1. 28 May, 2020 40 commits
    • Ravi Bangoria's avatar
      perf powerpc: Don't ignore sym-handling.c file · 39548e50
      Ravi Bangoria authored
      Commit 7eec00a7 ("perf symbols: Consolidate symbol fixup issue")
      removed powerpc specific sym-handling.c file from Build. This wasn't
      caught by build CI because all functions in this file are declared
      as __weak in common code. Fix it.
      
      Fixes: 7eec00a7 ("perf symbols: Consolidate symbol fixup issue")
      Reported-by: default avatarSandipan Das <sandipan@linux.ibm.com>
      Signed-off-by: default avatarRavi Bangoria <ravi.bangoria@linux.ibm.com>
      Reviewed-by: default avatarLeo Yan <leo.yan@linaro.org>
      Reviewed-by: default avatarNaveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
      Acked-by: default avatarSandipan Das <sandipan@linux.ibm.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Link: http://lore.kernel.org/lkml/20200509112113.174745-1-ravi.bangoria@linux.ibm.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      39548e50
    • Ian Rogers's avatar
      perf expr: Test parsing of floating point numbers · 63f11355
      Ian Rogers authored
      Add test for fix in:
      commit 5741da3dee4c ("perf expr: Parse numbers as doubles")
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: Jiri Olsa <jolsa@redhat.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200513062752.3681-1-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      63f11355
    • Anand K Mistry's avatar
      perf record: Use an eventfd to wakeup when done · da231338
      Anand K Mistry authored
      The setting and checking of 'done' contains a rare race where the signal
      handler setting 'done' is run after checking to break the loop, but
      before waiting in evlist__poll(). In this case, the main loop won't wake
      up until either another signal is sent, or the perf data fd causes a
      wake up.
      
      The following simple script can trigger this condition (but you might
      need to run it for several hours):
      
      for ((i = 0; i >= 0; i++)) ; do
        echo "Loop $i"
        delay=$(echo "scale=4; 0.1 * $RANDOM/32768" | bc)
        ./perf record -- sleep 30000000 >/dev/null&
        pid=$!
        sleep $delay
        kill -TERM $pid
        echo "PID $pid"
        wait $pid
      done
      
      At some point, the loop will stall. Adding logging, even though perf has
      received the SIGTERM and set 'done = 1', perf will remain sleeping until
      a second signal is sent.
      
      Committer notes:
      
      Make this dependent on HAVE_EVENTFD_SUPPORT, so that we continue
      building on older systems without the eventfd syscall.
      Signed-off-by: default avatarAnand K Mistry <amistry@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200513122012.v3.1.I4d7421c6bbb1f83ea58419082481082e19097841@changeidSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      da231338
    • Arnaldo Carvalho de Melo's avatar
      tools feature: Rename HAVE_EVENTFD to HAVE_EVENTFD_SUPPORT · ba35fe93
      Arnaldo Carvalho de Melo authored
      To be consistent with other such auto-detected features.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Anand K Mistry <amistry@google.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      ba35fe93
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Initialize evsel->per_pkg_mask to NULL in evsel__init() · f0aef475
      Arnaldo Carvalho de Melo authored
      Just like with the other fields, this probably isn't fixing anything
      observable as evsel__new() uses zalloc() for the whole 'struct evsel',
      but since evsels can be embedded in larger structures and maybe those
      larger structures don't use zalloc() for some reason, init it to NULL
      just in case.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: Ian Rogers <irogers@google.com>
      Cc: Jiri Olsa <jolsa@redhat.com>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      f0aef475
    • Ian Rogers's avatar
      perf evsel: Fix 2 memory leaks · 3efc899d
      Ian Rogers authored
      If allocated, perf_pkg_mask and metric_events need freeing.
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Reviewed-by: default avatarAndi Kleen <ak@linux.intel.com>
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Jiri Olsa <jolsa@redhat.com>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200512235918.10732-1-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      3efc899d
    • Arnaldo Carvalho de Melo's avatar
      perf parse-events: Fix incorrect conversion of 'if () free()' to 'zfree()' · 7fcdccd4
      Arnaldo Carvalho de Melo authored
      When applying a patch by Ian I incorrectly converted to zfree() an
      expression that involved testing some other struct member, not the one
      being freed, which lead to bugs reproduceable by:
      
        $ perf stat -e i/bs,tsc,L2/o sleep 1
        WARNING: multiple event parsing errors
        Segmentation fault (core dumped)
        $
      
      Fix it by restoring the test for pos->free_str before freeing
      pos->val.str, but continue using zfree(&pos->val.str) to set that member
      to NULL after freeing it.
      Reported-by: default avatarIan Rogers <irogers@google.com>
      Fixes: e8dfb818 ("perf parse-events: Fix memory leaks found on parse_events")
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: clang-built-linux@googlegroups.com
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      7fcdccd4
    • Jiri Olsa's avatar
      perf tools: Fix is_bpf_image function logic · e12a89ef
      Jiri Olsa authored
      Adrian reported that is_bpf_image is not working the way it was intended
      - passing on trampolines and dispatcher names. Instead it returned true
      for all the bpf names.
      
