Commit Graph

330 Commits

Author SHA1 Message Date
Arnaldo Carvalho de Melo c54d241b35 perf maps: Rename map_groups.h to maps.h
One more step in the merge of 'struct maps' with 'struct map_groups'.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-9ibtn3vua76f934t7woyf26w@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-26 11:07:46 -03:00
Arnaldo Carvalho de Melo f2eaea09d6 perf map_symbol: Rename ms->mg to ms->maps
One more step on the merge of 'struct maps' with 'struct map_groups'.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-61rra2wg392rhvdgw421wzpt@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-26 11:07:46 -03:00
Arnaldo Carvalho de Melo 79b6bb73f8 perf maps: Merge 'struct maps' with 'struct map_groups'
And pick the shortest name: 'struct maps'.

The split existed because we used to have two groups of maps, one for
functions and one for variables, but that only complicated things,
sometimes we needed to figure out what was at some address and then had
to first try it on the functions group and if that failed, fall back to
the variables one.

That split is long gone, so for quite a while we had only one struct
maps per struct map_groups, simplify things by combining those structs.

First patch is the minimum needed to merge both, follow up patches will
rename 'thread->mg' to 'thread->maps', etc.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-hom6639ro7020o708trhxh59@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-26 11:07:46 -03:00
Arnaldo Carvalho de Melo 94e44b9ca5 perf annotate: Stop using map->groups, use map_symbol->mg instead
These were the last uses of map->groups, next cset will nuke it.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-n3g0foos7l7uxq9nar0zo0vj@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-12 08:20:53 -03:00
Arnaldo Carvalho de Melo d46a4cdf49 pref tools: Make 'struct addr_map_symbol' contain 'struct map_symbol'
So that we pass that substructure around and with it consolidate lots of
functions that receive a (map, symbol) pair and now can receive just a
'struct map_symbol' pointer.

This further paves the way to add 'struct map_groups' to 'struct
map_symbol' so that we can have all we need for annotation so that we
can ditch 'struct map'->groups, i.e. have the map_groups pointer in a
more central place, avoiding the pointer in the 'struct map' that have
tons of instances.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-fs90ttd9q12l7989fo7pw81q@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-12 08:20:53 -03:00
Arnaldo Carvalho de Melo 2975489458 perf annotate: Pass a 'map_symbol' in places receiving a pair of 'map' and 'symbol' pointers
We are already passing things like:

  symbol__annotate(ms->sym, ms->map, ...)

So shorten the signature of such functions to receive the 'map_symbol'
pointer.

This also paves the way to having the 'struct map_groups' pointer in the
'struct map_symbol' so that we can get rid of 'struct map'->groups.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-23yx8v1t41nzpkpi7rdrozww@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-12 08:20:53 -03:00
Arnaldo Carvalho de Melo 9d355b381b perf map_groups: Pass the object to map_groups__find_ams()
We were just passing a map to look for and reuse its map->groups member,
but the idea is that this is going away, as a map can be in multiple
rb_trees when being reused via a map_node, so do as all the other
map_groups methods and pass as its first arg the object being operated
on.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-nmi2pbggqloogwl6vxrvex5a@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-12 08:20:53 -03:00
Ian Rogers 5c65b1c084 perf annotate: Fix heap overflow
Fix expand_tabs that copies the source lines '\0' and then appends
another '\0' at a potentially out of bounds address.

Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.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/20191026035644.217548-1-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-11-07 08:30:18 -03:00
Ingo Molnar 27a0a90d63 Merge tag 'perf-core-for-mingo-5.5-20191021' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

perf trace:

- Add syscall failure stats to -s/--summary and -S/--with-summary, works in
  combination with specifying just a set of syscalls, see below first with
  -s/--summary, then with -S/--with-summary just for the syscalls we saw failing
  with -s:

