linux-stable.git/tools/perf, branch v5.4.63

perf record/stat: Explicitly call out event modifiers in the documentation

2020-09-05T09:22:49+00:00

commit e48a73a312ebf19cc3d72aa74985db25c30757c1 upstream.

Event modifiers are not mentioned in the perf record or perf stat
manpages.  Add them to orient new users more effectively by pointing
them to the perf list manpage for details.

Fixes: 2055fdaf8703 ("perf list: Document precise event sampling for AMD IBS")
Signed-off-by: Kim Phillips 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Alexey Budankov 
Cc: Ian Rogers 
Cc: Jin Yao 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Paul Clarke 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tony Jones 
Cc: stable@vger.kernel.org
Link: http://lore.kernel.org/lkml/20200901215853.276234-1-kim.phillips@amd.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf probe: Fix memory leakage when the probe point is not found

2020-08-26T08:40:48+00:00

[ Upstream commit 12d572e785b15bc764e956caaa8a4c846fd15694 ]

Fix the memory leakage in debuginfo__find_trace_events() when the probe
point is not found in the debuginfo. If there is no probe point found in
the debuginfo, debuginfo__find_probes() will NOT return -ENOENT, but 0.

Thus the caller of debuginfo__find_probes() must check the tf.ntevs and
release the allocated memory for the array of struct probe_trace_event.

The current code releases the memory only if the debuginfo__find_probes()
hits an error but not checks tf.ntevs. In the result, the memory allocated
on *tevs are not released if tf.ntevs == 0.

This fixes the memory leakage by checking tf.ntevs == 0 in addition to
ret < 0.

Fixes: ff741783506c ("perf probe: Introduce debuginfo to encapsulate dwarf information")
Signed-off-by: Masami Hiramatsu 
Reviewed-by: Srikar Dronamraju 
Cc: Andi Kleen 
Cc: Oleg Nesterov 
Cc: stable@vger.kernel.org
Link: http://lore.kernel.org/lkml/159438668346.62703.10887420400718492503.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf bench mem: Always memset source before memcpy

2020-08-21T11:05:38+00:00

[ Upstream commit 1beaef29c34154ccdcb3f1ae557f6883eda18840 ]

For memcpy, the source pages are memset to zero only when --cycles is
used.  This leads to wildly different results with or without --cycles,
since all sources pages are likely to be mapped to the same zero page
without explicit writes.

Before this fix:

$ export cmd="./perf stat -e LLC-loads -- ./perf bench \
  mem memcpy -s 1024MB -l 100 -f default"
$ $cmd

         2,935,826      LLC-loads
       3.821677452 seconds time elapsed

$ $cmd --cycles

       217,533,436      LLC-loads
       8.616725985 seconds time elapsed

After this fix:

$ $cmd

       214,459,686      LLC-loads
       8.674301124 seconds time elapsed

$ $cmd --cycles

       214,758,651      LLC-loads
       8.644480006 seconds time elapsed

Fixes: 47b5757bac03c338 ("perf bench mem: Move boilerplate memory allocation to the infrastructure")
Signed-off-by: Vincent Whitchurch 
Cc: Alexander Shishkin 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: kernel@axis.com
Link: http://lore.kernel.org/lkml/20200810133404.30829-1-vincent.whitchurch@axis.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf intel-pt: Fix duplicate branch after CBR

2020-08-21T11:05:29+00:00

commit a58a057ce65b52125dd355b7d8b0d540ea267a5f upstream.

CBR events can result in a duplicate branch event, because the state
type defaults to a branch. Fix by clearing the state type.

Example: trace 'sleep' and hope for a frequency change

 Before:

   $ perf record -e intel_pt//u sleep 0.1
   [ perf record: Woken up 1 times to write data ]
   [ perf record: Captured and wrote 0.034 MB perf.data ]
   $ perf script --itrace=bpe > before.txt

 After:

