Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 28 2月, 2015 4 次提交
    • Y
      perf tools: Remove the '--(null)' long_name for --list-opts · 3ef1e65c
      由 Yunlong Song 提交于 
      If the long_name of a 'struct option' is defined as NULL, --list-opts
will incorrectly print '--(null)' in its output. As a result, '--(null)'
will finally appear in the case of bash completion, e.g. 'perf record
--'.

Example:

Before this patch:

 $ perf record --list-opts

 --event --filter --pid --tid --realtime --no-buffering --raw-samples
 --all-cpus --cpu --count --output --no-inherit --freq --mmap-pages
 --group --(null) --call-graph --verbose --quiet --stat --data
 --timestamp --period --no-samples --no-buildid-cache --no-buildid
 --cgroup --delay --uid --branch-any --branch-filter --weight
 --transaction --per-thread --intr-regs

After this patch:

 $ perf record --list-opts

 --event --filter --pid --tid --realtime --no-buffering --raw-samples
 --all-cpus --cpu --count --output --no-inherit --freq --mmap-pages
 --group --call-graph --verbose --quiet --stat --data --timestamp
 --period --no-samples --no-buildid-cache --no-buildid --cgroup --delay
 --uid --branch-any --branch-filter --weight --transaction --per-thread
 --intr-regs
Signed-off-by: NYunlong Song <yunlong.song@huawei.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1425032491-20224-7-git-send-email-yunlong.song@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      3ef1e65c
    • Y
      perf list: Avoid confusion of perf output and the next command prompt · ed457520
      由 Yunlong Song 提交于 
      Distinguish the output of 'perf list --list-opts' or 'perf --list-cmds'
with the next command prompt, which also happens in other cases (e.g.
record, report ...).

Example:

Before this patch:

 $perf list --list-opts
 --raw-dump $          <-- the output and the next command prompt are at
                           the same line

After this patch:

 $perf list --list-opts
 --raw-dump
 $                     <-- the new line
Signed-off-by: NYunlong Song <yunlong.song@huawei.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1425032491-20224-6-git-send-email-yunlong.song@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      ed457520
    • Y
      perf list: Allow listing events with 'tracepoint' prefix · 16114951
      由 Yunlong Song 提交于 
      If somebody happens to name an event with the beginning of 'tracepoint'
(e.g. tracepoint_foo), then it will never be showed with perf list
event_glob, thus we parse the argument 'tracepoint' more carefully for
accuracy.

Example:

Before this patch:

 $ perf list tracepoint_foo:*

   jbd2:jbd2_start_commit                             [Tracepoint event]
   jbd2:jbd2_commit_locking                           [Tracepoint event]
   jbd2:jbd2_run_stats                                [Tracepoint event]
   block:block_rq_issue                               [Tracepoint event]
   block:block_bio_complete                           [Tracepoint event]
   block:block_bio_backmerge                          [Tracepoint event]
   block:block_getrq                                  [Tracepoint event]
   ...                                                ...

As shown above, all of the tracepoint events are printed. In fact, the
command's real intention is to print the events of tracepoint_foo.

After this patch:

 $ perf list tracepoint_foo:*

   tracepoint_foo:tp_foo_enter                        [Tracepoint event]
   tracepoint_foo:tp_foo_exit                         [Tracepoint event]

As shown above, only the events of tracepoint_foo are printed.
Signed-off-by: NYunlong Song <yunlong.song@huawei.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1425032491-20224-3-git-send-email-yunlong.song@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      16114951
    • Y
      perf list: Sort the output of 'perf list' to view more clearly · ab0e4800
      由 Yunlong Song 提交于 
      Sort the output according to ASCII character list (using strcmp), which
supports both number sequence and alphabet sequence.

Example:

Before this patch:

 $ perf list

 List of pre-defined events (to be used in -e):
   cpu-cycles OR cycles                               [Hardware event]
   instructions                                       [Hardware event]
   cache-references                                   [Hardware event]
   cache-misses                                       [Hardware event]
   branch-instructions OR branches                    [Hardware event]
   branch-misses                                      [Hardware event]
   bus-cycles                                         [Hardware event]
   ...                                                ...

   jbd2:jbd2_start_commit                             [Tracepoint event]
   jbd2:jbd2_commit_locking                           [Tracepoint event]
   jbd2:jbd2_run_stats                                [Tracepoint event]
   block:block_rq_issue                               [Tracepoint event]
   block:block_bio_complete                           [Tracepoint event]
   block:block_bio_backmerge                          [Tracepoint event]
   block:block_getrq                                  [Tracepoint event]
   ...                                                ...