      The reason even this logic worked properly is that all bpf objects, even
      trampolines and dispatcher, were assigned DSO_BINARY_TYPE__BPF_IMAGE
      binary_type.
      
      The later for bpf_prog objects, the binary_type was fixed in bpf load
      event processing, which is executed after the ksymbol code.
      
      Fixing the is_bpf_image logic, so it properly recognizes trampoline and
      dispatcher objects.
      
      Fixes: 3c29d448 ("perf annotate: Add basic support for bpf_image")
      Reported-by: default avatarAdrian Hunter <adrian.hunter@intel.com>
      Signed-off-by: default avatarJiri Olsa <jolsa@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Michael Petlan <mpetlan@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200512122310.3154754-1-jolsa@kernel.orgSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      e12a89ef
    • Ian Rogers's avatar
      perf c2c: Fix 'perf c2c record -e list' to show the default events used · b027cc6f
      Ian Rogers authored
      When the event is passed as list, the default events should be listed as
      per 'perf mem record -e list'. Previous behavior is:
      
        $ perf c2c record -e list
        failed: event 'list' not found, use '-e list' to get list of available events
      
         Usage: perf c2c record [<options>] [<command>]
            or: perf c2c record [<options>] -- <command> [<options>]
      
            -e, --event <event>   event selector. Use 'perf mem record -e list' to list available events
        $
      
      New behavior:
      
        $ perf c2c record -e list
        ldlat-loads  : available
        ldlat-stores : available
      
      v3: is a rebase.
      v2: addresses review comments by Jiri Olsa.
      
      https://lore.kernel.org/lkml/20191127081844.GH32367@krava/Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Acked-by: default avatarJiri Olsa <jolsa@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200507220604.3391-1-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      b027cc6f
    • Paul A. Clarke's avatar
      perf vendor events power9: Add missing metrics to POWER9 'cpi_breakdown' · 63b5930f
      Paul A. Clarke authored
      Add the following metrics to the POWER9 'cpi_breakdown' metricgroup:
      
      - ict_noslot_br_mpred_cpi
      - ict_noslot_br_mpred_icmiss_cpi
      - ict_noslot_cyc_other_cpi
      - ict_noslot_disp_held_cpi
      - ict_noslot_disp_held_hb_full_cpi
      - ict_noslot_disp_held_issq_cpi
      - ict_noslot_disp_held_other_cpi
      - ict_noslot_disp_held_sync_cpi
      - ict_noslot_disp_held_tbegin_cpi
      - ict_noslot_ic_l2_cpi
      - ict_noslot_ic_l3_cpi
      - ict_noslot_ic_l3miss_cpi
      - ict_noslot_ic_miss_cpi
      Signed-off-by: default avatarPaul Clarke <pc@us.ibm.com>
      Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
      Reviewed-by: default avatarKajol Jain <kjain@linux.ibm.com>
      Tested-by: default avatarIan Rogers <irogers@google.com>
      Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
      Cc: Michael Ellerman <mpe@ellerman.id.au>
      Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
      Cc: Sukadev Bhattiprolu <sukadev@linux.ibm.com>
      Link: http://lore.kernel.org/lkml/1588868938-21933-3-git-send-email-pc@us.ibm.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      63b5930f
    • Ian Rogers's avatar
      perf record: Add dummy event during system wide synthesis · 0a892c1c
      Ian Rogers authored
      During the processing of /proc during event synthesis new processes may
      start. Add a dummy event if /proc is to be processed, to capture mmaps
      for starting processes. This reuses the existing logic for
      initial-delay.
      
      v3 fixes the attr test of test-record-C0
      v2 fixes the dummy event configuration and a branch stack issue.
      Suggested-by: default avatarStephane Eranian <eranian@google.com>
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Jiri Olsa <jolsa@redhat.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200422173615.59436-1-irogers@google.com
      [ split from a larger patch ]
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      0a892c1c
    • Ian Rogers's avatar
      perf evsel: Dummy events never triggers, no need to ask for PERF_SAMPLE_BRANCH_STACK · 5885a202
      Ian Rogers authored
      A dummy event never triggers any actual counter and therefore cannot be
      used with branch_stack
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Jiri Olsa <jolsa@redhat.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200422173615.59436-1-irogers@google.com
      [ split from a larger patch ]
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      5885a202
    • Jin Yao's avatar
      perf parse-events: Use strcmp() to compare the PMU name · 8510895b
      Jin Yao authored
      A big uncore event group is split into multiple small groups which only
      include the uncore events from the same PMU. This has been supported in
      the commit 3cdc5c2c ("perf parse-events: Handle uncore event
      aliases in small groups properly").
      