   $ perf script --itrace=bpe > after.txt
   $ diff -u before.txt after.txt
#  --- before.txt  2020-07-07 14:42:18.191508098 +0300
#  +++ after.txt   2020-07-07 14:42:36.587891753 +0300
   @@ -29673,7 +29673,6 @@
               sleep 93431 [007] 15411.619905:          1  branches:u:                 0 [unknown] ([unknown]) =>     7f0818abb2e0 clock_nanosleep@@GLIBC_2.17+0x0 (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
               sleep 93431 [007] 15411.619905:          1  branches:u:      7f0818abb30c clock_nanosleep@@GLIBC_2.17+0x2c (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>                0 [unknown] ([unknown])
               sleep 93431 [007] 15411.720069:         cbr:  cbr: 15 freq: 1507 MHz ( 56%)         7f0818abb30c clock_nanosleep@@GLIBC_2.17+0x2c (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
   -           sleep 93431 [007] 15411.720069:          1  branches:u:      7f0818abb30c clock_nanosleep@@GLIBC_2.17+0x2c (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>                0 [unknown] ([unknown])
               sleep 93431 [007] 15411.720076:          1  branches:u:                 0 [unknown] ([unknown]) =>     7f0818abb30e clock_nanosleep@@GLIBC_2.17+0x2e (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
               sleep 93431 [007] 15411.720077:          1  branches:u:      7f0818abb323 clock_nanosleep@@GLIBC_2.17+0x43 (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>     7f0818ac0eb7 __nanosleep+0x17 (/usr/lib/x86_64-linux-gnu/libc-2.31.so)
               sleep 93431 [007] 15411.720077:          1  branches:u:      7f0818ac0ebf __nanosleep+0x1f (/usr/lib/x86_64-linux-gnu/libc-2.31.so) =>     55cb7e4c2827 rpl_nanosleep+0x97 (/usr/bin/sleep)

Fixes: 91de8684f1cff ("perf intel-pt: Cater for CBR change in PSB+")
Fixes: abe5a1d3e4bee ("perf intel-pt: Decoder to output CBR changes immediately")
Signed-off-by: Adrian Hunter 
Reviewed-by: Andi Kleen 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: stable@vger.kernel.org
Link: http://lore.kernel.org/lkml/20200710151104.15137-3-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf intel-pt: Fix FUP packet state

2020-08-21T11:05:29+00:00

commit 401136bb084fd021acd9f8c51b52fe0a25e326b2 upstream.

While walking code towards a FUP ip, the packet state is
INTEL_PT_STATE_FUP or INTEL_PT_STATE_FUP_NO_TIP. That was mishandled
resulting in the state becoming INTEL_PT_STATE_IN_SYNC prematurely.  The
result was an occasional lost EXSTOP event.

Signed-off-by: Adrian Hunter 
Reviewed-by: Andi Kleen 
Cc: Jiri Olsa 
Cc: stable@vger.kernel.org
Link: http://lore.kernel.org/lkml/20200710151104.15137-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf bench: Share some global variables to fix build with gcc 10

2020-08-05T07:59:52+00:00

commit e4d9b04b973b2dbce7b42af95ea70d07da1c936d upstream.

Noticed with gcc 10 (fedora rawhide) that those variables were not being
declared as static, so end up with:

  ld: /tmp/build/perf/bench/epoll-wait.o:/git/perf/tools/perf/bench/epoll-wait.c:93: multiple definition of `end'; /tmp/build/perf/bench/futex-hash.o:/git/perf/tools/perf/bench/futex-hash.c:40: first defined here
  ld: /tmp/build/perf/bench/epoll-wait.o:/git/perf/tools/perf/bench/epoll-wait.c:93: multiple definition of `start'; /tmp/build/perf/bench/futex-hash.o:/git/perf/tools/perf/bench/futex-hash.c:40: first defined here
  ld: /tmp/build/perf/bench/epoll-wait.o:/git/perf/tools/perf/bench/epoll-wait.c:93: multiple definition of `runtime'; /tmp/build/perf/bench/futex-hash.o:/git/perf/tools/perf/bench/futex-hash.c:40: first defined here
  ld: /tmp/build/perf/bench/epoll-ctl.o:/git/perf/tools/perf/bench/epoll-ctl.c:38: multiple definition of `end'; /tmp/build/perf/bench/futex-hash.o:/git/perf/tools/perf/bench/futex-hash.c:40: first defined here
  ld: /tmp/build/perf/bench/epoll-ctl.o:/git/perf/tools/perf/bench/epoll-ctl.c:38: multiple definition of `start'; /tmp/build/perf/bench/futex-hash.o:/git/perf/tools/perf/bench/futex-hash.c:40: first defined here
  ld: /tmp/build/perf/bench/epoll-ctl.o:/git/perf/tools/perf/bench/epoll-ctl.c:38: multiple definition of `runtime'; /tmp/build/perf/bench/futex-hash.o:/git/perf/tools/perf/bench/futex-hash.c:40: first defined here
  make[4]: *** [/git/perf/tools/build/Makefile.build:145: /tmp/build/perf/bench/perf-in.o] Error 1

Prefix those with bench__ and add them to bench/bench.h, so that we can
share those on the tools needing to access those variables from signal
handlers.

Acked-by: Thomas Gleixner 
Cc: Adrian Hunter 
Cc: Davidlohr Bueso 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/20200303155811.GD13702@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo 
Cc: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

perf env: Do not return pointers to local variables

2020-08-05T07:59:52+00:00

commit ebcb9464a2ae3a547e97de476575c82ece0e93e2 upstream.

