linux.git/tools/perf/util, branch v6.2-rc3

perf tools: Fix build on uClibc systems by adding missing sys/types.h include

2023-01-04T19:44:01+00:00

Not all libc implementations define ssize_t as part of stdio.h like
glibc does since the standard only requires this type to be defined by
unistd.h and sys/types.h. For this reason the perf build is currently
broken for toolchains based on uClibc, for instance.

Include sys/types.h explicitly to fix that.

Committer notes:

In addition, in the past this worked in uClibc test systems as there was
another way to get to sys/types.h that got removed in that cset:

  tools/perf/util/trace-event.h
    /usr/include/traceevent/event_parse.h # This got removed from util/trace-event.h in 378ef0f5d9d7f465
      /usr/include/regex.h
        /usr/include/sys/types.h
          typedef __ssize_t ssize_t;

So the size_t that is used in tools/perf/util/trace-event.h was being
obtained indirectly, by chance.

Fixes: 378ef0f5d9d7f465 ("perf build: Use libtraceevent from the system")
Signed-off-by: Jesus Sanchez-Palencia 
Cc: Alexander Shishkin 
Cc: Ian Rogers 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/20230104193414.606905-1-jesussanp@google.com
Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Fix handling of --for-each-cgroup with --bpf-counters to match non BPF mode

2023-01-04T14:11:41+00:00

The --for-each-cgroup can have the same cgroup multiple times, but this
confuses BPF counters (since they have the same cgroup id), making only
the last cgroup events to be counted.

Let's check the cgroup name before adding a new entry to the cgroups
list.

Before:

  $ sudo ./perf stat -a --bpf-counters --for-each-cgroup /,/ sleep 1

   Performance counter stats for 'system wide':

        msec cpu-clock                        /
             context-switches                 /
             cpu-migrations                   /
             page-faults                      /
             cycles                           /
             instructions                     /
             branches                         /
             branch-misses                    /
            8,016.04 msec cpu-clock                        /                #    7.998 CPUs utilized
               6,152      context-switches                 /                #  767.461 /sec
                 250      cpu-migrations                   /                #   31.187 /sec
                 442      page-faults                      /                #   55.139 /sec
         613,111,487      cycles                           /                #    0.076 GHz
         280,599,604      instructions                     /                #    0.46  insn per cycle
          57,692,724      branches                         /                #    7.197 M/sec
           3,385,168      branch-misses                    /                #    5.87% of all branches

         1.002220125 seconds time elapsed

After it becomes similar to the non-BPF mode:

  $ sudo ./perf stat -a --bpf-counters --for-each-cgroup /,/  sleep 1

   Performance counter stats for 'system wide':

            8,013.38 msec cpu-clock                        /                #    7.998 CPUs utilized
               6,859      context-switches                 /                #  855.944 /sec
                 334      cpu-migrations                   /                #   41.680 /sec
                 345      page-faults                      /                #   43.053 /sec
         782,326,119      cycles                           /                #    0.098 GHz
         471,645,724      instructions                     /                #    0.60  insn per cycle
          94,963,430      branches                         /                #   11.851 M/sec
           3,685,511      branch-misses                    /                #    3.88% of all branches

         1.001864539 seconds time elapsed

Committer notes:

As a reminder, to test with BPF counters one has to use BUILD_BPF_SKEL=1
in the make command line and have clang/llvm installed when building
perf, otherwise the --bpf-counters option will not be available:

  # perf stat -a --bpf-counters --for-each-cgroup /,/ sleep 1
  Error: unknown option `bpf-counters'

   Usage: perf stat [] []

      -a, --all-cpus        system-wide collection from all CPUs
  
  #

Fixes: bb1c15b60b981d10 ("perf stat: Support regex pattern in --for-each-cgroup")
Signed-off-by: Namhyung Kim 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Adrian Hunter 
Cc: bpf@vger.kernel.org
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Song Liu 
Link: https://lore.kernel.org/r/20230104064402.1551516-5-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Fix handling of unsupported cgroup events when using BPF counters

2023-01-04T13:52:07+00:00

When --for-each-cgroup option is used, it fails when any of events is
not supported and exits immediately.  This is not how 'perf stat'
handles unsupported events.

