Skip to content

K
Kernel

openeuler / Kernel
接近 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 04 4月, 2018 1 次提交
    • A
      perf annotate browser: Show extra title line with event information · 6920e285
      由 Arnaldo Carvalho de Melo 提交于 
      So at the top we'll have two lines, like this, from 'perf report':

  # perf report --group --ignore-vmlinux
=====================================================================================================
Samples: 46  of events 'cycles', 4000 Hz, Event count (approx.): 5154895
_raw_spin_lock_irqsave  /proc/kcore
Percent              │      nop
                     │      push   %rbx
  0.00  14.29   0.00 │      pushfq
  9.09   0.00   0.00 │      pop    %rax
  9.09   0.00  20.00 │      nop
                     │      mov    %rax,%rbx
                     │      cli
  4.55   7.14   0.00 │      nop
                     │      xor    %eax,%eax
                     │      mov    $0x1,%edx
                     │      lock   cmpxchg %edx,(%rdi)
 77.27  78.57  70.00 │      test   %eax,%eax
                     │    ↓ jne    2b
                     │      mov    %rbx,%rax
  0.00   0.00  10.00 │      pop    %rbx
                     │    ← retq
                     │2b:   mov    %eax,%esi
                     │    → callq  queued_spin_lock_slowpath
                     │      mov    %rbx,%rax
                     │      pop    %rbx
Press 'h' for help on│key bindings
=====================================================================================================

 9.09 + 9.09 + 4.55 + 77.27 = 100
14.29 + 7.14 + 78.57 = 100
20 + 70 + 10 = 100

We can do the math by using 't' to toggle from 'percent' to nr

=====================================================================================================
Samples: 46  of events 'cycles', 4000 Hz, Event count (approx.): 5154895
_raw_spin_lock_irqsave  /proc/kcore
Period                              │      nop
                                    │      push   %rbx
          0       79273           0 │      pushfq
     190455           0           0 │      pop    %rax
     198038           0        3045 │      nop
                                    │      mov    %rax,%rbx
                                    │      cli
     217233       32562           0 │      nop
                                    │      xor    %eax,%eax
                                    │      mov    $0x1,%edx
                                    │      lock   cmpxchg %edx,(%rdi)
    3421649      979174       28273 │      test   %eax,%eax
                                    │    ↓ jne    2b
                                    │      mov    %rbx,%rax
          0           0        5193 │      pop    %rbx
                                    │    ← retq
                                    │2b:   mov    %eax,%esi
                                    │    → callq  queued_spin_lock_slowpath
                                    │      mov    %rbx,%rax
                                    │      pop    %rbx
Press 'h' for help on│key bindings
=====================================================================================================

79273 + 190455 + 198038 + 3045 + 217233 + 32562 + 3421649 + 979174 + 28273 + 5193 = 5154895

Or number of samples:

=====================================================================================================
ooSamples: 46  of events 'cycles', 4000 Hz, Event count (approx.): 5154895
_raw_spin_lock_irqsave  /proc/kcore
Samples              │      nop
                     │      push   %rbx
     0      2      0 │      pushfq
     2      0      0 │      pop    %rax
     2      0      2 │      nop
                     │      mov    %rax,%rbx
                     │      cli
     1      1      0 │      nop
                     │      xor    %eax,%eax
                     │      mov    $0x1,%edx
                     │      lock   cmpxchg %edx,(%rdi)
    17     11      7 │      test   %eax,%eax
                     │    ↓ jne    2b
                     │      mov    %rbx,%rax
     0      0      1 │      pop    %rbx
                     │    ← retq
                     │2b:   mov    %eax,%esi
                     │    → callq  queued_spin_lock_slowpath
                     │      mov    %rbx,%rax
                     │      pop    %rbx
Press 'h' for help on key bindings
=====================================================================================================

2 + 2 + 2 + 2 + 1 + 1 + 17 + 11 + 7 + 1 = 46
Suggested-by: NMartin Liška <mliska@suse.cz>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=196935
Link: https://lkml.kernel.org/n/tip-ezccyxld50wtwyt66np6aomo@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      6920e285
  3. 03 4月, 2018 4 次提交
  5. 24 3月, 2018 2 次提交
    • A
      perf annotate: Support jumping from one function to another · e4cc91b8
      由 Arnaldo Carvalho de Melo 提交于 
      For instance:

  entry_SYSCALL_64  /lib/modules/4.16.0-rc5-00086-gdf09348f/build/vmlinux
    5.50 │     → callq  do_syscall_64
   14.56 │       mov    0x58(%rsp),%rcx
    7.44 │       mov    0x80(%rsp),%r11
    0.32 │       cmp    %rcx,%r11
         │     → jne    swapgs_restore_regs_and_return_to_usermode
    0.32 │       shl    $0x10,%rcx
    0.32 │       sar    $0x10,%rcx
    3.24 │       cmp    %rcx,%r11
         │     → jne    swapgs_restore_regs_and_return_to_usermode
    2.27 │       cmpq   $0x33,0x88(%rsp)
    1.29 │     → jne    swapgs_restore_regs_and_return_to_usermode
         │       mov    0x30(%rsp),%r11
    8.74 │       cmp    %r11,0x90(%rsp)
         │     → jne    swapgs_restore_regs_and_return_to_usermode
    0.32 │       test   $0x10100,%r11
         │     → jne    swapgs_restore_regs_and_return_to_usermode
    0.32 │       cmpq   $0x2b,0xa0(%rsp)
    0.65 │     → jne    swapgs_restore_regs_and_return_to_usermode

It'll behave just like a "call" instruction, i.e. press enter or right
arrow over one such line and the browser will navigate to the annotated
disassembly of that function, which when exited, via left arrow or esc,
will come back to the calling function.

