perf: Fix wakeup storm for RO mmap()s

RO mmap()s don't update the tail pointer, so
comparing against it for determining the written data
size doesn't really do any good.

Keep track of when we last did a wakeup, and compare
against that.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20100521090710.684479310@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Peter Zijlstra 2010-05-20 16:21:55 +02:00 committed by Ingo Molnar
parent 0e2e63dd60
commit adb8e118f2
2 changed files with 3 additions and 3 deletions

View File

@ -599,7 +599,7 @@ struct perf_mmap_data {
local_t head; /* write position */
local_t nest; /* nested writers */
local_t events; /* event limit */
local_t wakeup; /* needs a wakeup */
local_t wakeup; /* wakeup stamp */
local_t lost; /* nr records lost */
long watermark; /* wakeup watermark */

View File

@ -3056,8 +3056,8 @@ int perf_output_begin(struct perf_output_handle *handle,
handle->offset = offset;
handle->head = head;
if (head - tail > data->watermark)
local_inc(&data->wakeup);
if (head - local_read(&data->wakeup) > data->watermark)
local_add(data->watermark, &data->wakeup);
if (have_lost) {
lost_event.header.type = PERF_RECORD_LOST;