From: Steven Rostedt (VMware) <[email protected]> commit 55ea4cf403800af2ce6b125bc3d853117e0c0456 upstream.
The write stamp, used to calculate deltas between events, was updated with the stale "ts" value in the "info" structure, and not with the updated "ts" variable. This caused the deltas between events to be inaccurate, and when crossing into a new sub buffer, had time go backwards. Link: https://lkml.kernel.org/r/[email protected] Cc: [email protected] Fixes: a389d86f7fd09 ("ring-buffer: Have nested events still record running time stamp") Reported-by: "J. Avila" <[email protected]> Tested-by: Daniel Mentz <[email protected]> Tested-by: Will McVicker <[email protected]> Signed-off-by: Steven Rostedt (VMware) <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]> --- kernel/trace/ring_buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -3291,7 +3291,7 @@ __rb_reserve_next(struct ring_buffer_per /* Nothing came after this event between C and E */ info->delta = ts - info->after; (void)rb_time_cmpxchg(&cpu_buffer->write_stamp, - info->after, info->ts); + info->after, ts); info->ts = ts; } else { /*