Let's ignore the failure and proceed with others so that the output is
similar to when BPF counters are not used:

Before:

  $ sudo ./perf stat -a --bpf-counters -e L1-icache-loads,L1-dcache-loads --for-each-cgroup system.slice,user.slice sleep 1
  Failed to open first cgroup events
  $

After it shows output similat to when --bpf-counters isn't specified:

  $ sudo ./perf stat -a --bpf-counters -e L1-icache-loads,L1-dcache-loads --for-each-cgroup system.slice,user.slice sleep 1

   Performance counter stats for 'system wide':

           L1-icache-loads                  system.slice
          29,892,418      L1-dcache-loads                  system.slice
           L1-icache-loads                  user.slice
          52,497,220      L1-dcache-loads                  user.slice
  $

Fixes: 944138f048f7d759 ("perf stat: Enable BPF counter with --for-each-cgroup")
Signed-off-by: Namhyung Kim 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Adrian Hunter 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Peter Zijlstra 
Cc: Song Liu 
Link: https://lore.kernel.org/r/20230104064402.1551516-4-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix segfault when trying to process tracepoints in perf.data and not linked with libtraceevent

2023-01-02T16:45:19+00:00

When we have a perf.data file with tracepoints, such as:

  # perf evlist -f
  probe_perf:lzma_decompress_to_file
  # Tip: use 'perf evlist --trace-fields' to show fields for tracepoint events
  #

We end up segfaulting when using perf built with NO_LIBTRACEEVENT=1 by
trying to find an evsel with a NULL 'event_name' variable:

  (gdb) run report --stdio -f
  Starting program: /root/bin/perf report --stdio -f

  Program received signal SIGSEGV, Segmentation fault.
  0x000000000055219d in find_evsel (evlist=0xfda7b0, event_name=0x0) at util/sort.c:2830
  warning: Source file is more recent than executable.
  2830		if (event_name[0] == '%') {
  Missing separate debuginfos, use: dnf debuginfo-install bzip2-libs-1.0.8-11.fc36.x86_64 cyrus-sasl-lib-2.1.27-18.fc36.x86_64 elfutils-debuginfod-client-0.188-3.fc36.x86_64 elfutils-libelf-0.188-3.fc36.x86_64 elfutils-libs-0.188-3.fc36.x86_64 glibc-2.35-20.fc36.x86_64 keyutils-libs-1.6.1-4.fc36.x86_64 krb5-libs-1.19.2-12.fc36.x86_64 libbrotli-1.0.9-7.fc36.x86_64 libcap-2.48-4.fc36.x86_64 libcom_err-1.46.5-2.fc36.x86_64 libcurl-7.82.0-12.fc36.x86_64 libevent-2.1.12-6.fc36.x86_64 libgcc-12.2.1-4.fc36.x86_64 libidn2-2.3.4-1.fc36.x86_64 libnghttp2-1.51.0-1.fc36.x86_64 libpsl-0.21.1-5.fc36.x86_64 libselinux-3.3-4.fc36.x86_64 libssh-0.9.6-4.fc36.x86_64 libstdc++-12.2.1-4.fc36.x86_64 libunistring-1.0-1.fc36.x86_64 libunwind-1.6.2-2.fc36.x86_64 libxcrypt-4.4.33-4.fc36.x86_64 libzstd-1.5.2-2.fc36.x86_64 numactl-libs-2.0.14-5.fc36.x86_64 opencsd-1.2.0-1.fc36.x86_64 openldap-2.6.3-1.fc36.x86_64 openssl-libs-3.0.5-2.fc36.x86_64 slang-2.3.2-11.fc36.x86_64 xz-libs-5.2.5-9.fc36.x86_64 zlib-1.2.11-33.fc36.x86_64
  (gdb) bt
  #0  0x000000000055219d in find_evsel (evlist=0xfda7b0, event_name=0x0) at util/sort.c:2830
  #1  0x0000000000552416 in add_dynamic_entry (evlist=0xfda7b0, tok=0xffb6eb "trace", level=2) at util/sort.c:2976
  #2  0x0000000000552d26 in sort_dimension__add (list=0xf93e00 , tok=0xffb6eb "trace", evlist=0xfda7b0, level=2) at util/sort.c:3193
  #3  0x0000000000552e1c in setup_sort_list (list=0xf93e00 , str=0xffb6eb "trace", evlist=0xfda7b0) at util/sort.c:3227
  #4  0x00000000005532fa in __setup_sorting (evlist=0xfda7b0) at util/sort.c:3381
  #5  0x0000000000553cdc in setup_sorting (evlist=0xfda7b0) at util/sort.c:3608
  #6  0x000000000042eb9f in cmd_report (argc=0, argv=0x7fffffffe470) at builtin-report.c:1596
  #7  0x00000000004aee7e in run_builtin (p=0xf64ca0 , argc=3, argv=0x7fffffffe470) at perf.c:330
  #8  0x00000000004af0f2 in handle_internal_command (argc=3, argv=0x7fffffffe470) at perf.c:384
  #9  0x00000000004af241 in run_argv (argcp=0x7fffffffe29c, argv=0x7fffffffe290) at perf.c:428
  #10 0x00000000004af5fc in main (argc=3, argv=0x7fffffffe470) at perf.c:562
  (gdb)