      If the event's PMU name starts to repeat, it must be a new event.
      That can be used to distinguish the leader from other members.
      But now it only compares the pointer of pmu_name
      (leader->pmu_name == evsel->pmu_name).
      
      If we use "perf stat -M LLC_MISSES.PCIE_WRITE -a" on cascadelakex,
      the event list is:
      
        evsel->name					evsel->pmu_name
        ---------------------------------------------------------------
        unc_iio_data_req_of_cpu.mem_write.part0		uncore_iio_4 (as leader)
        unc_iio_data_req_of_cpu.mem_write.part0		uncore_iio_2
        unc_iio_data_req_of_cpu.mem_write.part0		uncore_iio_0
        unc_iio_data_req_of_cpu.mem_write.part0		uncore_iio_5
        unc_iio_data_req_of_cpu.mem_write.part0		uncore_iio_3
        unc_iio_data_req_of_cpu.mem_write.part0		uncore_iio_1
        unc_iio_data_req_of_cpu.mem_write.part1		uncore_iio_4
        ......
      
      For the event "unc_iio_data_req_of_cpu.mem_write.part1" with
      "uncore_iio_4", it should be the event from PMU "uncore_iio_4".
      It's not a new leader for this PMU.
      
      But if we use "(leader->pmu_name == evsel->pmu_name)", the check
      would be failed and the event is stored to leaders[] as a new
      PMU leader.
      
      So this patch uses strcmp to compare the PMU name between events.
      
      Fixes: d4953f7e ("perf parse-events: Fix 3 use after frees found with clang ASAN")
      Signed-off-by: default avatarJin Yao <yao.jin@linux.intel.com>
      Acked-by: default avatarJiri Olsa <jolsa@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: Jin Yao <yao.jin@intel.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200430003618.17002-1-yao.jin@linux.intel.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      8510895b
    • Paul A. Clarke's avatar
      perf stat: Increase perf metric output resolution · d4d5ca0b
      Paul A. Clarke authored
      Add another digit of precision to the perf metrics output.
      
      Before:
      
        $ /usr/bin/perf stat --metrics run_cpi /bin/ls
        [...]
                 4,345,526      pm_run_cyc                #      1.1 run_cpi
                 3,818,069      pm_run_inst_cmpl
        [...]
        $ /usr/bin/perf stat --metrics run_cpi --metric-only /bin/ls
        [...]
                     run_cpi
                         1.1
        [...]
      
      After:
      
        $ perf stat --metrics run_cpi /bin/ls
        [...]
                 4,280,882      pm_run_cyc                #     1.12 run_cpi
                 3,817,016      pm_run_inst_cmpl
        [...]
        $ perf stat --metrics run_cpi --metric-only /bin/ls
        [...]
                     run_cpi
                        1.06
        [...]
      
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Kan Liang <kan.liang@intel.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarPaul Clarke <pc@us.ibm.com>
      LPU-Reference: 1588861087-31280-1-git-send-email-pc@us.ibm.com
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      d4d5ca0b
    • Ian Rogers's avatar
      perf expr: Print a debug message for division by zero · 9be27a5d
      Ian Rogers authored
      If an expression yields 0 and is then divided-by/modulus-by then the
      parsing aborts. Add a debug error message to better enable debugging
      when this happens.
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-11-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      9be27a5d
    • Ian Rogers's avatar
      perf metrics: Fix parse errors in power9 metrics · f2682a8f
      Ian Rogers authored
      Mismatched parentheses.
      
      Fixes: 7f3cf5ac (perf vendor events power9: Cpi_breakdown & estimated_dcache_miss_cpi metrics)
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Reviewed-by: default avatarPaul Clarke <pc@us.ibm.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-10-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      f2682a8f
    • Ian Rogers's avatar
      perf metrics: Fix parse errors in power8 metrics · 981d169f
      Ian Rogers authored
      Mismatched parentheses.
      
      Fixes: dd81eafa (perf vendor events power8: Cpi_breakdown & estimated_dcache_miss_cpi metrics)
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Reviewed-by: default avatarPaul Clarke <pc@us.ibm.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-9-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      981d169f
    • Ian Rogers's avatar
      perf expr: Debug lex if debugging yacc · e5e0e635
      Ian Rogers authored
      Only effects parser debugging (disabled by default). Enables displaying
      '--accepting rule at line .. ("...").
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-8-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      e5e0e635
    • Ian Rogers's avatar
      perf expr: Parse numbers as doubles · 7db2fd0b
      Ian Rogers authored
      This is expected in expr.y and metrics use floating point values such as
      x86 broadwell IFetch_Line_Utilization.
      