    # perf trace -s sleep 1

     Summary of events:

     sleep (16218), 80 events, 93.0%

       syscall     calls  errors  total      min      avg      max   stddev
                                  (msec)   (msec)   (msec)   (msec)    (%)
       ----------- -----  ------ -------- -------- -------- -------- ------
       nanosleep       1      0  1000.091 1000.091 1000.091 1000.091  0.00%
       mmap            8      0     0.045    0.005    0.006    0.008  7.09%
       mprotect        4      0     0.028    0.005    0.007    0.009 11.38%
       openat          3      0     0.021    0.005    0.007    0.009 14.07%
       munmap          1      0     0.017    0.017    0.017    0.017  0.00%
       brk             4      0     0.010    0.001    0.002    0.004 23.15%
       read            4      0     0.009    0.002    0.002    0.003  8.13%
       close           5      0     0.008    0.001    0.002    0.002 10.83%
       fstat           3      0     0.006    0.002    0.002    0.002  6.97%
       access          1      1     0.006    0.006    0.006    0.006  0.00%
       lseek           3      0     0.005    0.001    0.002    0.002  7.37%
       arch_prctl      2      1     0.004    0.001    0.002    0.002 17.64%
       execve          1      0     0.000    0.000    0.000    0.000  0.00%

    # perf trace -e access,arch_prctl -S sleep 1
         0.000 ( 0.006 ms): sleep/19503 arch_prctl(option: 0x3001, arg2: 0x7fff165996b0) = -1 EINVAL (Invalid argument)
         0.024 ( 0.006 ms): sleep/19503 access(filename: 0x2177e510, mode: R)            = -1 ENOENT (No such file or directory)
         0.136 ( 0.002 ms): sleep/19503 arch_prctl(option: SET_FS, arg2: 0x7f9421737580) = 0

     Summary of events:

     sleep (19503), 6 events, 50.0%

       syscall    calls  errors total    min    avg    max  stddev
                                (msec) (msec) (msec) (msec)    (%)
       ---------- -----  ------ ------ ------ ------ ------ ------
       arch_prctl   2       1    0.008  0.002  0.004  0.006 57.22%
       access       1       1    0.006  0.006  0.006  0.006  0.00%

    #

  - Introduce --errno-summary, to drill down a bit more in the errno stats:

    # perf trace --errno-summary -e access,arch_prctl -S sleep 1
         0.000 ( 0.006 ms): sleep/5587 arch_prctl(option: 0x3001, arg2: 0x7ffd6ba6aa00) = -1 EINVAL (Invalid argument)
         0.028 ( 0.007 ms): sleep/5587 access(filename: 0xb83d9510, mode: R)            = -1 ENOENT (No such file or directory)
         0.172 ( 0.003 ms): sleep/5587 arch_prctl(option: SET_FS, arg2: 0x7f45b8392580) = 0

     Summary of events:

     sleep (5587), 6 events, 50.0%

       syscall    calls  errors total    min    avg    max  stddev
                                (msec) (msec) (msec) (msec)   (%)
       ---------- -----  ------ ------ ------ ------ ------ ------
       arch_prctl     2     1    0.009  0.003  0.005  0.006 38.90%
			   EINVAL: 1
       access         1     1    0.007  0.007  0.007  0.007  0.00%
                           ENOENT: 1
    #

  - Filter own pid to avoid a feedback look in 'perf trace record -a'

  - Add the glue for the auto generated x86 IRQ vector array.

  - Show error message when not finding a field used in a filter expression

    # perf trace --max-events=4 -e syscalls:sys_enter_write --filter="cnt>32767"
    Failed to set filter "(cnt>32767) && (common_pid != 19938 && common_pid != 8922)" on event syscalls:sys_enter_write with 22 (Invalid argument)
    #
    # perf trace --max-events=4 -e syscalls:sys_enter_write --filter="count>32767"
         0.000 python3.5/17535 syscalls:sys_enter_write(fd: 3, buf: 0x564b0dc53600, count: 172086)
        12.641 python3.5.post/17535 syscalls:sys_enter_write(fd: 3, buf: 0x564b0db63660, count: 75994)
        27.738 python3.5.post/17535 syscalls:sys_enter_write(fd: 3, buf: 0x564b0db4b1e0, count: 41635)
       136.070 python3.5.post/17535 syscalls:sys_enter_write(fd: 3, buf: 0x564b0dbab510, count: 62232)
    #

  - Add a generator for x86's IRQ vectors -> strings

  - Introduce stroul() (string -> number) methods for the strarray and
    strarrays classes, also strtoul_flags, allowing to go from both strings
    and or-ed strings to numbers, allowing things like:

