提交 · d110162cafc80dad0622cfd40f3113aebb77e1bb · openeuler / Kernel

23 9月, 2020 4 次提交

perf tsc: Support cap_user_time_short for event TIME_CONV · d110162c

由 Leo Yan 提交于 9月 14, 2020

The synthesized event TIME_CONV doesn't contain the complete parameters
for counters, this will lead to wrong conversion between counter cycles
and timestamp.

This patch extends event TIME_CONV to record flags 'cap_user_time_zero'
which is used to indicate the counter parameters are valid or not, if
not will directly return 0 for timestamp calculation.  And record the
flag 'cap_user_time_short' and its relevant fields 'time_cycles' and
'time_mask' for cycle calibration.
Signed-off-by: NLeo Yan <leo.yan@linaro.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kemeng Shi <shikemeng@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nick Gasson <nick.gasson@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Remi Bernon <rbernon@codeweavers.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steve Maclean <steve.maclean@microsoft.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zou Wei <zou_wei@huawei.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lore.kernel.org/lkml/20200914115311.2201-5-leo.yan@linaro.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

d110162c

perf tsc: Calculate timestamp with cap_user_time_short · 78a93d4c

由 Leo Yan 提交于 9月 14, 2020

The perf mmap'ed buffer contains the flag 'cap_user_time_short' and two
extra fields 'time_cycles' and 'time_mask', perf tool needs to know them
for handling the counter wrapping case.

This patch is to reads out the relevant parameters from the head of the
first mmap'ed page and stores into the structure 'perf_tsc_conversion',
if the flag 'cap_user_time_short' has been set, it will firstly
calibrate cycle value for timestamp calculation.

Committer testing:

Before/after:

  # perf test tsc
  70: Convert perf time to TSC                                        : Ok
  #
  # perf test -v tsc
  70: Convert perf time to TSC                                        :
  --- start ---
  test child forked, pid 11059
  mmap size 528384B
  1st event perf time 996384576521 tsc 3850532906613
  rdtsc          time 996384578455 tsc 3850532913950
  2nd event perf time 996384578845 tsc 3850532915428
  test child finished with 0
  ---- end ----
  Convert perf time to TSC: Ok
  #
Signed-off-by: NLeo Yan <leo.yan@linaro.org>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kemeng Shi <shikemeng@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nick Gasson <nick.gasson@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Remi Bernon <rbernon@codeweavers.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steve Maclean <steve.maclean@microsoft.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zou Wei <zou_wei@huawei.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lore.kernel.org/lkml/20200914115311.2201-4-leo.yan@linaro.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

78a93d4c

perf tsc: Add rdtsc() for Arm64 · 4979e861

由 Leo Yan 提交于 9月 14, 2020

The system register CNTVCT_EL0 can be used to retrieve the counter from
user space.  Add rdtsc() for Arm64.
Signed-off-by: NLeo Yan <leo.yan@linaro.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kemeng Shi <shikemeng@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nick Gasson <nick.gasson@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Remi Bernon <rbernon@codeweavers.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steve Maclean <steve.maclean@microsoft.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zou Wei <zou_wei@huawei.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lore.kernel.org/lkml/20200914115311.2201-3-leo.yan@linaro.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

4979e861

perf tsc: Move out common functions from x86 · 03fca3af

由 Leo Yan 提交于 9月 14, 2020

Functions perf_read_tsc_conversion() and perf_event__synth_time_conv()
should work as common functions rather than x86 specific, so move these
two functions out from arch/x86 folder and place them into util/tsc.c.

Since the function perf_event__synth_time_conv() will be linked in
util/tsc.c, remove its weak version.