      Fixes: 26226a97 (perf expr: Move expr lexer to flex)
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Tested-by: default avatarKajol Jain <kjain@linux.ibm.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-7-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      7db2fd0b
    • Ian Rogers's avatar
      perf expr: Increase max other · f59d3f84
      Ian Rogers authored
      Large metrics such as Branch_Misprediction_Cost_SMT on x86 broadwell
      need more space.
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-6-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      f59d3f84
    • Ian Rogers's avatar
      perf expr: Allow ',' to be an other token · cb59fa79
      Ian Rogers authored
      Corrects parse errors in expr__find_other of expressions with min.
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-5-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      cb59fa79
    • Ian Rogers's avatar
      perf metrics: Fix parse errors in skylake metrics · 7db61f38
      Ian Rogers authored
      Remove over escaping with \\.
      
      Fixes: fd550098 (perf vendor events intel: Update metrics from TMAM 3.5)
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-4-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      7db61f38
    • Ian Rogers's avatar
      perf metrics: Fix parse errors in cascade lake metrics · 92aa1c2b
      Ian Rogers authored
      Remove over escaping with \\.
      Remove extraneous if 1 if 0 == 1 else 0 else 0.
      
      Fixes: fd550098 (perf vendor events intel: Update metrics from TMAM 3.5)
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@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: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kajol Jain <kjain@linux.ibm.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-3-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      92aa1c2b
    • Ian Rogers's avatar
      perf expr: Allow for unlimited escaped characters in a symbol · 5b3141d0
      Ian Rogers authored
      Current expression allows 2 escaped '-,=' characters. However, some
      metrics require more, for example Haswell DRAM_BW_Use.
      
      Fixes: 26226a97 (perf expr: Move expr lexer to flex)
      Signed-off-by: default avatarIan Rogers <irogers@google.com>
      Acked-by: default avatarJiri Olsa <jolsa@redhat.com>
      Tested-by: default avatarKajol Jain <kjain@linux.ibm.com>
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Andi Kleen <ak@linux.intel.com>
      Cc: Haiyan Song <haiyanx.song@intel.com>
      Cc: Jin Yao <yao.jin@linux.intel.com>
      Cc: John Garry <john.garry@huawei.com>
      Cc: Kan Liang <kan.liang@linux.intel.com>
      Cc: Leo Yan <leo.yan@linaro.org>
      Cc: Mark Rutland <mark.rutland@arm.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Clarke <pc@us.ibm.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
      Cc: Song Liu <songliubraving@fb.com>
      Cc: Stephane Eranian <eranian@google.com>
      Link: http://lore.kernel.org/lkml/20200501173333.227162-2-irogers@google.comSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      5b3141d0
    • Jiri Olsa's avatar
      perf script: Enable IP fields for callchains · 53fb1894
      Jiri Olsa authored
      In case the callchains were deleted in pipe mode, we need to ensure that
      the IP fields are enabled, otherwise the callchain is not displayed.
      
      Enabling IP and SYM, which should be enough for callchains.
      
      Committer testing:
      
      Before:
      
      Committer Testing:
      
      before:
      
        # ls
        # perf record -g -e 'syscalls:*' sleep 0.1 2>/dev/null | perf script | tail
             sleep 5677 [0] 5034.295882:         syscalls:sys_exit_mmap: 0x7fcbcfa74000
             sleep 5677 [0] 5034.295885:       syscalls:sys_enter_close: fd: 0x00000003
             sleep 5677 [0] 5034.295886:        syscalls:sys_exit_close: 0x0
             sleep 5677 [0] 5034.295911:   syscalls:sys_enter_nanosleep: rqtp: 0x7fff775b33a0, rmtp: 0x00000000
             sleep 5677 [0] 5034.396021:    syscalls:sys_exit_nanosleep: 0x0
             sleep 5677 [0] 5034.396027:       syscalls:sys_enter_close: fd: 0x00000001
             sleep 5677 [0] 5034.396028:        syscalls:sys_exit_close: 0x0
             sleep 5677 [0] 5034.396029:       syscalls:sys_enter_close: fd: 0x00000002
             sleep 5677 [0] 5034.396029:        syscalls:sys_exit_close: 0x0
             sleep 5677 [0] 5034.396032:  syscalls:sys_enter_exit_group: error_code: 0x00000000
        #
        # ls
        #
      
      After:
      