Now to support jump to an offset on a different function...
Reported-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-78o508mqvr8inhj63ddtw7mo@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      e4cc91b8
    • A
      perf annotate: Add "_local" to jump/offset validation routines · 2eff0611
      由 Arnaldo Carvalho de Melo 提交于 
      Because they all really check if we can access data structures/visual
constructs where a "jump" instruction targets code in the same function,
i.e. things like:

  __pthread_mutex_lock  /usr/lib64/libpthread-2.26.so
  1.95 │       mov    __pthread_force_elision,%ecx
       │    ┌──test   %ecx,%ecx
  0.07 │    ├──je     60
       │    │  test   $0x300,%esi
       │    │↓ jne    60
       │    │  or     $0x100,%esi
       │    │  mov    %esi,0x10(%rdi)
       │ 42:│  mov    %esi,%edx
       │    │  lea    0x16(%r8),%rsi
       │    │  mov    %r8,%rdi
       │    │  and    $0x80,%edx
       │    │  add    $0x8,%rsp
       │    │→ jmpq   __lll_lock_elision
       │    │  nop
  0.29 │ 60:└─→and    $0x80,%esi
  0.07 │       mov    $0x1,%edi
  0.29 │       xor    %eax,%eax
  2.53 │       lock   cmpxchg %edi,(%r8)

And not things like that "jmpq __lll_lock_elision", that instead should behave
like a "call" instruction and "jump" to the disassembly of "___lll_lock_elision".

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-3cwx39u3h66dfw9xjrlt7ca2@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      2eff0611
  7. 21 3月, 2018 30 次提交
  9. 17 3月, 2018 1 次提交
    • I
      perf report: Show zero counters as well in 'perf report --stat' · 39ce7fb3
      由 Ingo Molnar 提交于 
      When recently using 'perf report --stat' it was not clear to me from the
output whether a particular statistics field (LOST_SAMPLES) was not
present, or just zero:

  fomalhaut:~> perf report --stat

  Aggregated stats:
           TOTAL events:     495984
            MMAP events:         85
            COMM events:       3389
            EXIT events:       1605
        THROTTLE events:          2
      UNTHROTTLE events:          2
            FORK events:       3377
          SAMPLE events:     472629
           MMAP2 events:      14753
  FINISHED_ROUND events:        139
      THREAD_MAP events:          1
         CPU_MAP events:          1
       TIME_CONV events:          1

I had to check the output several times to ascertain that I'm not
misreading the output, that the field didn't change and that I didn't
misremember the name. In fact I had to look into the perf source to make
sure that zero fields are indeed not shown.

With the patch applied:

  fomalhaut:~> perf report --stat

  Aggregated stats:
           TOTAL events:     495984
            MMAP events:         85
            LOST events:          0
            COMM events:       3389
            EXIT events:       1605
        THROTTLE events:          2
      UNTHROTTLE events:          2
            FORK events:       3377
            READ events:          0
          SAMPLE events:     472629
           MMAP2 events:      14753
             AUX events:          0
    ITRACE_START events:          0
    LOST_SAMPLES events:          0
          SWITCH events:          0
 SWITCH_CPU_WIDE events:          0
      NAMESPACES events:          0
            ATTR events:          0
      EVENT_TYPE events:          0
    TRACING_DATA events:          0
        BUILD_ID events:          0
  FINISHED_ROUND events:        139
        ID_INDEX events:          0
   AUXTRACE_INFO events:          0
        AUXTRACE events:          0
  AUXTRACE_ERROR events:          0
      THREAD_MAP events:          1
         CPU_MAP events:          1
     STAT_CONFIG events:          0
            STAT events:          0
      STAT_ROUND events:          0
    EVENT_UPDATE events:          0
       TIME_CONV events:          1
         FEATURE events:          0

It's pretty clear at a glance that LOST_SAMPLES is present but zero.

The original output can still be gotten via:

  fomalhaut:~> perf report --stat | grep -vw 0

  Aggregated stats:
           TOTAL events:     495984
            MMAP events:         85
            COMM events:       3389
            EXIT events:       1605
        THROTTLE events:          2
      UNTHROTTLE events:          2
            FORK events:       3377
          SAMPLE events:     472629
           MMAP2 events:      14753
  FINISHED_ROUND events:        139
      THREAD_MAP events:          1
         CPU_MAP events:          1
       TIME_CONV events:          1

So I don't think there's any real loss in functionality.
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Acked-by: NJiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20180307152430.7e5h7e657b7bgd7q@gmail.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      39ce7fb3
  11. 08 3月, 2018 1 次提交
  13. 05 3月, 2018 1 次提交