After this patch:

 $ perf list

 List of pre-defined events (to be used in -e):
   branch-instructions OR branches                    [Hardware event]
   branch-misses                                      [Hardware event]
   bus-cycles                                         [Hardware event]
   cache-misses                                       [Hardware event]
   cache-references                                   [Hardware event]
   cpu-cycles OR cycles                               [Hardware event]
   instructions                                       [Hardware event]
   ...                                                ...

   block:block_bio_backmerge                          [Tracepoint event]
   block:block_bio_complete                           [Tracepoint event]
   block:block_getrq                                  [Tracepoint event]
   block:block_rq_issue                               [Tracepoint event]
   jbd2:jbd2_commit_locking                           [Tracepoint event]
   jbd2:jbd2_run_stats                                [Tracepoint event]
   jbd2:jbd2_start_commit                             [Tracepoint event]
   ...                                                ...
Signed-off-by: NYunlong Song <yunlong.song@huawei.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1425032491-20224-2-git-send-email-yunlong.song@huawei.com
[ Don't forget closedir({sys,evt}_dir) when handling errors ]
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      ab0e4800
  3. 27 2月, 2015 5 次提交
    • Y
      perf data: Fix sentinel setting for data_cmds array · 1f924c29
      由 Yunlong Song 提交于 
      The recent new patch "perf tools: Add new 'perf data' command" (commit
2245bf14 in acme's git repo perf/core) has caused a building error when
compiling the source code of perf:

 cc1: warnings being treated as errors
 builtin-data.c:89: error: missing initializer
 builtin-data.c:89: error: (near initialization for ‘data_cmds[1].summary’)
 make[2]: *** [builtin-data.o] Error 1
 make[2]: *** Waiting for unfinished jobs....
   LD       bench/perf-in.o
   LD       tests/perf-in.o
 make[1]: *** [perf-in.o] Error 2
 make: *** [all] Error 2

This patch fixes the building error above.
Signed-off-by: NYunlong Song <yunlong.song@huawei.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1425038026-27604-1-git-send-email-yunlong.song@huawei.com
[ .name == NULL ends the loop, use it instead of seting all fields to NULL ]
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      1f924c29
    • H
      perf probe: Fix a precedence bug · f56847c2
      由 He Kuang 提交于 
      The minus operator has higher precedence than ?: Add parentheses around
?: fix this.

Before this patch:

  $ echo 'p:myprobe do_sys_open' > /sys/kernel/debug/tracing/kprobe_events
  $ perf probe -l -k ../vmlinux
    kprobes:myprobe      (on do_sys_open)

After this patch:

  $ echo 'p:myprobe do_sys_open' > /sys/kernel/debug/tracing/kprobe_events
  $ perf probe -l -k ../vmlinux
    kprobes:myprobe      (on do_sys_open@linux.git/fs/open.c)
Signed-off-by: NHe Kuang <hekuang@huawei.com>
Acked-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1425034373-14511-1-git-send-email-hekuang@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      f56847c2
    • K
      perf diff: Support for different binaries · 94ba462d
      由 Kan Liang 提交于 
      Currently, the perf diff only works with same binaries. That's because
it compares the symbol start address. It doesn't work if the perf.data
comes from different binaries. This patch matches the symbol names.

Actually, perf diff once intended to compare the symbol names.  The
commit as below can look for a pair by name.

604c5c92 (perf diff: Change the default sort order to "dso,symbol")
However, at that time, perf diff used a global list of dsos. That means
the binaries which has same name can only be loaded once. That's a
problem for comparing different binaries.

For example, we have an old binary and an updated binary. They very
likely have same name and most of the functions, so only dsos from old
binary will be loaded. When processing the data from updated binary,
perf still use the symbol information from old binary. That's wrong.

Then the commit as below used IP to replace symbol name.
9c443dfd ("perf diff: Fix support for all --sort combinations")
>From that time, perf diff starts to compare the symbol address.

The global dsos is discarded from a patch in 2010.
a1645ce1 ("perf: 'perf kvm' tool for monitoring guest performance
from host")
However, at that time, perf diff already compared by address. So perf
diff cannot work for different binaries as well.

This patch actually rolls back the perf diff to original design. The
document is also changed, so everybody knows the original design is to
compare the symbol names.

Here are some examples:

The only difference between example_v1.c and example_v2.c is the
location of f2 and f3. There is no change in behavior, but the previous
perf diff display the wrong differential profile.

example_v1.c
noinline void f3(void)
{
        volatile int i;
        for (i = 0; i < 10000;) {

                if(i%2)
                        i++;
                else
                        i++;
        }
}

noinline void f2(void)
{
        volatile int a = 100, b, c;
        for (b = 0; b < 10000; b++)
                c = a * b;