        # perf record --call-graph=dwarf -e 'syscalls:sys_enter*' sleep 0.1 2>/dev/null | perf script | tail -37
        sleep 33010 [000]  5400.625269:              syscalls:sys_enter_nanosleep: rqtp: 0x7fff2d0e7860, rmtp: 0x00000000
        	    7f1406f131a7 __GI___nanosleep (inlined)
        	    561c4f996966 [unknown]
        	    561c4f99673f [unknown]
        	    561c4f9937af [unknown]
        	    7f1406e6c1a2 __libc_start_main
        	    561c4f99388d [unknown]
      
        sleep 33010 [000]  5400.725391:                  syscalls:sys_enter_close: fd: 0x00000001
        	    7f1406f3c3cb __GI___close_nocancel (inlined)
        	    7f1406ec7d6f _IO_new_file_close_it (inlined)
        	    7f1406ebafa5 _IO_new_fclose (inlined)
        	    561c4f996a40 [unknown]
        	    561c4f993d79 [unknown]
        	    7f1406e83e86 __run_exit_handlers
        	    7f1406e8403f __GI_exit (inlined)
        	    7f1406e6c1a9 __libc_start_main
        	    561c4f99388d [unknown]
      
        sleep 33010 [000]  5400.725395:                  syscalls:sys_enter_close: fd: 0x00000002
        	    7f1406f3c3cb __GI___close_nocancel (inlined)
        	    7f1406ec7d6f _IO_new_file_close_it (inlined)
        	    7f1406ebafa5 _IO_new_fclose (inlined)
        	    561c4f996a40 [unknown]
        	    561c4f993da2 [unknown]
        	    7f1406e83e86 __run_exit_handlers
        	    7f1406e8403f __GI_exit (inlined)
        	    7f1406e6c1a9 __libc_start_main
        	    561c4f99388d [unknown]
      
        sleep 33010 [000]  5400.725399:             syscalls:sys_enter_exit_group: error_code: 0x00000000
        	    7f1406f13466 __GI__exit (inlined)
        	    7f1406e83fa1 __run_exit_handlers
        	    7f1406e8403f __GI_exit (inlined)
        	    7f1406e6c1a9 __libc_start_main
        	    561c4f99388d [unknown]
        #
      
      And, if we install coreutils-debuginfo, we'll have those [unknown] resolved,
      those are for the /usr/bin/sleep binary, use:
      
        # dnf debuginfo-install coreutils
      
      On Fedora and derivatives, then:
      
        # perf record --call-graph=dwarf -e 'syscalls:sys_enter*' sleep 0.1 2>/dev/null | perf script | tail -37
        sleep 33046 [009]  5533.910074:              syscalls:sys_enter_nanosleep: rqtp: 0x7ffea6fa7ab0, rmtp: 0x00000000
        	    7f5f786e81a7 __GI___nanosleep (inlined)
        	    564472454966 rpl_nanosleep
        	    56447245473f xnanosleep
        	    5644724517af main
        	    7f5f786411a2 __libc_start_main
        	    56447245188d _start
      
        sleep 33046 [009]  5534.010218:                  syscalls:sys_enter_close: fd: 0x00000001
        	    7f5f787113cb __GI___close_nocancel (inlined)
        	    7f5f7869cd6f _IO_new_file_close_it (inlined)
        	    7f5f7868ffa5 _IO_new_fclose (inlined)
        	    564472454a40 close_stream
        	    564472451d79 close_stdout
        	    7f5f78658e86 __run_exit_handlers
        	    7f5f7865903f __GI_exit (inlined)
        	    7f5f786411a9 __libc_start_main
        	    56447245188d _start
      
        sleep 33046 [009]  5534.010224:                  syscalls:sys_enter_close: fd: 0x00000002
        	    7f5f787113cb __GI___close_nocancel (inlined)
        	    7f5f7869cd6f _IO_new_file_close_it (inlined)
        	    7f5f7868ffa5 _IO_new_fclose (inlined)
        	    564472454a40 close_stream
        	    564472451da2 close_stdout
        	    7f5f78658e86 __run_exit_handlers
        	    7f5f7865903f __GI_exit (inlined)
        	    7f5f786411a9 __libc_start_main
        	    56447245188d _start
      
        sleep 33046 [009]  5534.010229:             syscalls:sys_enter_exit_group: error_code: 0x00000000
        	    7f5f786e8466 __GI__exit (inlined)
        	    7f5f78658fa1 __run_exit_handlers
        	    7f5f7865903f __GI_exit (inlined)
        	    7f5f786411a9 __libc_start_main
        	    56447245188d _start
      
        #
      Reported-by: default avatarPaul Khuong <pvk@pvk.ca>
      Signed-off-by: default avatarJiri Olsa <jolsa@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Ian Rogers <irogers@google.com>
      Cc: Michael Petlan <mpetlan@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200507095024.2789147-6-jolsa@kernel.orgSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      53fb1894
    • Jiri Olsa's avatar
      perf callchain: Setup callchain properly in pipe mode · 0d71a2b2
      Jiri Olsa authored
      Callchains are automatically initialized by checking on event's
      sample_type. For pipe mode we need to put this check into attr event
      code.
      