Committer testing:

Before/after:

  # perf test tsc
  70: Convert perf time to TSC                                        : Ok
  #
  # perf test -v tsc
  70: Convert perf time to TSC                                        :
  --- start ---
  test child forked, pid 8520
  mmap size 528384B
  1st event perf time 592110439891 tsc 2317172044331
  rdtsc          time 592110441915 tsc 2317172052010
  2nd event perf time 592110442336 tsc 2317172053605
  test child finished with 0
  ---- end ----
  Convert perf time to TSC: Ok
  #
Signed-off-by: NLeo Yan <leo.yan@linaro.org>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kemeng Shi <shikemeng@huawei.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nick Gasson <nick.gasson@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Remi Bernon <rbernon@codeweavers.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Steve Maclean <steve.maclean@microsoft.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zou Wei <zou_wei@huawei.com>
Link: http://lore.kernel.org/lkml/20200914115311.2201-2-leo.yan@linaro.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

03fca3af

18 9月, 2020 11 次提交

perf probe: Fall back to debuginfod query if debuginfo and source not found locally · 7cd5738d

由 Masami Hiramatsu 提交于 9月 18, 2020

Since 'perf probe' heavily depends on debuginfo, debuginfod gives us
many benefits on the 'perf probe' command on remote machine.

Especially, this will be helpful for the embedded devices which will not
have enough storage, or boot with a cross-build kernel whose source code
is in the host machine.

This will work as similar to commit c7a14fdc ("perf build-ids:
Fall back to debuginfod query if debuginfo not found")

Tested with:

  (host) $ cd PATH/TO/KBUILD/DIR/
  (host) $ debuginfod -F .
  ...

  (remote) # perf probe -L vfs_read
  Failed to find the path for the kernel: No such file or directory
    Error: Failed to show lines.

  (remote) # export DEBUGINFOD_URLS="http://$HOST_IP:8002/"
  (remote) # perf probe -L vfs_read
  <vfs_read@...>
        0  ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
           {
        2         ssize_t ret;

                  if (!(file->f_mode & FMODE_READ))
                          return -EBADF;
        6         if (!(file->f_mode & FMODE_CAN_READ))
                          return -EINVAL;
        8         if (unlikely(!access_ok(buf, count)))
                          return -EFAULT;

       11         ret = rw_verify_area(READ, file, pos, count);
       12         if (ret)
                          return ret;
                  if (count > MAX_RW_COUNT)
  ...

  (remote) # perf probe -a "vfs_read count"
  Added new event:
    probe:vfs_read       (on vfs_read with count)

  (remote) # perf probe -l
    probe:vfs_read       (on vfs_read@ksrc/linux/fs/read_write.c with count)
Signed-off-by: NMasami Hiramatsu <mhiramat@kernel.org>
Reviewed-by: NFrank Ch. Eigler <fche@redhat.com>
Cc: Aaron Merey <amerey@redhat.com>
Cc: Daniel Thompson <daniel.thompson@linaro.org>
Link: http://lore.kernel.org/lkml/160041610083.912668.13659563860278615846.stgit@devnote2Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

7cd5738d

perf probe: Fix to adjust symbol address with correct reloc_sym address · ac7a75d1

由 Masami Hiramatsu 提交于 9月 18, 2020

'perf probe' uses ref_reloc_sym to adjust symbol offset address from
debuginfo address or ref_reloc_sym based address, but that is misusing
reloc_sym->addr and reloc_sym->unrelocated_addr.  If map is not
relocated (map->reloc == 0), we can use reloc_sym->addr as unrelocated
address instead of reloc_sym->unrelocated_addr.

This usually does not happen. If we have a non-stripped ELF binary, we
will use it for map and debuginfo, if not, we use only kallsyms without
debuginfo. Thus, the map is always relocated (ELF and DWARF binary) or
not relocated (kallsyms).

However, if we allow the combination of debuginfo and kallsyms based map
(like using debuginfod), we have to check the map->reloc and choose the
collect address of reloc_sym.
Signed-off-by: NMasami Hiramatsu <mhiramat@kernel.org>
Reviewed-by: NFrank Ch. Eigler <fche@redhat.com>
Cc: Aaron Merey <amerey@redhat.com>
Cc: Daniel Thompson <daniel.thompson@linaro.org>
Link: http://lore.kernel.org/lkml/160041609047.912668.14314639291419159274.stgit@devnote2Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

ac7a75d1

perf intel-pt: Fix "context_switch event has no tid" error · 7d537a8d

由 Adrian Hunter 提交于 9月 09, 2020

A context_switch event can have no tid because pids can be detached from
a task while the task is still running (in do_exit()). Note this won't
happen with per-task contexts because then tracing stops at
perf_event_exit_task()

