scan_file()
>
> o Add feature checks to .gitignore (Alexander Shishkin)
>
> o "perf kvm stat report" improvements by Alexander Yarygin:
> o Save pid string in opts.target.pid
> o Enable the target.system_wide flag
> o Unify the title bar output
>
> o Fix build issue on powerpc when DWARF support is disabled (Anton Blanchard)
>
> o Allow to specify lib compile variable for spec usage (Jiri Olsa)
>
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
>
> ----------------------------------------------------------------
> Adrian Hunter (6):
> perf tools: Let a user specify a PMU event without any config terms
> perf tools: Add perf-with-kcore script
> perf tools: Build programs to copy 32-bit compatibility VDSOs
> perf tools: Add support for 32-bit compatibility VDSOs
> perf tools: Let default config be defined for a PMU
> perf tools: Add perf_pmu__scan_file()
>
> Alexander Shishkin (1):
> perf tools: Add feature checks to .gitignore
>
> Alexander Yarygin (3):
> perf kvm stat report: Save pid string in opts.target.pid
> perf kvm stat report: Enable the target.system_wide flag
> perf kvm stat report: Unify the title bar output
>
> Anton Blanchard (1):
> perf tools powerpc: Fix build issue when DWARF support is disabled
>
> Jiri Olsa (2):
> perf tools: Add +field argument support for --sort option
> perf tools: Allow to specify lib compile variable for spec usage
>
> tools/perf/.gitignore | 2 +
> tools/perf/Makefile.perf | 42 ++++++-
> tools/perf/arch/powerpc/Makefile | 2 +-
> tools/perf/builtin-kvm.c | 23 ++--
> tools/perf/config/Makefile | 33 ++++-
> tools/perf/config/Makefile.arch | 8 ++
> tools/perf/perf-read-vdso.c | 59 +++++++++
> tools/perf/perf-with-kcore.sh | 259 +++++++++++++++++++++++++++++++++++++++
> tools/perf/tests/pmu.c | 2 +-
> tools/perf/util/kvm-stat.h | 1 -
> tools/perf/util/parse-events.c | 13 +-
> tools/perf/util/parse-events.y | 10 ++
> tools/perf/util/pmu.c | 79 +++++++++---
> tools/perf/util/pmu.h | 12 +-
> tools/perf/util/sort.c | 37 +++++-
> tools/perf/util/vdso.c | 170 ++++++++++++++++++++++++-
> tools/perf/util/vdso.h | 4 +-
> 17 files changed, 712 insertions(+), 44 deletions(-)
> create mode 100644 tools/perf/perf-read-vdso.c
> create mode 100644 tools/perf/perf-with-kcore.sh

Hm, so I'm getting this error, when trying to profile a 32-bit
ELF binary on a 64-bit kernel, on a testbox:

$ perf record ~/hackbench 10
Time: 0.115
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.373 MB perf.data (~16279 samples) ]
sh: perf-read-vdso32: command not found

comet:~/tip> file ~/hackbench
/home/mingo/hackbench: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.2.5, not stripped

64-bit executables work fine:

$ perf record perf bench sched pipe
# Running 'sched/pipe' benchmark:
# Executed 1000000 pipe operations between two processes

Total time: 13.374 [sec]

13.374352 usecs/op
74769 ops/sec
[ perf record: Woken up 14 times to write data ]
[ perf record: Captured and wrote 3.602 MB perf.data (~157354 samples) ]

The kernel is an older one:

$ uname -a
Linux comet 3.14.17-100.fc19.x86_64 #1 SMP Thu Aug 14 17:17:26
UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

Thanks,

Ingo


\
 
 \ /
  Last update: 2014-09-09 07:21    [W:0.125 / U:0.264 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site