      Moving the callchains setup code into callchain_param_setup function and
      calling it from attr event process code.
      
      This enables pipe output having callchains, like:
      
        # perf record -g -e 'raw_syscalls:sys_enter' true | perf script
        # perf record -g -e 'raw_syscalls:sys_enter' true | perf report
      
      Committer notes:
      
      We still need the next patch for the above output to work.
      Reported-by: default avatarPaul Khuong <pvk@pvk.ca>
      Signed-off-by: default avatarJiri Olsa <jolsa@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Ian Rogers <irogers@google.com>
      Cc: Michael Petlan <mpetlan@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200507095024.2789147-5-jolsa@kernel.orgSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      0d71a2b2
    • Jiri Olsa's avatar
      perf session: Try to read pipe data from file · 14d3d540
      Jiri Olsa authored
      Ian came with the idea of having support to read the pipe data also from
      file. Currently pipe mode files fail like:
      
        $ perf record -o - sleep 1 > /tmp/perf.pipe.data
        $ perf report -i /tmp/perf.pipe.data
        incompatible file format (rerun with -v to learn more)
      
      This patch adds the support to do that by trying the pipe header first,
      and if its successfully detected, switching the perf data to pipe mode.
      
      Committer testing:
      
        # ls
        # perf record -a -o - sleep 1 > /tmp/perf.pipe.data
        [ perf record: Woken up 1 times to write data ]
        [ perf record: Captured and wrote 0.000 MB - ]
        # ls
        # perf report -i /tmp/perf.pipe.data | head -25
        # To display the perf.data header info, please use --header/--header-only options.
        #
        #
        # Total Lost Samples: 0
        #
        # Samples: 511  of event 'cycles'
        # Event count (approx.): 178447276
        #
        # Overhead  Command   Shared Object      Symbol
        # ........  ........  .................  ...........................................................................................
        #
            65.49%  swapper   [kernel.kallsyms]  [k] native_safe_halt
             6.45%  chromium  libblink_core.so   [.] blink::SelectorChecker::CheckOne
             4.08%  chromium  libblink_core.so   [.] blink::SelectorQuery::ExecuteForTraverseRoot<blink::AllElementsSelectorQueryTrait>
             2.25%  chromium  libblink_core.so   [.] blink::SelectorQuery::FindTraverseRootsAndExecute<blink::AllElementsSelectorQueryTrait>
             2.11%  chromium  libblink_core.so   [.] blink::SelectorChecker::MatchSelector
             1.91%  chromium  libblink_core.so   [.] blink::Node::OwnerShadowHost
             1.31%  chromium  libblink_core.so   [.] blink::Node::parentNode@plt
             1.22%  chromium  libblink_core.so   [.] blink::Node::parentNode
             0.59%  chromium  libblink_core.so   [.] blink::AnyAttributeMatches
             0.58%  chromium  libv8.so           [.] v8::internal::GlobalHandles::Create
             0.58%  chromium  libblink_core.so   [.] blink::NodeTraversal::NextAncestorSibling
             0.55%  chromium  libv8.so           [.] v8::internal::RegExpGlobalCache::RegExpGlobalCache
             0.55%  chromium  libblink_core.so   [.] blink::Node::ContainingShadowRoot
             0.55%  chromium  libblink_core.so   [.] blink::NodeTraversal::NextAncestorSibling@plt
        #
      Original-patch-by: default avatarIan Rogers <irogers@google.com>
      Signed-off-by: default avatarJiri Olsa <jolsa@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Ian Rogers <irogers@google.com>
      Cc: Michael Petlan <mpetlan@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Khuong <pvk@pvk.ca>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200507095024.2789147-4-jolsa@kernel.orgSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      14d3d540
    • Jiri Olsa's avatar
      perf tools: Do not seek in pipe fd during tracing data processing · b491198d
      Jiri Olsa authored
      There's no need to set 'fd' position in pipe mode, the file descriptor
      is already in proper place. Moreover the lseek will fail on pipe
      descriptor and that's why it's been working properly.
      
      I was tempted to remove the lseek calls completely, because it seems
      that tracing data event was always synthesized only in pipe mode, so
      there's no need for 'file' mode handling. But I guess there was a reason
      behind this and there might (however unlikely) be a perf.data that we
      could break processing for.
      Signed-off-by: default avatarJiri Olsa <jolsa@kernel.org>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Ian Rogers <irogers@google.com>
      Cc: Michael Petlan <mpetlan@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Khuong <pvk@pvk.ca>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200507095024.2789147-3-jolsa@kernel.orgSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      b491198d
    • Jiri Olsa's avatar
      perf tools: Do not display extra info when there is nothing to build · fdb071f8
      Jiri Olsa authored
      Even with fully built tree, we still display extra output when make is
      invoked, like:
      
        $ make
          BUILD:   Doing 'make -j8' parallel build
          DESCEND  plugins
        make[3]: Nothing to be done for 'plugins/libtraceevent-dynamic-list'.
      