If a task with no tid gets preempted, or a dying task gets preempted and
its parent releases it, when it subsequently gets switched back in,
Intel PT will not be able to determine what task is running and prints
an error "context_switch event has no tid". However, it is not really an
error because the task is in kernel space and the decoder can continue
to decode successfully. Fix by changing the error to be only a logged
message, and make allowance for tid == -1.

Example:

  Using 5.9-rc4 with Preemptible Kernel (Low-Latency Desktop) e.g.
  $ uname -r
  5.9.0-rc4
  $ grep PREEMPT .config
  # CONFIG_PREEMPT_NONE is not set
  # CONFIG_PREEMPT_VOLUNTARY is not set
  CONFIG_PREEMPT=y
  CONFIG_PREEMPT_COUNT=y
  CONFIG_PREEMPTION=y
  CONFIG_PREEMPT_RCU=y
  CONFIG_PREEMPT_NOTIFIERS=y
  CONFIG_DRM_I915_PREEMPT_TIMEOUT=640
  CONFIG_DEBUG_PREEMPT=y
  # CONFIG_PREEMPT_TRACER is not set
  # CONFIG_PREEMPTIRQ_DELAY_TEST is not set

Before:

  $ cat forkit.c

  #include <sys/types.h>
  #include <unistd.h>
  #include <sys/wait.h>

  int main()
  {
          pid_t child;
          int status = 0;

          child = fork();
          if (child == 0)
                  return 123;
          wait(&status);
          return 0;
  }

  $ gcc -o forkit forkit.c
  $ sudo ~/bin/perf record --kcore -a -m,64M -e intel_pt/cyc/k &
  [1] 11016
  $ taskset 2 ./forkit
  $ sudo pkill perf
  $ [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 17.262 MB perf.data ]

  [1]+  Terminated              sudo ~/bin/perf record --kcore -a -m,64M -e intel_pt/cyc/k
  $ sudo ~/bin/perf script --show-task-events --show-switch-events --itrace=iqqe-o -C 1 --ns | grep -C 2 forkit
  context_switch event has no tid
           taskset 11019 [001] 66663.270045029:          1 instructions:k:  ffffffffb1d9f844 strnlen_user+0xb4 ([kernel.kallsyms])
           taskset 11019 [001] 66663.270201816:          1 instructions:k:  ffffffffb1a83121 unmap_page_range+0x561 ([kernel.kallsyms])
            forkit 11019 [001] 66663.270327553: PERF_RECORD_COMM exec: forkit:11019/11019
            forkit 11019 [001] 66663.270420028:          1 instructions:k:  ffffffffb1db9537 __clear_user+0x27 ([kernel.kallsyms])
            forkit 11019 [001] 66663.270648704:          1 instructions:k:  ffffffffb18829e6 do_user_addr_fault+0xf6 ([kernel.kallsyms])
            forkit 11019 [001] 66663.270833163:          1 instructions:k:  ffffffffb230a825 irqentry_exit_to_user_mode+0x15 ([kernel.kallsyms])
            forkit 11019 [001] 66663.271092359:          1 instructions:k:  ffffffffb1aea3d9 lock_page_memcg+0x9 ([kernel.kallsyms])
            forkit 11019 [001] 66663.271207092: PERF_RECORD_FORK(11020:11020):(11019:11019)
            forkit 11019 [001] 66663.271234775: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid: 11020/11020
            forkit 11020 [001] 66663.271238407: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 11019/11019
            forkit 11020 [001] 66663.271312066:          1 instructions:k:  ffffffffb1a88140 handle_mm_fault+0x10 ([kernel.kallsyms])
            forkit 11020 [001] 66663.271476225: PERF_RECORD_EXIT(11020:11020):(11019:11019)
            forkit 11020 [001] 66663.271497488: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 11019/11019
            forkit 11019 [001] 66663.271500523: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 11020/11020
            forkit 11019 [001] 66663.271517241:          1 instructions:k:  ffffffffb24012cd error_entry+0x6d ([kernel.kallsyms])
            forkit 11019 [001] 66663.271664080: PERF_RECORD_EXIT(11019:11019):(1386:1386)