    # perf trace -e syscalls:sys_enter_mmap --filter="flags==DENYWRITE|PRIVATE|FIXED" sleep 1
         0.000 sleep/22588 syscalls:sys_enter_mmap(addr: 0x7f42d2aa5000, len: 1363968, prot: READ|EXEC, flags: PRIVATE|FIXED|DENYWRITE, fd: 3, off: 0x22000)
         0.011 sleep/22588 syscalls:sys_enter_mmap(addr: 0x7f42d2bf2000, len: 311296, prot: READ, flags: PRIVATE|FIXED|DENYWRITE, fd: 3, off: 0x16f000)
         0.015 sleep/22588 syscalls:sys_enter_mmap(addr: 0x7f42d2c3f000, len: 24576, prot: READ|WRITE, flags: PRIVATE|FIXED|DENYWRITE, fd: 3, off: 0x1bb000)
    #

  Allowing to narrow down from the complete set of mmap calls for that workload:

    # perf trace -e syscalls:sys_enter_mmap sleep 1
         0.000 sleep/22695 syscalls:sys_enter_mmap(len: 134773, prot: READ, flags: PRIVATE, fd: 3)
         0.041 sleep/22695 syscalls:sys_enter_mmap(len: 8192, prot: READ|WRITE, flags: PRIVATE|ANONYMOUS)
         0.053 sleep/22695 syscalls:sys_enter_mmap(len: 1857472, prot: READ, flags: PRIVATE|DENYWRITE, fd: 3)
         0.069 sleep/22695 syscalls:sys_enter_mmap(addr: 0x7fd23ffb6000, len: 1363968, prot: READ|EXEC, flags: PRIVATE|FIXED|DENYWRITE, fd: 3, off: 0x22000)
         0.077 sleep/22695 syscalls:sys_enter_mmap(addr: 0x7fd240103000, len: 311296, prot: READ, flags: PRIVATE|FIXED|DENYWRITE, fd: 3, off: 0x16f000)
         0.083 sleep/22695 syscalls:sys_enter_mmap(addr: 0x7fd240150000, len: 24576, prot: READ|WRITE, flags: PRIVATE|FIXED|DENYWRITE, fd: 3, off: 0x1bb000)
         0.095 sleep/22695 syscalls:sys_enter_mmap(addr: 0x7fd240156000, len: 14272, prot: READ|WRITE, flags: PRIVATE|FIXED|ANONYMOUS)
         0.339 sleep/22695 syscalls:sys_enter_mmap(len: 217750512, prot: READ, flags: PRIVATE, fd: 3)
    #

  Works with all targets, so, for system wide, looking at who calls mmap with flags set to just "PRIVATE":

    # perf trace --max-events=5 -e syscalls:sys_enter_mmap --filter="flags==PRIVATE"
         0.000 pool/2242 syscalls:sys_enter_mmap(len: 756, prot: READ, flags: PRIVATE, fd: 14)
         0.050 pool/2242 syscalls:sys_enter_mmap(len: 756, prot: READ, flags: PRIVATE, fd: 14)
         0.062 pool/2242 syscalls:sys_enter_mmap(len: 756, prot: READ, flags: PRIVATE, fd: 14)
         0.145 goa-identity-s/2240 syscalls:sys_enter_mmap(len: 756, prot: READ, flags: PRIVATE, fd: 18)
         0.183 goa-identity-s/2240 syscalls:sys_enter_mmap(len: 756, prot: READ, flags: PRIVATE, fd: 18)
    #

  # perf trace --max-events=2 -e syscalls:sys_enter_lseek --filter="whence==SET && offset != 0"
         0.000 Cache2 I/O/12047 syscalls:sys_enter_lseek(fd: 277, offset: 43, whence: SET)
      1142.070 mozStorage #5/12302 syscalls:sys_enter_lseek(fd: 44</home/acme/.mozilla/firefox/ina67tev.default/cookies.sqlite-wal>, offset: 393536, whence: SET)
  #

perf annotate:

  - Fix objdump --no-show-raw-insn flag to work with goth gcc and clang.