}

noinline void f1(void)
{
                f2();
                f3();
}

int main()
{
        int i;
        for (i = 0; i < 100000; i++)
                f1();
}

example_v2.c
noinline void f2(void)
{
        volatile int a = 100, b, c;
        for (b = 0; b < 10000; b++)
                c = a * b;
}

noinline void f3(void)
{
        volatile int i;
        for (i = 0; i < 10000;) {
                if(i%2)
                        i++;
                else
                        i++;
        }
}

noinline void f1(void)
{
                f2();
                f3();
}

int main()
{
        int i;
        for (i = 0; i < 100000; i++)
                f1();
}

[lk@localhost perf_diff]$ gcc example_v1.c -o example
[lk@localhost perf_diff]$ perf record -o example_v1.data ./example
[ perf record: Woken up 4 times to write data ]
[ perf record: Captured and wrote 0.813 MB example_v1.data (~35522 samples) ]

[lk@localhost perf_diff]$ gcc example_v2.c -o example
[lk@localhost perf_diff]$ perf record -o example_v2.data ./example
[ perf record: Woken up 4 times to write data ]
[ perf record: Captured and wrote 0.824 MB example_v2.data (~36015 samples) ]

Old perf diff result:

[lk@localhost perf_diff]$ perf diff example_v1.data example_v2.data
 Event 'cycles'
 Baseline    Delta  Shared Object     Symbol
 ........  .......  ................  ...............................

                     [kernel.vmlinux]  [k] __perf_event_task_sched_out
     0.00%           [kernel.vmlinux]  [k] apic_timer_interrupt
                     [kernel.vmlinux]  [k] idle_cpu
                     [kernel.vmlinux]  [k] intel_pstate_timer_func
                     [kernel.vmlinux]  [k] native_read_msr_safe
     0.00%           [kernel.vmlinux]  [k] native_read_tsc
     0.00%           [kernel.vmlinux]  [k] native_write_msr_safe
                     [kernel.vmlinux]  [k] ntp_tick_length
     0.00%           [kernel.vmlinux]  [k] rb_erase
     0.00%           [kernel.vmlinux]  [k] tick_sched_timer
     0.00%           [kernel.vmlinux]  [k] unmap_single_vma
     0.00%           [kernel.vmlinux]  [k] update_wall_time
     0.00%           example           [.] f1
    46.24%           example           [.] f2
    53.71%   -7.55%  example           [.] f3
            +53.81%  example           [.] f3
     0.02%           example           [.] main

New perf diff result:

[lk@localhost perf_diff]$ perf diff example_v1.data example_v2.data
                     [kernel.vmlinux]  [k] __perf_event_task_sched_out
     0.00%           [kernel.vmlinux]  [k] apic_timer_interrupt
                     [kernel.vmlinux]  [k] idle_cpu
                     [kernel.vmlinux]  [k] intel_pstate_timer_func
                     [kernel.vmlinux]  [k] native_read_msr_safe
     0.00%           [kernel.vmlinux]  [k] native_read_tsc
     0.00%           [kernel.vmlinux]  [k] native_write_msr_safe
                     [kernel.vmlinux]  [k] ntp_tick_length
     0.00%           [kernel.vmlinux]  [k] rb_erase
     0.00%           [kernel.vmlinux]  [k] tick_sched_timer
     0.00%           [kernel.vmlinux]  [k] unmap_single_vma
     0.00%           [kernel.vmlinux]  [k] update_wall_time
     0.00%           example           [.] f1
    46.24%   -0.08%  example           [.] f2
    53.71%   +0.11%  example           [.] f3
     0.02%           example           [.] main
Signed-off-by: NKan Liang <kan.liang@intel.com>
Acked-by: NJiri Olsa <jolsa@kernel.org>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1423460384-11645-1-git-send-email-kan.liang@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      94ba462d
    • M
      perf buildid-cache: Add new buildid cache if update target is not cached · a50d11a1
      由 Masami Hiramatsu 提交于 
      Add new buildid cache if the update target file is not cached.

This can happen when an old binary is replaced by new one after caching
the old one. In this case, user sees his operation just failed.

But it does not look straight, since user just pass the binary "path",
not "build-id".

  ----
  # ./perf buildid-cache --add ./perf
  (update ./perf to new binary)
  # ./perf buildid-cache --update ./perf
  ./perf wasn't in the cache
  #
  ----

This patch adds given new binary to cache if the new binary is
not cached. So we'll not see the above error.

  ----
  # ./perf buildid-cache --add ./perf
  (update ./perf to new binary)
  # ./perf buildid-cache --update ./perf
  #
  ----
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20150226065440.23912.1494.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      a50d11a1
    • A
      perf probe: Handle strdup() failure · 38ae502b
      由 Arnaldo Carvalho de Melo 提交于 
      We could end up returning 0 (Ok) with a NULL raw_path. Fix it.
Acked-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naohiro Aota <naota@elisp.net>
Link: http://lkml.kernel.org/n/tip-l0kcbcg5f4nnzqt01cv42vec@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      38ae502b
  5. 26 2月, 2015 5 次提交
    • M
      perf probe: Fix get_real_path to free allocated memory in error path · eb47cb2e
      由 Masami Hiramatsu 提交于 
      Fix get_real_path to free allocated memory when comp_dir is used for
complementing path and getting an error.
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naohiro Aota <naota@elisp.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20150226082504.28125.74506.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      eb47cb2e
    • M
      perf probe: Check kprobes blacklist when adding new events · 9aaf5a5f
      由 Masami Hiramatsu 提交于 
      Recent linux kernel provides a blacklist of the functions which can not
be probed. perf probe can now check this blacklist before setting new
events and indicate better error message for users.