After:

  $ sudo ~/bin/perf script --show-task-events --show-switch-events --itrace=iqqe-o -C 1 --ns | grep -C 2 forkit
           taskset 11019 [001] 66663.270045029:          1 instructions:k:  ffffffffb1d9f844 strnlen_user+0xb4 ([kernel.kallsyms])
           taskset 11019 [001] 66663.270201816:          1 instructions:k:  ffffffffb1a83121 unmap_page_range+0x561 ([kernel.kallsyms])
            forkit 11019 [001] 66663.270327553: PERF_RECORD_COMM exec: forkit:11019/11019
            forkit 11019 [001] 66663.270420028:          1 instructions:k:  ffffffffb1db9537 __clear_user+0x27 ([kernel.kallsyms])
            forkit 11019 [001] 66663.270648704:          1 instructions:k:  ffffffffb18829e6 do_user_addr_fault+0xf6 ([kernel.kallsyms])
            forkit 11019 [001] 66663.270833163:          1 instructions:k:  ffffffffb230a825 irqentry_exit_to_user_mode+0x15 ([kernel.kallsyms])
            forkit 11019 [001] 66663.271092359:          1 instructions:k:  ffffffffb1aea3d9 lock_page_memcg+0x9 ([kernel.kallsyms])
            forkit 11019 [001] 66663.271207092: PERF_RECORD_FORK(11020:11020):(11019:11019)
            forkit 11019 [001] 66663.271234775: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid: 11020/11020
            forkit 11020 [001] 66663.271238407: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 11019/11019
            forkit 11020 [001] 66663.271312066:          1 instructions:k:  ffffffffb1a88140 handle_mm_fault+0x10 ([kernel.kallsyms])
            forkit 11020 [001] 66663.271476225: PERF_RECORD_EXIT(11020:11020):(11019:11019)
            forkit 11020 [001] 66663.271497488: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 11019/11019
            forkit 11019 [001] 66663.271500523: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 11020/11020
            forkit 11019 [001] 66663.271517241:          1 instructions:k:  ffffffffb24012cd error_entry+0x6d ([kernel.kallsyms])
            forkit 11019 [001] 66663.271664080: PERF_RECORD_EXIT(11019:11019):(1386:1386)
            forkit 11019 [001] 66663.271688752: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:    -1/-1
               :-1    -1 [001] 66663.271692086: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 11019/11019
                :-1    -1 [001] 66663.271707466:          1 instructions:k:  ffffffffb18eb096 update_load_avg+0x306 ([kernel.kallsyms])

Fixes: 86c27869 ("perf intel-pt: Add support for PERF_RECORD_SWITCH")
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Yu-cheng Yu <yu-cheng.yu@intel.com>
Link: http://lore.kernel.org/lkml/20200909084923.9096-3-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

7d537a8d

perf script: Display negative tid in non-sample events · fc18380f

由 Adrian Hunter 提交于 9月 09, 2020

The kernel can release tasks while they are still running. This can
result in a task having no tid, in which case perf records a tid of -1.
Improve the perf script output in that case.

Example:

Before:

  # cat ./autoreap.c

  #include <sys/types.h>
  #include <unistd.h>
  #include <sys/wait.h>
  #include <signal.h>

  struct sigaction act = {
          .sa_handler = SIG_IGN,
  };

  int main()
  {
          pid_t child;
          int status = 0;

          sigaction(SIGCHLD, &act, NULL);
          child = fork();
          if (child == 0)
                  return 123;
          wait(&status);
          return 0;
  }

