Peter Zijlstra 9b231d9f47 perf/core: Fix time on IOC_ENABLE ... Vince reported that when we do IOC_ENABLE/IOC_DISABLE while the task is SIGSTOP'ed state the timestamps go wobbly. It turns out we indeed fail to correctly account time while in 'OFF' state and doing IOC_ENABLE without getting scheduled in exposes the problem. Further thinking about this problem, it occurred to me that we can suffer a similar fate when we migrate an uncore event between CPUs. The perf_event_install() on the 'new' CPU will do add_event_to_ctx() which will reset all the time stamp, resulting in a subsequent update_event_times() to overwrite the total_time_* fields with smaller values. Reported-by: Vince Weaver <vincent.weaver@maine.edu> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@kernel.org>		2017-08-10 12:01:09 +02:00
..
callchain.c	perf/callchain: Force USER_DS when invoking perf_callchain_user()	2017-05-10 07:54:00 +02:00
core.c	perf/core: Fix time on IOC_ENABLE	2017-08-10 12:01:09 +02:00
hw_breakpoint.c	perf: Collapse and fix event_function_call() users	2016-01-21 18:54:24 +01:00
internal.h	bpf, events: fix offset in skb copy handler	2016-07-25 10:34:11 -07:00
Makefile	ftrace: allow architectures to specify ftrace compile options	2015-01-29 09:19:19 +01:00
ring_buffer.c	perf/aux: Correct return code of rb_alloc_aux() if !has_aux(ev)	2017-06-21 11:58:30 +02:00
uprobes.c	sched/headers: Prepare for new header dependencies before moving code to <linux/sched/coredump.h>	2017-03-02 08:42:28 +01:00