Without this patch,
  ----
  # perf probe --add vmalloc_fault
  Added new event:
  Failed to write event: Invalid argument
    Error: Failed to add events.
  ----
With this patch
  ----
  # perf probe --add vmalloc_fault
  Added new event:
  Warning: Skipped probing on blacklisted function: vmalloc_fault
  ----
Reported-by: NArnaldo Carvalho de Melo <acme@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20150219143113.14434.5387.stgit@localhost.localdomainSigned-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      9aaf5a5f
    • D
      perf trace: Fix SIGBUS failures due to misaligned accesses · 55d43bca
      由 David Ahern 提交于 
      On Sparc64 perf-trace is failing in many spots due to extended load
instructions being used on misaligned accesses.

(gdb) run trace ls
Starting program: /tmp/perf/perf trace ls
[Thread debugging using libthread_db enabled]
Detaching after fork from child process 169460.

<ls output removed>

Program received signal SIGBUS, Bus error.
0x000000000014f4dc in tp_field__u64 (field=0x4cc700, sample=0x7feffffa098) at builtin-trace.c:61
warning: Source file is more recent than executable.
61      TP_UINT_FIELD(64);

(gdb) bt
 0  0x000000000014f4dc in tp_field__u64 (field=0x4cc700, sample=0x7feffffa098) at builtin-trace.c:61
 1  0x0000000000156ad4 in trace__sys_exit (trace=0x7feffffc268, evsel=0x4cc580, event=0xfffffc0104912000,
    sample=0x7feffffa098) at builtin-trace.c:1701
 2  0x0000000000158c14 in trace__run (trace=0x7feffffc268, argc=1, argv=0x7fefffff360) at builtin-trace.c:2160
 3  0x000000000015b78c in cmd_trace (argc=1, argv=0x7fefffff360, prefix=0x0) at builtin-trace.c:2609
 4  0x0000000000107d94 in run_builtin (p=0x4549c8, argc=2, argv=0x7fefffff360) at perf.c:341
 5  0x0000000000108140 in handle_internal_command (argc=2, argv=0x7fefffff360) at perf.c:400
 6  0x0000000000108308 in run_argv (argcp=0x7feffffef2c, argv=0x7feffffef20) at perf.c:444
 7  0x0000000000108728 in main (argc=2, argv=0x7fefffff360) at perf.c:559

(gdb) p *sample
$1 = {ip = 4391276, pid = 169472, tid = 169472, time = 6303014583281250, addr = 0, id = 72082,
  stream_id = 18446744073709551615, period = 1, weight = 0, transaction = 0, cpu = 73, raw_size = 36,
  data_src = 84410401, flags = 0, insn_len = 0, raw_data = 0xfffffc010491203c, callchain = 0x0,
  branch_stack = 0x0, user_regs = {abi = 0, mask = 0, regs = 0x0, cache_regs = 0x7feffffa098, cache_mask = 0},
  intr_regs = {abi = 0, mask = 0, regs = 0x0, cache_regs = 0x7feffffa098, cache_mask = 0}, user_stack = {
    offset = 0, size = 0, data = 0x0}, read = {time_enabled = 0, time_running = 0, {group = {nr = 0,
        values = 0x0}, one = {value = 0, id = 0}}}}
(gdb) p *field
$2 = {offset = 16, {integer = 0x14f4a8 <tp_field__u64>, pointer = 0x14f4a8 <tp_field__u64>}}

sample->raw_data is guaranteed to not be 8-byte aligned because it is preceded
by the size as a u3. So accessing raw data with an extended load instruction causes
the SIGBUS. Resolve by using memcpy to a temporary variable of appropriate size.
Signed-off-by: NDavid Ahern <david.ahern@oracle.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1424376022-140608-1-git-send-email-david.ahern@oracle.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      55d43bca
    • S
      perf data: Add a 'perf' prefix to the generic fields · 54cf776a
      由 Sebastian Andrzej Siewior 提交于 
      Some of the tracers bring their own id or pid fields and we can end up
having two of them. This patch adds a "perf_" prefix to the 'generic'
fields so we avoid a clash of the member names.