So check if we have tracepoint events in add_dynamic_entry() and bail
out instead:

  # perf report --stdio -f
  This perf binary isn't linked with libtraceevent, can't process probe_perf:lzma_decompress_to_file
  Error:
  Unknown --sort key: `trace'
  #

Fixes: 378ef0f5d9d7f465 ("perf build: Use libtraceevent from the system")
Acked-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Link: http://lore.kernel.org/lkml/Y7MDb7kRaHZB6APC@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Don't include signature in version strings

2023-01-02T15:34:06+00:00

This explodes the build if HEAD is signed, since the generated version
is gpg: Signature made Mon 26 Dec 2022 20:34:48 CET, then a few more
lines, then the SHA.

Signed-off-by: Ahelenia Ziemiańska 
Acked-by: Namhyung Kim 
Cc: Alexander Shishkin 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Peter Zijlstra 
Link: http://lore.kernel.org/lkml/7c9637711271f50ec2341fb8a7c29585335dab04.1672174189.git.nabijaczleweli@nabijaczleweli.xyz
Signed-off-by: Arnaldo Carvalho de Melo

perf help: Use HAVE_LIBTRACEEVENT to filter out unsupported commands

2023-01-02T14:51:53+00:00

Commands such as kmem, kwork, lock, sched, trace and timechart depend on
libtraceevent, these commands need to be isolated using HAVE_LIBTRACEEVENT
macro when cmdlist generation.

The output of the generate-cmdlist.sh script is as follows:

  # ./util/generate-cmdlist.sh
  /* Automatically generated by ./util/generate-cmdlist.sh */
  struct cmdname_help
  {
      char name[16];
      char help[80];
  };

  static struct cmdname_help common_cmds[] = {
    {"annotate", "Read perf.data (created by perf record) and display annotated code"},
    {"archive", "Create archive with object files with build-ids found in perf.data file"},
    {"bench", "General framework for benchmark suites"},
    {"buildid-cache", "Manage build-id cache."},
    {"buildid-list", "List the buildids in a perf.data file"},
    {"c2c", "Shared Data C2C/HITM Analyzer."},
    {"config", "Get and set variables in a configuration file."},
    {"daemon", "Run record sessions on background"},
    {"data", "Data file related processing"},
    {"diff", "Read perf.data files and display the differential profile"},
    {"evlist", "List the event names in a perf.data file"},
    {"ftrace", "simple wrapper for kernel's ftrace functionality"},
    {"inject", "Filter to augment the events stream with additional information"},
    {"iostat", "Show I/O performance metrics"},
    {"kallsyms", "Searches running kernel for symbols"},
    {"kvm", "Tool to trace/measure kvm guest os"},
    {"list", "List all symbolic event types"},
    {"mem", "Profile memory accesses"},
    {"record", "Run a command and record its profile into perf.data"},
    {"report", "Read perf.data (created by perf record) and display the profile"},
    {"script", "Read perf.data (created by perf record) and display trace output"},
    {"stat", "Run a command and gather performance counter statistics"},
    {"test", "Runs sanity tests."},
    {"top", "System profiling tool."},
    {"version", "display the version of perf binary"},
  #ifdef HAVE_LIBELF_SUPPORT
    {"probe", "Define new dynamic tracepoints"},
  #endif /* HAVE_LIBELF_SUPPORT */
  #if defined(HAVE_LIBTRACEEVENT) && (defined(HAVE_LIBAUDIT_SUPPORT) || defined(HAVE_SYSCALL_TABLE_SUPPORT))
    {"trace", "strace inspired tool"},
  #endif /* HAVE_LIBTRACEEVENT && (HAVE_LIBAUDIT_SUPPORT || HAVE_SYSCALL_TABLE_SUPPORT) */
  #ifdef HAVE_LIBTRACEEVENT
    {"kmem", "Tool to trace/measure kernel memory properties"},
    {"kwork", "Tool to trace/measure kernel work properties (latencies)"},
    {"lock", "Analyze lock events"},
    {"sched", "Tool to trace/measure scheduler properties (latencies)"},
    {"timechart", "Tool to visualize total system behavior during a workload"},
  #endif /* HAVE_LIBTRACEEVENT */
  };

Fixes: 378ef0f5d9d7f465 ("perf build: Use libtraceevent from the system")
Signed-off-by: Yang Jihong 
Acked-by: Namhyung Kim 
Cc: Alexander Shishkin 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Peter Zijlstra 
Link: https://lore.kernel.org/r/20221226085703.95081-1-yangjihong1@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix resources leak in perf_data__open_dir()

2023-01-02T14:45:43+00:00

In perf_data__open_dir(), opendir() opens the directory stream.  Add
missing closedir() to release it after use.

Fixes: eb6176709b235b96 ("perf data: Add perf_data__open_dir_data function")
Reviewed-by: Adrian Hunter 
Signed-off-by: Miaoqian Lin 
Cc: Alexander Shishkin 
Cc: Alexey Bayduraev 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: https://lore.kernel.org/r/20221229090903.1402395-1-linmq006@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo

perf python: Fix splitting CC into compiler and options

2022-12-22T14:34:30+00:00

Noticed this build failure on archlinux:base when building with clang:

  clang-14: error: optimization flag '-ffat-lto-objects' is not supported [-Werror,-Wignored-optimization-argument]

In tools/perf/util/setup.py we check if clang supports that option, but
since commit 3cad53a6f9cdbafa ("perf python: Account for multiple words
in CC") this got broken as in the common case where CC="clang":

  >>> cc="clang"
  >>> print(cc.split()[0])
  clang
  >>> option="-ffat-lto-objects"
  >>> print(str(cc.split()[1:]) + option)
  []-ffat-lto-objects
  >>>

And then the Popen will call clang with that bogus option name that in
turn will not produce the b"unknown argument" or b"is not supported"
that this function uses to detect if the option is not available and
thus later on clang will be called with an unknown/unsupported option.

Fix it by looking if really there are options in the provided CC
variable, and if so override 'cc' with the first token and append the
options to the 'option' variable.

Fixes: 3cad53a6f9cdbafa ("perf python: Account for multiple words in CC")
Cc: Adrian Hunter 
Cc: Fangrui Song 
Cc: Florian Fainelli 
Cc: Ian Rogers 
Cc: Jiri Olsa 
Cc: John Keeping 
Cc: Khem Raj 
Cc: Leo Yan 
Cc: Michael Petlan 
Cc: Namhyung Kim 
Cc: Nathan Chancellor 
Cc: Nick Desaulniers 
Cc: Sedat Dilek 
Link: http://lore.kernel.org/lkml/Y6Rq5F5NI0v1QQHM@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf scripting python: Don't be strict at handling libtraceevent enumerations

2022-12-21T20:30:38+00:00

The build was failing on archlinux because it has a newer libtraceevent
that added a new entry to the tep_print_arg_type enum:

    19.72 archlinux:base                : FAIL gcc version 12.2.0 (GCC)
    util/scripting-engines/trace-event-python.c: In function ‘define_event_symbols’:
    util/scripting-engines/trace-event-python.c:281:9: error: enumeration value ‘TEP_PRINT_CPUMASK’ not handled in switch [-Werror=switch-enum]
      281 |         switch (args->type) {
          |         ^~~~~~
    cc1: all warnings being treated as errors

Since we build with distros that have different versions of
libtraceevent and there is no way to easily test if these enum entries
are available, just disable -Werror=switch-enum for that specific
object.

Signed-off-by: Arnaldo Carvalho de Melo

perf hist: Improve srcline_{from,to} sort key performance

2022-12-21T17:52:40+00:00

Likewise, modify ->cmp() callback to compare sample address and map
address.  And add ->collapse() and ->sort() to check the actual
srcfile string.  Also add ->init() to make sure it has the srcfile.

Signed-off-by: Namhyung Kim 
Acked-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Andi Kleen 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Leo Yan 
Cc: Milian Wolff 
Cc: Peter Zijlstra 
Link: https://lore.kernel.org/r/20221215192817.2734573-10-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo