linux-stable.git/tools/perf/util, branch linux-6.2.y

perf stat: Separate bperf from bpf_profiler

2023-05-17T11:58:54+00:00

[ Upstream commit ecc68ee216c6c5b2f84915e1441adf436f1b019b ]

It seems that perf stat -b  doesn't produce any results:

    $ perf stat -e cycles -b 4 -I 10000 -vvv
    Control descriptor is not initialized
    cycles: 0 0 0
                time        counts unit      events
	10.007641640          cycles

Looks like this happens because fentry/fexit progs are getting loaded, but the
corresponding perf event is not enabled and not added into the events bpf map.
I think there is some mixing up between two type of bpf support, one for bperf
and one for bpf_profiler. Both are identified via evsel__is_bpf, based on which
perf events are enabled, but for the latter (bpf_profiler) a perf event is
required. Using evsel__is_bperf to check only bperf produces expected results:

    $ perf stat -e cycles -b 4 -I 10000 -vvv
    Control descriptor is not initialized
    ------------------------------------------------------------
    perf_event_attr:
      size                             136
      sample_type                      IDENTIFIER
      read_format                      TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING
      disabled                         1
      exclude_guest                    1
    ------------------------------------------------------------
    sys_perf_event_open: pid -1  cpu 0  group_fd -1  flags 0x8 = 3
    ------------------------------------------------------------
    [...perf_event_attr for other CPUs...]
    ------------------------------------------------------------
    cycles: 309426 169009 169009
		time             counts unit events
	10.010091271             309426      cycles

The final numbers correspond (at least in the level of magnitude) to the
same metric obtained via bpftool.

Fixes: 112cb56164bc2108 ("perf stat: Introduce config stat.bpf-counter-events")
Reviewed-by: Song Liu 
Signed-off-by: Dmitrii Dolgov <9erthalion6@gmail.com>
Tested-by: Song Liu 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Song Liu 
Link: https://lore.kernel.org/r/20230412182316.11628-1-9erthalion6@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf tracepoint: Fix memory leak in is_valid_tracepoint()

2023-05-17T11:58:54+00:00

[ Upstream commit 9b86c49710eec7b4fbb78a0232b2dd0972a2b576 ]

When is_valid_tracepoint() returns 1, need to call put_events_file() to
free `dir_path`.

Fixes: 25a7d914274de386 ("perf parse-events: Use get/put_events_file()")
Signed-off-by: Yang Jihong 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: https://lore.kernel.org/r/20230421025953.173826-1-yangjihong1@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf symbols: Fix return incorrect build_id size in elf_read_build_id()

2023-05-17T11:58:53+00:00

[ Upstream commit 1511e4696acb715a4fe48be89e1e691daec91c0e ]

In elf_read_build_id(), if gnu build_id is found, should return the size of
the actually copied data. If descsz is greater thanBuild_ID_SIZE,
write_buildid data access may occur.

Fixes: be96ea8ffa788dcc ("perf symbols: Fix issue with binaries using 16-bytes buildids (v2)")
Reported-by: Will Ochowicz 
Signed-off-by: Yang Jihong 
Tested-by: Will Ochowicz 
Acked-by: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Leo Yan 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: https://lore.kernel.org/lkml/CWLP265MB49702F7BA3D6D8F13E4B1A719C649@CWLP265MB4970.GBRP265.PROD.OUTLOOK.COM/T/
Link: https://lore.kernel.org/r/20230427012841.231729-1-yangjihong1@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf cs-etm: Fix timeless decode mode detection

2023-05-17T11:58:53+00:00

[ Upstream commit 449067f3fc9f340da54e383738286881e6634d0b ]

In this context, timeless refers to the trace data rather than the perf
event data. But when detecting whether there are timestamps in the trace
data or not, the presence of a timestamp flag on any perf event is used.

Since commit f42c0ce573df ("perf record: Always get text_poke events
with --kcore option") timestamps were added to a tracking event when
--kcore is used which breaks this detection mechanism. Fix it by
detecting if trace timestamps exist by looking at the ETM config flags.
This would have always been a more accurate way of doing it anyway.