  # gcc -o autoreap autoreap.c
  # ./perf record -a -e dummy --switch-events ./autoreap
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.948 MB perf.data ]
  # ./perf script --show-task-events --show-switch-events | grep -C2 'autoreap\|4294967295\|-1'
           swapper     0 [004] 18462.673613: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25189/25189
              perf 25189 [004] 18462.673614: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
          autoreap 25189 [004] 18462.673800: PERF_RECORD_COMM exec: autoreap:25189/25189
          autoreap 25189 [004] 18462.674042: PERF_RECORD_FORK(25191:25191):(25189:25189)
          autoreap 25189 [004] 18462.674050: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [004] 18462.674051: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 25189/25189
           swapper     0 [005] 18462.674083: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25191/25191
          autoreap 25191 [005] 18462.674084: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
           swapper     0 [003] 18462.674121: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid:    11/11
       rcu_preempt    11 [003] 18462.674121: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
       rcu_preempt    11 [003] 18462.674124: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [003] 18462.674124: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:    11/11
          autoreap 25191 [005] 18462.674138: PERF_RECORD_EXIT(25191:25191):(25189:25189)
  PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [005] 18462.674149: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 4294967295/4294967295
           swapper     0 [004] 18462.674182: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25189/25189
          autoreap 25189 [004] 18462.674183: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
          autoreap 25189 [004] 18462.674218: PERF_RECORD_EXIT(25189:25189):(25188:25188)
          autoreap 25189 [004] 18462.674225: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [004] 18462.674226: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 25189/25189
           swapper     0 [007] 18462.674257: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25188/25188

After:

  # ./perf script --show-task-events --show-switch-events | grep -C2 'autoreap\|4294967295\|-1'
           swapper     0 [004] 18462.673613: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25189/25189
              perf 25189 [004] 18462.673614: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
          autoreap 25189 [004] 18462.673800: PERF_RECORD_COMM exec: autoreap:25189/25189
          autoreap 25189 [004] 18462.674042: PERF_RECORD_FORK(25191:25191):(25189:25189)
          autoreap 25189 [004] 18462.674050: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [004] 18462.674051: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 25189/25189
           swapper     0 [005] 18462.674083: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25191/25191
          autoreap 25191 [005] 18462.674084: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
           swapper     0 [003] 18462.674121: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid:    11/11
       rcu_preempt    11 [003] 18462.674121: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
       rcu_preempt    11 [003] 18462.674124: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [003] 18462.674124: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:    11/11
          autoreap 25191 [005] 18462.674138: PERF_RECORD_EXIT(25191:25191):(25189:25189)
               :-1    -1 [005] 18462.674149: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [005] 18462.674149: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:    -1/-1
           swapper     0 [004] 18462.674182: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25189/25189
          autoreap 25189 [004] 18462.674183: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid:     0/0
          autoreap 25189 [004] 18462.674218: PERF_RECORD_EXIT(25189:25189):(25188:25188)
          autoreap 25189 [004] 18462.674225: PERF_RECORD_SWITCH_CPU_WIDE OUT          next pid/tid:     0/0
           swapper     0 [004] 18462.674226: PERF_RECORD_SWITCH_CPU_WIDE IN           prev pid/tid: 25189/25189
           swapper     0 [007] 18462.674257: PERF_RECORD_SWITCH_CPU_WIDE OUT preempt  next pid/tid: 25188/25188
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Yu-cheng Yu <yu-cheng.yu@intel.com>
Link: http://lore.kernel.org/lkml/20200909084923.9096-2-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

fc18380f

perf docs: Improve help information in perf.txt · 99f63817

由 Zejiang Tang 提交于 9月 09, 2020

perf has many undocumented options, such as:-vv, --exec-path,
--html-path, -p, --paginate,--no-pager, --debugfs-dir, --buildid-dir,
--list-cmds, --list-opts.