The change is visible in the babeltrace output:

Before:
  $ babeltrace ./ctf-data/
  [03:19:13.962131936] (+0.000001935) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 8 }
  [03:19:13.962133732] (+0.000001796) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 114 }
  ...

Now:
  $ babeltrace ./ctf-data/
  [03:19:13.962131936] (+0.000001935) cycles: { }, { perf_ip = 0xFFFFFFFF8105443A, perf_tid = 20714, perf_pid = 20714, perf_period = 8 }
  [03:19:13.962133732] (+0.000001796) cycles: { }, { perf_ip = 0xFFFFFFFF8105443A, perf_tid = 20714, perf_pid = 20714, perf_period = 114 }
  ...
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Reviewed-by: NDavid Ahern <dsahern@gmail.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jeremie Galarneau <jgalar@efficios.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1424470628-5969-5-git-send-email-jolsa@kernel.orgSigned-off-by: NJiri Olsa <jolsa@kernel.org>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      54cf776a
    • J
      perf data: Add perf data to CTF conversion support · edbe9817
      由 Jiri Olsa 提交于 
      Adding 'perf data convert' to convert perf data file into different
format. This patch adds support for CTF format conversion.

To convert perf.data into CTF run:
  $ perf data convert --to-ctf=./ctf-data/
  [ perf data convert: Converted 'perf.data' into CTF data './ctf-data/' ]
  [ perf data convert: Converted and wrote 11.268 MB (100230 samples) ]

The command will create CTF metadata out of perf.data file (or one
specified via -i option) and then convert all sample events into single
CTF stream.

Each sample_type bit is translated into separated CTF event field apart
from following exceptions:

  PERF_SAMPLE_RAW          - added in next patch
  PERF_SAMPLE_READ         - TODO
  PERF_SAMPLE_CALLCHAIN    - TODO
  PERF_SAMPLE_BRANCH_STACK - TODO
  PERF_SAMPLE_REGS_USER    - TODO
  PERF_SAMPLE_STACK_USER   - TODO

  $ perf --debug=data-convert=2 data convert ...

The converted CTF data could be analyzed by CTF tools, like babletrace
or tracecompass [1].

  $ babeltrace ./ctf-data/
  [03:19:13.962125533] (+?.?????????) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 1 }
  [03:19:13.962130001] (+0.000004468) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 1 }
  [03:19:13.962131936] (+0.000001935) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 8 }
  [03:19:13.962133732] (+0.000001796) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 114 }
  [03:19:13.962135557] (+0.000001825) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 2087 }
  [03:19:13.962137627] (+0.000002070) cycles: { }, { ip = 0xFFFFFFFF81361938, tid = 20714, pid = 20714, period = 37582 }
  [03:19:13.962161091] (+0.000023464) cycles: { }, { ip = 0xFFFFFFFF8124218F, tid = 20714, pid = 20714, period = 600246 }
  [03:19:13.962517569] (+0.000356478) cycles: { }, { ip = 0xFFFFFFFF811A75DB, tid = 20714, pid = 20714, period = 1325731 }
  [03:19:13.969518008] (+0.007000439) cycles: { }, { ip = 0x34080917B2, tid = 20714, pid = 20714, period = 1144298 }

The following members to the ctf-environment were decided to be added to
distinguish and specify perf CTF data:

  - domain

    It says "kernel" because it contains a kernel trace (not to be
    confused with a user space like lttng-ust does)

  - tracer_name

    It says perf. This can be used to distinguish between lttng and perf
    CTF based trace.

  - version

    The kernel version from stream. In addition to release, this is what
    it looks like on a Debian kernel:

      release = "3.14-1-amd64";
      version = "3.14.0";

[1] http://projects.eclipse.org/projects/tools.tracecompassSigned-off-by: NJiri Olsa <jolsa@kernel.org>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Reviewed-by: NDavid Ahern <dsahern@gmail.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jeremie Galarneau <jgalar@efficios.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1424470628-5969-4-git-send-email-jolsa@kernel.orgSigned-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      edbe9817
  7. 25 2月, 2015 6 次提交
  9. 23 2月, 2015 13 次提交
  11. 19 2月, 2015 2 次提交
    • K
      perf tools: Construct LBR call chain · 384b6055
      由 Kan Liang 提交于 
      LBR call stack only has user-space callchains. It is output in the
PERF_SAMPLE_BRANCH_STACK data format. For kernel callchains, it's
still in the form of PERF_SAMPLE_CALLCHAIN.

The perf tool has to handle both data sources to construct a
complete callstack.

For the "perf report -D" option, both lbr and fp information will be
displayed.

A new call chain recording option "lbr" is introduced into the perf
tool for LBR call stack. The user can use --call-graph lbr to get
the call stack information from hardware.

Here are some examples.