  - Streamline objdump execution, preserving the right error codes for better
    reporting to user.

perf report:

  - Add warning when libunwind not compiled in.

perf stat:

  Jin Yao:

  - Support --all-kernel/--all-user, to match options available in 'perf record',
    asking that all the events specified work just with kernel or user events.

perf list:

  Jin Yao:

  - Hide deprecated events by default, allow showing them with --deprecated.

libbperf:

  Jiri Olsa:

  - Allow to build with -ltcmalloc.

  - Finish mmap interface, getting more stuff from tools/perf while adding
    abstractions to avoid pulling too much stuff, to get libperf to grow as
    tools needs things like auxtrace, etc.

perf scripting engines:

  Steven Rostedt (VMware):

  - Iterate on tep event arrays directly, fixing script generation with
    '-g python' when having multiple tracepoints in a perf.data file.

core:

  - Allow to build with -ltcmalloc.

perf test:

  Leo Yan:

  - Report failure for mmap events.

  - Avoid infinite loop for task exit case.

  - Remove needless headers for bp_account test.

  - Add dedicated checking helper is_supported().

  - Disable bp_signal testing for arm64.

Vendor events:

arm64:

  John Garry:

  - Fix Hisi hip08 DDRC PMU eventname.

  - Add some missing events for Hisi hip08 DDRC, L3C and HHA PMUs.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-10-22 01:15:45 +02:00
Ingo Molnar aa7a7b7297 Merge branch 'perf/urgent' into perf/core, to pick up fixes
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-10-22 01:15:32 +02:00
Gustavo A. R. Silva f948eb45e3 perf annotate: Fix multiple memory and file descriptor leaks
Store SYMBOL_ANNOTATE_ERRNO__BPF_MISSING_BTF in variable *ret*, instead
of returning in the middle of the function and leaking multiple
resources: prog_linfo, btf, s and bfdf.

Addresses-Coverity-ID: 1454832 ("Structurally dead code")
Fixes: 11aad897f6 ("perf annotate: Don't return -1 for error when doing BPF disassembly")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
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/20191014171047.GA30850@embeddedor
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-15 12:00:01 -03:00
Ian Rogers c5baf90892 perf annotate: Fix objdump --no-show-raw-insn flag
Remove redirection of objdump's stderr to /dev/null to help diagnose
failures.

Fix the '--no-show-raw' flag to be '--no-show-raw-insn' which binutils
is permissive and allows, but fails with LLVM objdump.

Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@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: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: clang-built-linux@googlegroups.com
Link: http://lore.kernel.org/lkml/20191010183649.23768-6-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-15 08:39:42 -03:00
Ian Rogers b34b45eef1 perf annotate: Don't pipe objdump output through 'expand' command
Avoiding a pipe allows objdump command failures to surface.  Move to the
caller of symbol__parse_objdump_line the call to strim that removes
leading and trailing tabs.  Add a new expand_tabs function that if a tab
is present allocate a new line in which tabs are expanded.  In
symbol__parse_objdump_line the line had no leading spaces, so simplify
the line_ip processing.

Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@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: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: clang-built-linux@googlegroups.com
Link: http://lore.kernel.org/lkml/20191010183649.23768-5-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-15 08:39:42 -03:00
Ian Rogers 7a675de428 perf annotate: Don't pipe objdump output through 'grep' command
Simplify the objdump command by not piping the output of objdump through
grep. Instead, drop lines that match the grep pattern during the reading
loop.

Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@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: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: clang-built-linux@googlegroups.com
Link: http://lore.kernel.org/lkml/20191010183649.23768-4-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-15 08:39:42 -03:00
Ian Rogers 4235949944 perf annotate: Use libsubcmd's run-command.h to fork objdump
Reduce duplicated logic by using the subcmd library. Ensure when errors
occur they are reported to the caller. Before this patch, if no lines
are read the error status is 0.

Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@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: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: clang-built-linux@googlegroups.com
Link: http://lore.kernel.org/lkml/20191010183649.23768-3-irogers@google.com
Link: http://lore.kernel.org/lkml/20191015003418.62563-1-irogers@google.com
[ merged follow up fix for NULL termination as in the 2nd link above ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-15 08:39:01 -03:00
Ian Rogers 353dcaa2f9 perf annotate: Avoid reallocation in objdump parsing
Objdump output is parsed using getline which allocates memory for the
read. Getline will realloc if the memory is too small, but currently the
line is always freed after the call.

Simplify parse_objdump_line by performing the reading in symbol__disassemble.

Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@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: Song Liu <songliubraving@fb.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: clang-built-linux@googlegroups.com
Link: http://lore.kernel.org/lkml/20191010183649.23768-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-15 08:36:22 -03:00
Jin Yao cebf7d51a6 perf diff: Report noisy for cycles diff
This patch prints the stddev and hist for the cycles diff of program
block. It can help us to understand if the cycles is noisy or not.

This patch is inspired by Andi Kleen's patch:

  https://lwn.net/Articles/600471/

We create new option '--cycles-hist'.

Example:

  perf record -b ./div
  perf record -b ./div
  perf diff -c cycles

  # Baseline                                [Program Block Range] Cycles Diff  Shared Object      Symbol
  # ........  .......................................................... ....  .................  ............................
  #
      46.72%                                      [div.c:40 -> div.c:40]    0  div                [.] main
      46.72%                                      [div.c:42 -> div.c:44]    0  div                [.] main
      46.72%                                      [div.c:42 -> div.c:39]    0  div                [.] main
      20.54%                          [random_r.c:357 -> random_r.c:394]    1  libc-2.27.so       [.] __random_r
      20.54%                          [random_r.c:357 -> random_r.c:380]    0  libc-2.27.so       [.] __random_r
      20.54%                          [random_r.c:388 -> random_r.c:388]    0  libc-2.27.so       [.] __random_r
      20.54%                          [random_r.c:388 -> random_r.c:391]    0  libc-2.27.so       [.] __random_r
      17.04%                              [random.c:288 -> random.c:291]    0  libc-2.27.so       [.] __random
      17.04%                              [random.c:291 -> random.c:291]    0  libc-2.27.so       [.] __random
      17.04%                              [random.c:293 -> random.c:293]    0  libc-2.27.so       [.] __random
      17.04%                              [random.c:295 -> random.c:295]    0  libc-2.27.so       [.] __random
      17.04%                              [random.c:295 -> random.c:295]    0  libc-2.27.so       [.] __random
      17.04%                              [random.c:298 -> random.c:298]    0  libc-2.27.so       [.] __random
       8.40%                                      [div.c:22 -> div.c:25]    0  div                [.] compute_flag
       8.40%                                      [div.c:27 -> div.c:28]    0  div                [.] compute_flag
       5.14%                                    [rand.c:26 -> rand.c:27]    0  libc-2.27.so       [.] rand
       5.14%                                    [rand.c:28 -> rand.c:28]    0  libc-2.27.so       [.] rand
       2.15%                                  [rand@plt+0 -> rand@plt+0]    0  div                [.] rand@plt
       0.00%                                                                   [kernel.kallsyms]  [k] __x86_indirect_thunk_rax
       0.00%                                [do_mmap+714 -> do_mmap+732]  -10  [kernel.kallsyms]  [k] do_mmap
       0.00%                                [do_mmap+737 -> do_mmap+765]    1  [kernel.kallsyms]  [k] do_mmap
       0.00%                                [do_mmap+262 -> do_mmap+299]    0  [kernel.kallsyms]  [k] do_mmap
       0.00%  [__x86_indirect_thunk_r15+0 -> __x86_indirect_thunk_r15+0]    7  [kernel.kallsyms]  [k] __x86_indirect_thunk_r15
       0.00%            [native_sched_clock+0 -> native_sched_clock+119]   -1  [kernel.kallsyms]  [k] native_sched_clock
       0.00%                 [native_write_msr+0 -> native_write_msr+16]  -13  [kernel.kallsyms]  [k] native_write_msr

When we enable the option '--cycles-hist', the output is

  perf diff -c cycles --cycles-hist

  # Baseline                                [Program Block Range] Cycles Diff        stddev/Hist  Shared Object      Symbol
  # ........  .......................................................... ....  .................  .................  ............................
  #
      46.72%                                      [div.c:40 -> div.c:40]    0  ± 37.8% ▁█▁▁██▁█   div                [.] main
      46.72%                                      [div.c:42 -> div.c:44]    0  ± 49.4% ▁▁▂█▂▂▂▂   div                [.] main
      46.72%                                      [div.c:42 -> div.c:39]    0  ± 24.1% ▃█▂▄▁▃▂▁   div                [.] main
      20.54%                          [random_r.c:357 -> random_r.c:394]    1  ± 33.5% ▅▂▁█▃▁▂▁   libc-2.27.so       [.] __random_r
      20.54%                          [random_r.c:357 -> random_r.c:380]    0  ± 39.4% ▁▁█▁██▅▁   libc-2.27.so       [.] __random_r
      20.54%                          [random_r.c:388 -> random_r.c:388]    0                     libc-2.27.so       [.] __random_r
      20.54%                          [random_r.c:388 -> random_r.c:391]    0  ± 41.2% ▁▃▁▂█▄▃▁   libc-2.27.so       [.] __random_r
      17.04%                              [random.c:288 -> random.c:291]    0  ± 48.8% ▁▁▁▁███▁   libc-2.27.so       [.] __random
      17.04%                              [random.c:291 -> random.c:291]    0  ±100.0% ▁█▁▁▁▁▁▁   libc-2.27.so       [.] __random
      17.04%                              [random.c:293 -> random.c:293]    0  ±100.0% ▁█▁▁▁▁▁▁   libc-2.27.so       [.] __random
      17.04%                              [random.c:295 -> random.c:295]    0  ±100.0% ▁█▁▁▁▁▁▁   libc-2.27.so       [.] __random
      17.04%                              [random.c:295 -> random.c:295]    0                     libc-2.27.so       [.] __random
      17.04%                              [random.c:298 -> random.c:298]    0  ± 75.6% ▃█▁▁▁▁▁▁   libc-2.27.so       [.] __random
       8.40%                                      [div.c:22 -> div.c:25]    0  ± 42.1% ▁▃▁▁███▁   div                [.] compute_flag
       8.40%                                      [div.c:27 -> div.c:28]    0  ± 41.8% ██▁▁▄▁▁▄   div                [.] compute_flag
       5.14%                                    [rand.c:26 -> rand.c:27]    0  ± 37.8% ▁▁▁████▁   libc-2.27.so       [.] rand
       5.14%                                    [rand.c:28 -> rand.c:28]    0                     libc-2.27.so       [.] rand
       2.15%                                  [rand@plt+0 -> rand@plt+0]    0                     div                [.] rand@plt
       0.00%                                                                                      [kernel.kallsyms]  [k] __x86_indirect_thunk_rax
       0.00%                                [do_mmap+714 -> do_mmap+732]  -10                     [kernel.kallsyms]  [k] do_mmap
       0.00%                                [do_mmap+737 -> do_mmap+765]    1                     [kernel.kallsyms]  [k] do_mmap
       0.00%                                [do_mmap+262 -> do_mmap+299]    0                     [kernel.kallsyms]  [k] do_mmap
       0.00%  [__x86_indirect_thunk_r15+0 -> __x86_indirect_thunk_r15+0]    7                     [kernel.kallsyms]  [k] __x86_indirect_thunk_r15
       0.00%            [native_sched_clock+0 -> native_sched_clock+119]   -1  ± 38.5% ▄█▁        [kernel.kallsyms]  [k] native_sched_clock
       0.00%                 [native_write_msr+0 -> native_write_msr+16]  -13  ± 47.1% ▁█▇▃▁▁     [kernel.kallsyms]  [k] native_write_msr

 v8:
 ---
 Rebase to perf/core branch

 v7:
 ---
 1. v6 got Jiri's ACK.
 2. Rebase to latest perf/core branch.

 v6:
 ---
 1. Jiri provides better code for using data__hpp_register() in ui_init().
    Use this code in v6.

 v5:
 ---
 1. Refine the use of data__hpp_register() in ui_init() according to
    Jiri's suggestion.

 v4:
 ---
 1. Rename the new option from '--noisy' to '--cycles-hist'
 2. Remove the option '-n'.
 3. Only update the spark value and stats when '--cycles-hist' is enabled.
 4. Remove the code of printing '..'.

 v3:
 ---
 1. Move the histogram to a separate column
 2. Move the svals[] out of struct stats

 v2:
 ---
 Jiri got a compile error,

  CC       builtin-diff.o
  builtin-diff.c: In function ‘compute_cycles_diff’:
  builtin-diff.c:712:10: error: taking the absolute value of unsigned type ‘u64’ {aka ‘long unsigned int’} has no effect [-Werror=absolute-value]
  712 |          labs(pair->block_info->cycles_spark[i] -
      |          ^~~~