Add entris for these options in perf.txt.
Signed-off-by: NZejiang Tang <tangzejiang@loongson.cn>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Link: http://lore.kernel.org/lkml/1599645194-8438-1-git-send-email-tangzejiang@loongson.cnSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

99f63817

perf metric: Remove duplicate include · a803fbe6

由 YueHaibing 提交于 9月 15, 2020

Remove duplicate header which is included twice.
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20200915081541.41004-1-yuehaibing@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

a803fbe6

perf tools: Add documentation for topdown metrics · 328781df

由 Andi Kleen 提交于 9月 11, 2020

Add some documentation how to use the topdown metrics in ring 3.
Signed-off-by: NAndi Kleen <ak@linux.intel.com>
Co-developed-by: NKan Liang <kan.liang@linux.intel.com>
Acked-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200911144808.27603-5-kan.liang@linux.intel.comSigned-off-by: NKan Liang <kan.liang@linux.intel.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

328781df

perf stat: Support new per thread TopDown metrics · 55c36a9f

由 Andi Kleen 提交于 9月 11, 2020

Icelake has support for reporting per thread TopDown metrics.

These are reported differently than the previous TopDown support,
each metric is standalone, but scaled to pipeline "slots".

We don't need to do anything special for HyperThreading anymore.
Teach perf stat --topdown to handle these new metrics and
print them in the same way as the previous TopDown metrics.

The restrictions of only being able to report information per core is
gone.
Signed-off-by: NAndi Kleen <ak@linux.intel.com>
Co-developed-by: NKan Liang <kan.liang@linux.intel.com>
Acked-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200911144808.27603-4-kan.liang@linux.intel.comSigned-off-by: NKan Liang <kan.liang@linux.intel.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

55c36a9f

perf record: Support sample-read topdown metric group · acb65150

由 Kan Liang 提交于 9月 11, 2020

With the hardware TopDown metrics feature, sample-read feature should be
supported for a topdown group, e.g., sample a non-topdown event and read
a topdown metric group. But the current perf record code errors out.

For a topdown metric group, the slots event must be the leader of the
group, but the leader slots event doesn't support sampling.

To support sample-read the topdown metric group, use the 2nd event of
the group as the "leader" for the purposes of sampling.

Only the platform with Topdown metic feature supports sample-read the
topdown group. Add arch_topdown_sample_read() to indicate whether the
topdown group supports sample-read.
Signed-off-by: NKan Liang <kan.liang@linux.intel.com>
Acked-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200911144808.27603-3-kan.liang@linux.intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

acb65150

perf tools: Rename group to topdown · 687986bb

由 Kan Liang 提交于 9月 11, 2020

The group.h/c only include TopDown group related functions. The name
"group" is too generic and inaccurate. Use the name "topdown" to replace
it.

Move topdown related functions to a dedicated file, topdown.c.
Signed-off-by: NKan Liang <kan.liang@linux.intel.com>
Acked-by: NJiri Olsa <jolsa@redhat.com>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200911144808.27603-2-kan.liang@linux.intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

687986bb

perf machine: Add machine__for_each_dso() function · c57f5eaa

由 Jiri Olsa 提交于 9月 13, 2020

Add the machine__for_each_dso() to iterate over all dso objects defined
for the within a machine object. It will be used in the MMAP3 patch
series.
Signed-off-by: NJiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Budankov <alexey.budankov@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200913210313.1985612-22-jolsa@kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

c57f5eaa

16 9月, 2020 1 次提交

perf stat: Fix the ratio comments of miss-events · ce9c13f3

由 Qi Liu 提交于 9月 16, 2020

'perf stat' displays miss ratio of L1-dcache, L1-icache, dTLB cache,
iTLB cache and LL-cache. Take L1-dcache for example, miss ratio is
caculated as "L1-dcache-load-misses/L1-dcache-loads". So "of all
L1-dcache hits" is unsuitable to describe it, and "of all L1-dcache
accesses" seems better.

The comments of L1-icache, dTLB cache, iTLB cache and LL-cache are
fixed in the same way.
Signed-off-by: NQi Liu <liuqi115@huawei.com>
Reviewed-by: NAndi Kleen <ak@linux.intel.com>
Cc: linuxarm@huawei.com
Link: http://lore.kernel.org/lkml/1600253331-10535-1-git-send-email-liuqi115@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

ce9c13f3

15 9月, 2020 17 次提交

perf test: Free formats for perf pmu parse test · d26383dc