      Changing the make descend directly to plugins directory, which quiets
      those messages down.
      Reported-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Signed-off-by: default avatarJiri Olsa <jolsa@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
      Cc: Ian Rogers <irogers@google.com>
      Cc: Michael Petlan <mpetlan@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: Paul Khuong <pvk@pvk.ca>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Link: http://lore.kernel.org/lkml/20200507095024.2789147-2-jolsa@kernel.orgSigned-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      fdb071f8
    • Masami Hiramatsu's avatar
      perf probe: Do not show the skipped events · f41ebe9d
      Masami Hiramatsu authored
      When a probe point is expanded to several places (like inlined) and if
      some of them are skipped because of blacklisted or __init function,
      those trace_events has no event name. It must be skipped while showing
      results.
      
      Without this fix, you can see "(null):(null)" on the list,
      
        # ./perf probe request_resource
        reserve_setup is out of .text, skip it.
        Added new events:
          (null):(null)        (on request_resource)
          probe:request_resource (on request_resource)
      
        You can now use it in all perf tools, such as:
      
        	perf record -e probe:request_resource -aR sleep 1
      
        #
      
      With this fix, it is ignored:
      
        # ./perf probe request_resource
        reserve_setup is out of .text, skip it.
        Added new events:
          probe:request_resource (on request_resource)
      
        You can now use it in all perf tools, such as:
      
        	perf record -e probe:request_resource -aR sleep 1
      
        #
      
      Fixes: 5a51fcd1 ("perf probe: Skip kernel symbols which is out of .text")
      Signed-off-by: default avatarMasami Hiramatsu <mhiramat@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: stable@vger.kernel.org
      Link: http://lore.kernel.org/lkml/158763968263.30755.12800484151476026340.stgit@devnote2Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      f41ebe9d
    • Masami Hiramatsu's avatar
      perf probe: Check address correctness by map instead of _etext · 2ae5d0d7
      Masami Hiramatsu authored
      Since commit 03db8b58 ("perf tools: Fix
      maps__find_symbol_by_name()") introduced map address range check in
      maps__find_symbol_by_name(), we can not get "_etext" from kernel map
      because _etext is placed on the edge of the kernel .text section (=
      kernel map in perf.)
      
      To fix this issue, this checks the address correctness by map address
      range information (map->start and map->end) instead of using _etext
      address.
      
      This can cause an error if the target inlined function is embedded in
      both __init function and normal function.
      
      For exaample, request_resource() is a normal function but also embedded
      in __init reserve_setup(). In this case, the probe point in
      reserve_setup() must be skipped.
      
      However, without this fix, it failes to setup all probe points:
      
        # ./perf probe -v request_resource
        probe-definition(0): request_resource
        symbol:request_resource file:(null) line:0 offset:0 return:0 lazy:(null)
        0 arguments
        Looking at the vmlinux_path (8 entries long)
        Using /usr/lib/debug/lib/modules/5.5.17-200.fc31.x86_64/vmlinux for symbols
        Open Debuginfo file: /usr/lib/debug/lib/modules/5.5.17-200.fc31.x86_64/vmlinux
        Try to find probe point from debuginfo.
        Matched function: request_resource [15e29ad]
        found inline addr: 0xffffffff82fbf892
        Probe point found: reserve_setup+204
        found inline addr: 0xffffffff810e9790
        Probe point found: request_resource+0
        Found 2 probe_trace_events.
        Opening /sys/kernel/debug/tracing//kprobe_events write=1
        Opening /sys/kernel/debug/tracing//README write=0
        Writing event: p:probe/request_resource _text+33290386
        Failed to write event: Invalid argument
          Error: Failed to add events. Reason: Invalid argument (Code: -22)
        #
      
      With this fix,
      
        # ./perf probe request_resource
        reserve_setup is out of .text, skip it.
        Added new events:
          (null):(null)        (on request_resource)
          probe:request_resource (on request_resource)
      
        You can now use it in all perf tools, such as:
      
        	perf record -e probe:request_resource -aR sleep 1
      
        #
      
      Fixes: 03db8b58 ("perf tools: Fix maps__find_symbol_by_name()")
      Reported-by: default avatarArnaldo Carvalho de Melo <acme@kernel.org>
      Signed-off-by: default avatarMasami Hiramatsu <mhiramat@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: stable@vger.kernel.org
      Link: http://lore.kernel.org/lkml/158763967332.30755.4922496724365529088.stgit@devnote2Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      2ae5d0d7
    • Masami Hiramatsu's avatar
      perf probe: Fix to check blacklist address correctly · 80526491
      Masami Hiramatsu authored
      Fix to check kprobe blacklist address correctly with relocated address
      by adjusting debuginfo address.
      