This fixes the following error message when using --kcore with
Coresight:

  $ perf record --kcore -e cs_etm// --per-thread
  $ perf report
  The perf.data/data data has no samples!

Fixes: f42c0ce573df79d1 ("perf record: Always get text_poke events with --kcore option")
Reported-by: Yang Shi 
Signed-off-by: James Clark 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: John Garry 
Cc: Leo Yan 
Cc: Mark Rutland 
Cc: Mathieu Poirier 
Cc: Mike Leach 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Suzuki Poulouse 
Cc: Will Deacon 
Cc: coresight@lists.linaro.org
Cc: denik@google.com
Cc: linux-arm-kernel@lists.infradead.org
Link: https://lore.kernel.org/lkml/CAHbLzkrJQTrYBtPkf=jf3OpQ-yBcJe7XkvQstX9j2frz4WF-SQ@mail.gmail.com/
Link: https://lore.kernel.org/r/20230424134748.228137-2-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf map: Delete two variable initialisations before null pointer checks in sort__sym_from_cmp()

2023-05-17T11:58:53+00:00

[ Upstream commit c160118a90d4acf335993d8d59b02ae2147a524e ]

Addresses of two data structure members were determined before
corresponding null pointer checks in the implementation of the function
“sort__sym_from_cmp”.

Thus avoid the risk for undefined behaviour by removing extra
initialisations for the local variables “from_l” and “from_r” (also
because they were already reassigned with the same value behind this
pointer check).

This issue was detected by using the Coccinelle software.

Fixes: 1b9e97a2a95e4941 ("perf tools: Fix report -F symbol_from for data without branch info")
Signed-off-by: 
Acked-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Andi Kleen 
Cc: German Gomez 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Kan Liang 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Link: https://lore.kernel.org/cocci/54a21fea-64e3-de67-82ef-d61b90ffad05@web.de/
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf pmu: zfree() expects a pointer to a pointer to zero it after freeing its contents

2023-05-17T11:58:52+00:00

[ Upstream commit 57f14b5ae1a97537f2abd2828ee7212cada7036e ]

An audit showed just this one problem with zfree(), fix it.

Fixes: 9fbc61f832ebf432 ("perf pmu: Add support for PMU capabilities")
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf hist: Improve srcfile sort key performance (really)

2023-05-17T11:58:52+00:00

[ Upstream commit 6094c7744bb0563e833e81d8df8513f9a4e7a257 ]

The earlier commit f0cdde28fecc0d7f ("perf hist: Improve srcfile sort
key performance") updated the srcfile logic but missed to change the
->cmp() callback which is called for every sample.

It should use the same logic like in the srcline to speed up the
processing because it'd return the same information repeatedly for the
same address.  The real processing will be done in
sort__srcfile_collapse().

Fixes: f0cdde28fecc0d7f ("perf hist: Improve srcfile sort key performance")
Signed-off-by: Namhyung Kim 
Cc: Adrian Hunter 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: https://lore.kernel.org/r/20230323025005.191239-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf script: Fix Python support when no libtraceevent

2023-05-17T11:58:52+00:00

[ Upstream commit 80c3a7d9f20401169283b5670dbb8d7ac07a1d55 ]

Python scripting can be used without libtraceevent. In particular,
scripting for Intel PT does not use tracepoints, and so does not need
libtraceevent support.

Alter the build and employ conditional compilation to allow Python
scripting without libtraceevent.

Example:

 Before:

    $ ldd `which perf` | grep -i python
    $ ldd `which perf` | grep -i libtraceevent
    $ perf record -e intel_pt//u uname
    Linux
    [ perf record: Woken up 1 times to write data ]
    [ perf record: Captured and wrote 0.031 MB perf.data ]
    $ perf script intel-pt-events.py |& head -3
      Error: Couldn't find script `intel-pt-events.py'

     See perf script -l for available scripts.

