Skip to content

Commit

Permalink
perf: Add PERF_EVENT_STATE_EXIT state for events with exited task
Browse files Browse the repository at this point in the history
Adding new perf event state to indicate that the monitored task has
exited.  In this case the event stays alive until the owner task exits
or close the event fd while providing the last data through the read
syscall and ring buffer.

Instead it needs to propagate the error info (monitored task has died)
via poll and read  syscalls by  returning POLLHUP and 0 respectively.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20140811120102.GY9918@twins.programming.kicks-ass.net
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jean Pihet <jean.pihet@linaro.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/n/tip-t5y3w8jjx6tfo5w8y6oajsjq@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
  • Loading branch information
Jiri Olsa authored and Arnaldo Carvalho de Melo committed Aug 24, 2014
1 parent 61b6768 commit 179033b
Show file tree
Hide file tree
Showing 2 changed files with 10 additions and 1 deletion.
1 change: 1 addition & 0 deletions include/linux/perf_event.h
Original file line number Diff line number Diff line change
Expand Up @@ -269,6 +269,7 @@ struct pmu {
* enum perf_event_active_state - the states of a event
*/
enum perf_event_active_state {
PERF_EVENT_STATE_EXIT = -3,
PERF_EVENT_STATE_ERROR = -2,
PERF_EVENT_STATE_OFF = -1,
PERF_EVENT_STATE_INACTIVE = 0,
Expand Down
10 changes: 9 additions & 1 deletion kernel/events/core.c
Original file line number Diff line number Diff line change
Expand Up @@ -3600,7 +3600,8 @@ perf_read_hw(struct perf_event *event, char __user *buf, size_t count)
* error state (i.e. because it was pinned but it couldn't be
* scheduled on to the CPU at some point).
*/
if (event->state == PERF_EVENT_STATE_ERROR)
if ((event->state == PERF_EVENT_STATE_ERROR) ||
(event->state == PERF_EVENT_STATE_EXIT))
return 0;

if (count < event->read_size)
Expand Down Expand Up @@ -3630,6 +3631,10 @@ static unsigned int perf_poll(struct file *file, poll_table *wait)
unsigned int events = POLLHUP;

poll_wait(file, &event->waitq, wait);

if (event->state == PERF_EVENT_STATE_EXIT)
return events;

/*
* Pin the event->rb by taking event->mmap_mutex; otherwise
* perf_event_set_output() can swizzle our rb and make us miss wakeups.
Expand Down Expand Up @@ -7588,6 +7593,9 @@ __perf_event_exit_task(struct perf_event *child_event,
if (child_event->parent) {
sync_child_event(child_event, child);
free_event(child_event);
} else {
child_event->state = PERF_EVENT_STATE_EXIT;
perf_event_wakeup(child_event);
}
}

Expand Down

0 comments on commit 179033b

Please sign in to comment.