      Since the address in the debuginfo is same as objdump, it is different
      from relocated kernel address with KASLR.  Thus, 'perf probe' always
      misses to catch the blacklisted addresses.
      
      Without this patch, 'perf probe' can not detect the blacklist addresses
      on a KASLR enabled kernel.
      
        # perf probe kprobe_dispatcher
        Failed to write event: Invalid argument
          Error: Failed to add events.
        #
      
      With this patch, it correctly shows the error message.
      
        # perf probe kprobe_dispatcher
        kprobe_dispatcher is blacklisted function, skip it.
        Probe point 'kprobe_dispatcher' not found.
          Error: Failed to add events.
        #
      
      Fixes: 9aaf5a5f ("perf probe: Check kprobes blacklist when adding new events")
      Signed-off-by: default avatarMasami Hiramatsu <mhiramat@kernel.org>
      Tested-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: stable@vger.kernel.org
      Link: http://lore.kernel.org/lkml/158763966411.30755.5882376357738273695.stgit@devnote2Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      80526491
    • Masami Hiramatsu's avatar
      perf probe: Accept the instance number of kretprobe event · c6aab66a
      Masami Hiramatsu authored
      Since the commit 6a13a0d7 ("ftrace/kprobe: Show the maxactive number
      on kprobe_events") introduced to show the instance number of kretprobe
      events, the length of the 1st format of the kprobe event will not 1, but
      it can be longer.  This caused a parser error in perf-probe.
      
      Skip the length check the 1st format of the kprobe event to accept this
      instance number.
      
      Without this fix:
      
        # perf probe -a vfs_read%return
        Added new event:
          probe:vfs_read__return (on vfs_read%return)
      
        You can now use it in all perf tools, such as:
      
        	perf record -e probe:vfs_read__return -aR sleep 1
      
        # perf probe -l
        Semantic error :Failed to parse event name: r16:probe/vfs_read__return
          Error: Failed to show event list.
      
      And with this fixes:
      
        # perf probe -a vfs_read%return
        ...
        # perf probe -l
          probe:vfs_read__return (on vfs_read%return)
      
      Fixes: 6a13a0d7 ("ftrace/kprobe: Show the maxactive number on kprobe_events")
      Reported-by: default avatarYuxuan Shui <yshuiv7@gmail.com>
      Signed-off-by: default avatarMasami Hiramatsu <mhiramat@kernel.org>
      Tested-by: default avatarYuxuan Shui <yshuiv7@gmail.com>
      Cc: Jiri Olsa <jolsa@redhat.com>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Cc: stable@vger.kernel.org
      Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=207587
      Link: http://lore.kernel.org/lkml/158877535215.26469.1113127926699134067.stgit@devnote2Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      c6aab66a
    • Arnaldo Carvalho de Melo's avatar
      perf counts: Rename perf_evsel__*counts() to evsel__*counts() · 7d1e239e
      Arnaldo Carvalho de Melo authored
      As these are 'struct evsel' methods, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      7d1e239e
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Rename perf_evsel__[hs]w_cache* to evsel__[hs]w_cache* · c64e85e1
      Arnaldo Carvalho de Melo authored
      As those are 'struct evsel' methods, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      c64e85e1
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Rename perf_evsel__new*() to evsel__new*() · 8f6725a2
      Arnaldo Carvalho de Melo authored
      As these are 'struct evsel' methods, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      8f6725a2
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Rename *perf_evsel__get_config_term() & friends to evsel__env() · 35ac0cad
      Arnaldo Carvalho de Melo authored
      As it is a 'struct evsel' method, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      35ac0cad
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Rename perf_evsel__fprintf() to evsel__fprintf() · 2dbfc945
      Arnaldo Carvalho de Melo authored
      As it is a 'struct evsel' method, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      2dbfc945
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Rename perf_evsel__resort*() to evsel__resort*() · 10c513f7
      Arnaldo Carvalho de Melo authored
      As it is a 'struct evsel' method, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      10c513f7
    • Arnaldo Carvalho de Melo's avatar
      perf evsel: Rename perf_evsel__object_config() to evsel__object_config() · 4c703828
      Arnaldo Carvalho de Melo authored
      As it is a 'struct evsel' method, not part of tools/lib/perf/, aka
      libperf, to whom the perf_ prefix belongs.
      
      Cc: Adrian Hunter <adrian.hunter@intel.com>
      Cc: Jiri Olsa <jolsa@kernel.org>
      Cc: Namhyung Kim <namhyung@kernel.org>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      4c703828