diff options
| author | Ken Chen <kenchen@google.com> | 2010-11-11 14:05:16 -0800 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@suse.de> | 2010-12-09 13:32:53 -0800 |
| commit | a7a001eb003abe7156c245609af7825bf65f2698 (patch) | |
| tree | aa04ab7fbf97c7c709da03646da5c1ef2a971cbf /kernel | |
| parent | 8301e7e3480ecabce25e116f1e6072b88f6167b4 (diff) | |
latencytop: fix per task accumulator
commit 38715258aa2e8cd94bd4aafadc544e5104efd551 upstream.
Per task latencytop accumulator prematurely terminates due to erroneous
placement of latency_record_count. It should be incremented whenever a
new record is allocated instead of increment on every latencytop event.
Also fix search iterator to only search known record events instead of
blindly searching all pre-allocated space.
Signed-off-by: Ken Chen <kenchen@google.com>
Reviewed-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'kernel')
| -rw-r--r-- | kernel/latencytop.c | 17 |
1 files changed, 8 insertions, 9 deletions
diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 877fb306d415..17110a4a4fc2 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -194,14 +194,7 @@ __account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) account_global_scheduler_latency(tsk, &lat); - /* - * short term hack; if we're > 32 we stop; future we recycle: - */ - tsk->latency_record_count++; - if (tsk->latency_record_count >= LT_SAVECOUNT) - goto out_unlock; - - for (i = 0; i < LT_SAVECOUNT; i++) { + for (i = 0; i < tsk->latency_record_count; i++) { struct latency_record *mylat; int same = 1; @@ -227,8 +220,14 @@ __account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) } } + /* + * short term hack; if we're > 32 we stop; future we recycle: + */ + if (tsk->latency_record_count >= LT_SAVECOUNT) + goto out_unlock; + /* Allocated a new one: */ - i = tsk->latency_record_count; + i = tsk->latency_record_count++; memcpy(&tsk->latency_record[i], &lat, sizeof(struct latency_record)); out_unlock: |