When profiling bc(1) on Fedora 19:

  echo 'scale=2000; 4*a(1)' > cmd; perf record --call-graph lbr bc -l < cmd

If enabling LBR, perf report output looks like:

    50.36%       bc  bc                 [.] bc_divide
                 |
                 --- bc_divide
                     execute
                     run_code
                     yyparse
                     main
                     __libc_start_main
                     _start
    33.66%       bc  bc                 [.] _one_mult
                 |
                 --- _one_mult
                     bc_divide
                     execute
                     run_code
                     yyparse
                     main
                     __libc_start_main
                     _start
     7.62%       bc  bc                 [.] _bc_do_add
                 |
                 --- _bc_do_add
                    |
                    |--99.89%-- 0x2000186a8
                     --0.11%-- [...]
     6.83%       bc  bc                 [.] _bc_do_sub
                 |
                 --- _bc_do_sub
                    |
                    |--99.94%-- bc_add
                    |          execute
                    |          run_code
                    |          yyparse
                    |          main
                    |          __libc_start_main
                    |          _start
                     --0.06%-- [...]
     0.46%       bc  libc-2.17.so       [.] __memset_sse2
                 |
                 --- __memset_sse2
                    |
                    |--54.13%-- bc_new_num
                    |          |
                    |          |--51.00%-- bc_divide
                    |          |          execute
                    |          |          run_code
                    |          |          yyparse
                    |          |          main
                    |          |          __libc_start_main
                    |          |          _start
                    |          |
                    |          |--30.46%-- _bc_do_sub
                    |          |          bc_add
                    |          |          execute
                    |          |          run_code
                    |          |          yyparse
                    |          |          main
                    |          |          __libc_start_main
                    |          |          _start
                    |          |
                    |           --18.55%-- _bc_do_add
                    |                     bc_add
                    |                     execute
                    |                     run_code
                    |                     yyparse
                    |                     main
                    |                     __libc_start_main
                    |                     _start
                    |
                     --45.87%-- bc_divide
                               execute
                               run_code
                               yyparse
                               main
                               __libc_start_main
                               _start

If using FP, perf report output looks like:

  echo 'scale=2000; 4*a(1)' > cmd; perf record --call-graph fp bc -l < cmd

    50.49%       bc  bc                 [.] bc_divide
                 |
                 --- bc_divide
    33.57%       bc  bc                 [.] _one_mult
                 |
                 --- _one_mult
     7.61%       bc  bc                 [.] _bc_do_add
                 |
                 --- _bc_do_add
                     0x2000186a8
     6.88%       bc  bc                 [.] _bc_do_sub
                 |
                 --- _bc_do_sub
     0.42%       bc  libc-2.17.so       [.] __memcpy_ssse3_back
                 |
                 --- __memcpy_ssse3_back

If using LBR, perf report -D output looks like:

3458145275743 0x2fd750 [0xd8]: PERF_RECORD_SAMPLE(IP, 0x2): 9748/9748: 0x408ea8 period: 609644 addr: 0
... LBR call chain: nr:8
.....  0: fffffffffffffe00
.....  1: 0000000000408e50
.....  2: 000000000040a458
.....  3: 000000000040562e
.....  4: 0000000000408590
.....  5: 00000000004022c0
.....  6: 00000000004015dd
.....  7: 0000003d1cc21b43
... FP chain: nr:2
.....  0: fffffffffffffe00
.....  1: 0000000000408ea8
 ... thread: bc:9748
 ...... dso: /usr/bin/bc

The LBR call stack has the following known limitations:

 - Zero length calls are not filtered out by the hardware

 - Exception handing such as setjmp/longjmp will have calls/returns not
   match

 - Pushing different return address onto the stack will have
   calls/returns not match

 - If callstack is deeper than the LBR, only the last entries are
   captured
Tested-by: NJiri Olsa <jolsa@kernel.org>
Signed-off-by: NKan Liang <kan.liang@intel.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Simon Que <sque@chromium.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1420482185-29830-3-git-send-email-kan.liang@intel.comSigned-off-by: NIngo Molnar <mingo@kernel.org>
      384b6055
    • K
      perf tools: Enable LBR call stack support · aad2b21c
      由 Kan Liang 提交于 
      Currently, there are two call chain recording options, fp and dwarf.

Haswell has a new feature that utilizes the existing LBR facility to
record call chains. Kernel side LBR support code provides this as a
third option to record call chains. This patch enables the lbr call
stack support on the tooling side.

LBR call stack has some limitations:

 - It reuses current LBR facility, so LBR call stack and branch record
   can not be enabled at the same time.

 - It is only available for user-space callchains.

However, it also offers some advantages:

 - LBR call stack can work on user apps which don't have frame-pointers
   or dwarf debug info compiled. It is a good alternative when nothing
   else works.
Tested-by: NJiri Olsa <jolsa@kernel.org>
Signed-off-by: NKan Liang <kan.liang@intel.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Cody P Schafer <cody@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Jacob Shin <jacob.w.shin@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masanari Iida <standby24x7@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rodrigo Campos <rodrigo@sdfg.com.ar>
Cc: Stephane Eranian <eranian@google.com>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1420482185-29830-2-git-send-email-kan.liang@intel.comSigned-off-by: NIngo Molnar <mingo@kernel.org>
      aad2b21c
  13. 14 2月, 2015 3 次提交
    • A
      perf trace: Support --events foo:bar --no-syscalls · 726f3234
      由 Arnaldo Carvalho de Melo 提交于 
      I.e. support tracing just tracepoints, without strace like
raw_syscalls:*.

[acme@ssdandy linux]$ trace --no-sys --ev sched:*exec,sched:*switch,sched:*exit usleep 1
  0.048 (     ): sched:sched_process_exec:filename=/usr/bin/usleep pid=27298 old_pid=27298)
  0.369 (     ): sched:sched_switch:usleep:27298 [120] S ==> swapper/5:0 [120])
  0.452 (     ): sched:sched_process_exit:comm=usleep pid=27298 prio=120)
[acme@ssdandy linux]$

TODO: remove that (...) thing when --no-syscalls is specified.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-vn0hsixsbhm31b2rpj97r96k@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      726f3234
    • A
      perf trace: Allow mixing with other events · 14a052df
      由 Arnaldo Carvalho de Melo 提交于 
      Basically adopting 'perf record' --event command line argument syntax:

 # trace -e \!mprotect,mmap,munmap,open,close,read,fstat,access,arch_prctl --event sched:*switch,sched:*exec,sched:*exit usleep 1
  0.048 (        ): sched:sched_process_exec:filename=/bin/usleep pid=24732 old_pid=24732)
  0.078 (0.002 ms): usleep/24732 brk(                          ) = 0x78f000
  0.430 (0.002 ms): usleep/24732 brk(                          ) = 0x78f000
  0.434 (0.003 ms): usleep/24732 brk(brk: 0x7b0000             ) = 0x7b0000
  0.438 (0.001 ms): usleep/24732 brk(                          ) = 0x7b0000
  0.460 (0.004 ms): usleep/24732 nanosleep(rqtp: 0x7ffff3696a40) ...
  0.460 (        ): sched:sched_switch:prev_comm=usleep prev_pid=24732 prev_prio=120 prev_state=S ==> next_comm=swapper/1 next_pid=0 next_prio=120)
  0.515 (0.058 ms): usleep/24732  ... [continued]: nanosleep()) = 0
  0.520 (0.000 ms): usleep/24732 exit_group(
  0.550 (        ): sched:sched_process_exit:comm=usleep pid=24732 prio=120)
 #

Next steps, probably in this order:

1) Use ordered_events code, the logic in trace needs the events to be
   time ordered when needed, i.e. when multiple CPUs are involved.

2) Callchains!

3) Automatically account for interruptions when saying how long things
   took.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-gpst8mph575yb4wgf91qibyb@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      14a052df
    • A
      perf trace: Handle multiple threads better wrt syscalls being intermixed · e596663e
      由 Arnaldo Carvalho de Melo 提交于 
       $ trace time taskset -c 0 usleep 1
   0.845 ( 0.021 ms): time/16722 wait4(upid: 4294967295, stat_addr: 0x7fff17f443d4, ru: 0x7fff17f44438 ) ...
   0.865 ( 0.008 ms): time/16723 execve(arg0: 140733595272004, arg1: 140733595272720, arg2: 140733595272768, arg3: 139755107218496, arg4: 7307199665339051828, arg5: 3) = -2
   2.395 ( 1.523 ms): taskset/16723 execve(arg0: 140733595272013, arg1: 140733595272720, arg2: 140733595272768, arg3: 139755107218496, arg4: 7307199665339051828, arg5: 3) = 0
   2.411 ( 0.002 ms): taskset/16723 brk(                                                                  ) = 0x1915000
   3.300 ( 0.058 ms): usleep/16723 nanosleep(rqtp: 0x7ffff4ada190                                        ) = 0
 <SNIP>
   3.305 ( 0.000 ms): usleep/16723 exit_group(
   3.363 ( 2.539 ms): time/16722  ... [continued]: wait4()) = 16723
   3.366 ( 0.001 ms): time/16722 rt_sigaction(sig: INT, act: 0x7fff17f44160, oact: 0x7fff17f44200, sigsetsize: 8) = 0

We we're not seeing this line:

  0.845 ( 0.021 ms): time/16722 wait4(upid: 4294967295, stat_addr: 0x7fff17f443d4, ru: 0x7fff17f44438 ) ...

just the one when it finishes:

  3.363 ( 2.539 ms): time/16722  ... [continued]: wait4()) = 16723

Still some issues left till we move to ordered_samples when multiple
CPUs/threads are involved...

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-zq9x30a1ky3djqewqn2v3ja3@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      e596663e
  15. 13 2月, 2015 2 次提交
    • A
      perf trace: Print thread info when following children · 42052bea
      由 Arnaldo Carvalho de Melo 提交于 
      The default for 'trace workload' is to set perf_event_attr.inherit to 1,
i.e. to make it equivalent to 'strace -f workload', so we were ending
with syscalls for multiple processes mixed up, fix it:

Before:

  [root@ssdandy ~]# trace -e brk time usleep 1
     0.071 ( 0.002 ms): brk(              ) = 0x100e000
     0.802 ( 0.001 ms): brk(              ) = 0x1d99000
     1.132 ( 0.003 ms): brk(              ) = 0x1d99000
     1.136 ( 0.003 ms): brk(brk: 0x1dba000) = 0x1dba000
     1.140 ( 0.001 ms): brk(              ) = 0x1dba000
  0.00user 0.00system 0:00.00elapsed 63%CPU (0avgtext+0avgdata 528maxresident)k
  0inputs+0outputs (0major+181minor)pagefaults 0swaps
  [root@ssdandy ~]#

After:

  [root@ssdandy ~]# trace -f -e brk time usleep 1
     0.072 ( 0.002 ms): time/26308 brk(               ) = 0x1e6e000
     0.860 ( 0.001 ms): usleep/26309 brk(             ) = 0xb91000
     1.193 ( 0.003 ms): usleep/26309 brk(             ) = 0xb91000
     1.197 ( 0.003 ms): usleep/26309 brk(brk: 0xbb2000) = 0xbb2000
     1.201 ( 0.001 ms): usleep/26309 brk(             ) = 0xbb2000
  0.00user 0.00system 0:00.00elapsed 0%CPU (0avgtext+0avgdata 524maxresident)k
  0inputs+0outputs (0major+180minor)pagefaults 0swaps
  [root@ssdandy ~]#

BTW: to achieve the 'strace workload' behaviour, i.e. without a explicit
'-f', one has to use --no-inherit.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
echo Link: http://lkml.kernel.org/n/tip-`ranpwd -l 24`@git.kernel.org
Link: http://lkml.kernel.org/n/tip-2wu2d5n65msxoq1i7vtcaft2@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      42052bea
    • Y
      perf list: Place the header text in its right position · 619a303c
      由 Yunlong Song 提交于 
      The hearer text 'List of pre-defined events (to be used in -e):' is
placed in an improper function, which causes an abnormal output, e.g.
'perf list hw' shows no guiding text at all, and 'perf list hw
L1-dcache*' shows the guiding text incorrectly in the middle of the
output.

Example
Before this patch:

 $ perf list hw L1-dcache*

   branch-instructions OR branches                    [Hardware event]
   branch-misses                                      [Hardware event]
   bus-cycles                                         [Hardware event]
   cache-misses                                       [Hardware event]
   cache-references                                   [Hardware event]
   cpu-cycles OR cycles                               [Hardware event]
   instructions                                       [Hardware event]
   stalled-cycles-backend OR idle-cycles-backend      [Hardware event]
   stalled-cycles-frontend OR idle-cycles-frontend    [Hardware event]

 List of pre-defined events (to be used in -e):              <-- incorrect position
   L1-dcache-load-misses                              [Hardware cache event]
   L1-dcache-loads                                    [Hardware cache event]
   L1-dcache-prefetch-misses                          [Hardware cache event]
   L1-dcache-prefetches                               [Hardware cache event]
   L1-dcache-store-misses                             [Hardware cache event]
   L1-dcache-stores                                   [Hardware cache event]

After this patch:

 $ perf list hw L1-dcache*

 List of pre-defined events (to be used in -e):              <-- correct position

   branch-instructions OR branches                    [Hardware event]
   branch-misses                                      [Hardware event]
   bus-cycles                                         [Hardware event]
   cache-misses                                       [Hardware event]
   cache-references                                   [Hardware event]
   cpu-cycles OR cycles                               [Hardware event]
   instructions                                       [Hardware event]
   stalled-cycles-backend OR idle-cycles-backend      [Hardware event]
   stalled-cycles-frontend OR idle-cycles-frontend    [Hardware event]

   L1-dcache-load-misses                              [Hardware cache event]
   L1-dcache-loads                                    [Hardware cache event]
   L1-dcache-prefetch-misses                          [Hardware cache event]
   L1-dcache-prefetches                               [Hardware cache event]
   L1-dcache-store-misses                             [Hardware cache event]
   L1-dcache-stores                                   [Hardware cache event]
Signed-off-by: NYunlong Song <yunlong.song@huawei.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1423833115-11199-8-git-send-email-yunlong.song@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      619a303c