diff options
| author | Namhyung Kim <namhyung@kernel.org> | 2022-08-02 22:10:04 +0300 |
|---|---|---|
| committer | Arnaldo Carvalho de Melo <acme@redhat.com> | 2022-08-03 00:03:31 +0300 |
| commit | 6d499a6b3d90277dbb7e408e7c70d68b507ef383 (patch) | |
| tree | c7b31abbff7da55a7a83e42f7f149377f17ba46c /tools/perf/util/bpf_skel | |
| parent | ceb13bfc01d55388092c00796a5b76aa3dfa483a (diff) | |
| download | linux-6d499a6b3d90277dbb7e408e7c70d68b507ef383.tar.xz | |
perf lock: Print the number of lost entries for BPF
Like the normal 'perf lock contention' output, it'd print the number of
lost entries for BPF if exists or -v option is passed.
Currently it uses BROKEN_CONTENDED stat for the lost count (due to full
stack maps).
$ sudo perf lock con -a -b --map-nr-entries 128 sleep 5
...
=== output for debug===
bad: 43, total: 14903
bad rate: 0.29 %
histogram of events caused bad sequence
acquire: 0
acquired: 0
contended: 43
release: 0
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Blake Jones <blakejones@google.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <songliubraving@fb.com>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20220802191004.347740-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Diffstat (limited to 'tools/perf/util/bpf_skel')
| -rw-r--r-- | tools/perf/util/bpf_skel/lock_contention.bpf.c | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/tools/perf/util/bpf_skel/lock_contention.bpf.c b/tools/perf/util/bpf_skel/lock_contention.bpf.c index 67d46533e518..9e8b94eb6320 100644 --- a/tools/perf/util/bpf_skel/lock_contention.bpf.c +++ b/tools/perf/util/bpf_skel/lock_contention.bpf.c @@ -12,7 +12,7 @@ #define MAX_ENTRIES 10240 struct contention_key { - __u32 stack_id; + __s32 stack_id; }; struct contention_data { @@ -27,7 +27,7 @@ struct tstamp_data { __u64 timestamp; __u64 lock; __u32 flags; - __u32 stack_id; + __s32 stack_id; }; /* callstack storage */ @@ -73,6 +73,9 @@ int enabled; int has_cpu; int has_task; +/* error stat */ +unsigned long lost; + static inline int can_record(void) { if (has_cpu) { @@ -116,6 +119,8 @@ int contention_begin(u64 *ctx) pelem->flags = (__u32)ctx[1]; pelem->stack_id = bpf_get_stackid(ctx, &stacks, BPF_F_FAST_STACK_CMP); + if (pelem->stack_id < 0) + lost++; return 0; } |