 After:

    $ ldd `which perf` | grep -i python
            libpython3.10.so.1.0 => /lib/x86_64-linux-gnu/libpython3.10.so.1.0 (0x00007f4bac400000)
    $ ldd `which perf` | grep -i libtraceevent
    $ perf script intel-pt-events.py | head
    Intel PT Branch Trace, Power Events, Event Trace and PTWRITE
         Switch In    8021/8021  [000]     11234.097713404     0/0
           perf-exec  8021/8021  [000]     11234.098041726       psb                        offset: 0x0                0 [unknown] ([unknown])
           perf-exec  8021/8021  [000]     11234.098041726       cbr                         45  freq: 4505 MHz  (161%)                0 [unknown] ([unknown])
               uname  8021/8021  [000]     11234.098082170  branches:uH  tr strt                              0 [unknown] ([unknown]) => 7f3a8b9422b0 _start+0x0 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2)
               uname  8021/8021  [000]     11234.098082379  branches:uH  tr end                    7f3a8b9422b0 _start+0x0 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2) => 0 [unknown] ([unknown])
               uname  8021/8021  [000]     11234.098083629  branches:uH  tr strt                              0 [unknown] ([unknown]) => 7f3a8b9422b0 _start+0x0 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2)
               uname  8021/8021  [000]     11234.098083629  branches:uH  call                      7f3a8b9422b3 _start+0x3 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2) => 7f3a8b943050 _dl_start+0x0 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2)
               uname  8021/8021  [000]     11234.098083837  branches:uH  tr end                    7f3a8b943060 _dl_start+0x10 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2) => 0 [unknown] ([unknown])  IPC: 0.01 (9/938)
               uname  8021/8021  [000]     11234.098084670  branches:uH  tr strt                              0 [unknown] ([unknown]) => 7f3a8b943060 _dl_start+0x10 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2)

Fixes: 378ef0f5d9d7f465 ("perf build: Use libtraceevent from the system")
Signed-off-by: Adrian Hunter 
Cc: Ian Rogers 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: https://lore.kernel.org/r/20230315084321.14563-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin

perf intel-pt: Fix CYC timestamps after standalone CBR

2023-05-11T14:11:36+00:00

commit 430635a0ef1ce958b7b4311f172694ece2c692b8 upstream.

After a standalone CBR (not associated with TSC), update the cycles
reference timestamp and reset the cycle count, so that CYC timestamps
are calculated relative to that point with the new frequency.

Fixes: cc33618619cefc6d ("perf tools: Add Intel PT support for decoding CYC packets")
Signed-off-by: Adrian Hunter 
Cc: Adrian Hunter 
Cc: Ian Rogers 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230403154831.8651-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf auxtrace: Fix address filter entire kernel size

2023-05-11T14:11:36+00:00

commit 1f9f33ccf0320be21703d9195dd2b36a1c9a07cb upstream.

kallsyms is not completely in address order.

In find_entire_kern_cb(), calculate the kernel end from the maximum
address not the last symbol.

Example:

 Before:

    $ sudo cat /proc/kallsyms | grep ' [twTw] ' | tail -1
    ffffffffc00b8bd0 t bpf_prog_6deef7357e7b4530    [bpf]
    $ sudo cat /proc/kallsyms | grep ' [twTw] ' | sort | tail -1
    ffffffffc15e0cc0 t iwl_mvm_exit [iwlmvm]
    $ perf.d093603a05aa record -v --kcore -e intel_pt// --filter 'filter *' -- uname |& grep filter
    Address filter: filter 0xffffffff93200000/0x2ceba000

 After:

    $ perf.8fb0f7a01f8e record -v --kcore -e intel_pt// --filter 'filter *' -- uname |& grep filter
    Address filter: filter 0xffffffff93200000/0x2e3e2000

Fixes: 1b36c03e356936d6 ("perf record: Add support for using symbols in address filters")
Signed-off-by: Adrian Hunter 
Cc: Adrian Hunter 
Cc: Ian Rogers 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230403154831.8651-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman