perf: Add PERF_EVENT_STATE_EXIT state for events with exited task (179033b3) · Commits · 戴 / test

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>

include/linux/perf_event.h

+1 −0

Original line number	Diff line number	Diff line
		@@ -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,

kernel/events/core.c

+9 −1

Original line number	Diff line number	Diff line
		@@ -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)
		@@ -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.
		@@ -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);
		}
		}

Admin message