 Because the result of u64 - u64 is still u64. Now we change the type of
 cycles_spark[] to s64.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20190925011446.30678-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-10-11 10:57:00 -03:00
Arnaldo Carvalho de Melo 11aad897f6 perf annotate: Don't return -1 for error when doing BPF disassembly
Return errno when open_memstream() fails and add two new speciall error
codes for when an invalid, non BPF file or one without BTF is passed to
symbol__disassemble_bpf(), so that its callers can rely on
symbol__strerror_disassemble() to convert that to a human readable error
message that can help figure out what is wrong, with hints even.

Cc: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: Song Liu <songliubraving@fb.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/n/tip-usevw9r2gcipfcrbpaueurw0@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-30 17:30:06 -03:00
Arnaldo Carvalho de Melo 16ed3c1e91 perf annotate: Return appropriate error code for allocation failures
We should return errno or the annotation extra range understood by
symbol__strerror_disassemble() instead of -1, fix it, returning ENOMEM
instead.

Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/n/tip-8of1cmj3rz0mppfcshc9bbqq@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-30 17:30:04 -03:00
Arnaldo Carvalho de Melo 42d7a9107d perf annotate: Fix arch specific ->init() failure errors
They are called from symbol__annotate() and to propagate errors that can
help understand the problem make them return what
symbol__strerror_disassemble() known, i.e. errno codes and other
annotation specific errors in a special, out of errnos, range.

Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/n/tip-pqx7srcv7tixgid251aeboj6@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-30 17:30:03 -03:00
Arnaldo Carvalho de Melo 211f493b61 perf annotate: Propagate the symbol__annotate() error return
We were just returning -1 in symbol__annotate() when symbol__annotate()
failed, propagate its error as it is used later to pass to
symbol__strerror_disassemble() to present a error message to the user,
that in some cases were getting:

  "Invalid -1 error code"

Fix it to propagate the error.

Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/n/tip-0tj89rs9g7nbcyd5skadlvuu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-30 17:30:01 -03:00
Arnaldo Carvalho de Melo 28f4417c33 perf annotate: Fix the signedness of failure returns
Callers of symbol__annotate() expect a errno value or some other
extended error value range in symbol__strerror_disassemble() to
convert to a proper error string, fix it when propagating a failure to
find the arch specific annotation routines via arch__find(arch_name).

Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/n/tip-o0k6dw7cas0vvmjjvgsyvu1i@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-30 17:30:00 -03:00
Arnaldo Carvalho de Melo a66fa0619a perf annotate: Propagate perf_env__arch() error
The callers of symbol__annotate2() use symbol__strerror_disassemble() to
convert its failure returns into a human readable string, so
propagate error values from functions it calls, starting with
perf_env__arch() that when fails the right thing to do is to look at
'errno' to see why its possible call to uname() failed.

Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/n/tip-it5d83kyusfhb1q1b0l4pxzs@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-30 17:29:58 -03:00
Arnaldo Carvalho de Melo 252a2fdc74 perf tools: Replace needless mmap.h with what is needed, event.h
The perf_sample struct definition and the event_attr_init() are in
util/event.h, but some places were getting it thru an otherwise needless
util/mmap.h header, fix it by including util/event.h directly.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-p1anwyjdbbvghrkl9dlxv7h5@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-25 16:26:40 -03:00
Arnaldo Carvalho de Melo e0fcfb086f perf evlist: Adopt backwards ring buffer state enum
As this isn't used at all in mmap.h but in evlist.h, so to cut down the
header dependency tree, move it to where it is used.

Also add mmap.h to the places using it but previously getting it
indirectly via evlist.h.

Add missing pthread.h to evlist.h, as it has a pthread_t struct member
and was getting the header via mmap.h.

Noticed while processing a Jiri's libperf batch touching mmap.h, where
almost everything gets rebuilt because evlist.h is so popular, so cut
down't this rebuild the world party.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Song Liu <songliubraving@fb.com>
Link: https://lkml.kernel.org/n/tip-he0uljeftl0xfveh3d6vtode@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-25 09:51:45 -03:00