It is possible to return a pointer to a local variable when looking up
the architecture name for the running system and no normalization is
done on that value, i.e. we may end up returning the uts.machine local
variable.

While this doesn't happen on most arches, as normalization takes place,
lets fix this by making that a static variable and optimize it a bit by
not always running uname(), only the first time.

Noticed in fedora rawhide running with:

  [perfbuilder@a5ff49d6e6e4 ~]$ gcc --version
  gcc (GCC) 10.0.1 20200216 (Red Hat 10.0.1-0.8)

Reported-by: Jiri Olsa 
Cc: Adrian Hunter 
Cc: Namhyung Kim 
Signed-off-by: Arnaldo Carvalho de Melo 
Cc: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

perf tests bp_account: Make global variable static

2020-08-05T07:59:52+00:00

commit cff20b3151ccab690715cb6cf0f5da5cccb32adf upstream.

To fix the build with newer gccs, that without this patch exit with:

    LD       /tmp/build/perf/tests/perf-in.o
  ld: /tmp/build/perf/tests/bp_account.o:/git/perf/tools/perf/tests/bp_account.c:22: multiple definition of `the_var'; /tmp/build/perf/tests/bp_signal.o:/git/perf/tools/perf/tests/bp_signal.c:38: first defined here
  make[4]: *** [/git/perf/tools/build/Makefile.build:145: /tmp/build/perf/tests/perf-in.o] Error 1

First noticed in fedora:rawhide/32 with:

  [perfbuilder@a5ff49d6e6e4 ~]$ gcc --version
  gcc (GCC) 10.0.1 20200216 (Red Hat 10.0.1-0.8)

Reported-by: Jiri Olsa 
Cc: Adrian Hunter 
Cc: Namhyung Kim 
Signed-off-by: Arnaldo Carvalho de Melo 
Cc: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

perf tools: Fix record failure when mixed with ARM SPE event

2020-08-05T07:59:49+00:00

[ Upstream commit bd3c628f8fafa6cbd6a1ca440034b841f0080160 ]

When recording with cache-misses and arm_spe_x event, I found that it
will just fail without showing any error info if i put cache-misses
after 'arm_spe_x' event.

  [root@localhost 0620]# perf record -e cache-misses \
				-e arm_spe_0/ts_enable=1,pct_enable=1,pa_enable=1,load_filter=1,jitter=1,store_filter=1,min_latency=0/ sleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.067 MB perf.data ]
  [root@localhost 0620]#
  [root@localhost 0620]# perf record -e arm_spe_0/ts_enable=1,pct_enable=1,pa_enable=1,load_filter=1,jitter=1,store_filter=1,min_latency=0/ \
				     -e  cache-misses sleep 1
  [root@localhost 0620]#

The current code can only work if the only event to be traced is an
'arm_spe_x', or if it is the last event to be specified. Otherwise the
last event type will be checked against all the arm_spe_pmus[i]->types,
none will match and an out of bound 'i' index will be used in
arm_spe_recording_init().

We don't support concurrent multiple arm_spe_x events currently, that
is checked in arm_spe_recording_options(), and it will show the relevant
info. So add the check and record of the first found 'arm_spe_pmu' to
fix this issue here.

Fixes: ffd3d18c20b8 ("perf tools: Add ARM Statistical Profiling Extensions (SPE) support")
Signed-off-by: Wei Li 
Reviewed-by: Mathieu Poirier 
Tested-by-by: Leo Yan 
Cc: Alexander Shishkin 
Cc: Hanjun Guo 
Cc: Jiri Olsa 
Cc: Kim Phillips 
Cc: Mark Rutland 
Cc: Mike Leach 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Suzuki Poulouse 
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lore.kernel.org/lkml/20200724071111.35593-2-liwei391@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf stat: Zero all the 'ena' and 'run' array slot stats for interval mode

2020-07-22T07:33:05+00:00

commit 0e0bf1ea1147fcf74eab19c2d3c853cc3740a72f upstream.

As the code comments in perf_stat_process_counter() say, we calculate
counter's data every interval, and the display code shows ps->res_stats
avg value. We need to zero the stats for interval mode.

But the current code only zeros the res_stats[0], it doesn't zero the
res_stats[1] and res_stats[2], which are for ena and run of counter.

This patch zeros the whole res_stats[] for interval mode.

Fixes: 51fd2df1e882 ("perf stat: Fix interval output values")
Signed-off-by: Jin Yao 
Cc: Alexander Shishkin 
Cc: Andi Kleen 
Cc: Jin Yao 
Cc: Jiri Olsa 
Cc: Kan Liang 
Cc: Peter Zijlstra 
Link: http://lore.kernel.org/lkml/20200409070755.17261-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman