Re: [PATCH v2 1/2] trace: Default to using trace_global_clock if sched_clock is unstable

From: Steven Rostedt
Date: Mon Apr 02 2018 - 11:17:44 EST

On Fri, 30 Mar 2018 16:01:31 +0100
Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> wrote:

> Across suspend, we may see a very large drift in timestamps if the sched
> clock is unstable, prompting the global trace's ringbuffer code to warn
> and suggest switching to the global clock. Preempt this request by
> detecting when the sched clock is unstable (determined during
> late_initcall) and automatically switching the default clock over to
> trace_global_clock.
> This should prevent requiring user interaction to resolve warnings such
> as:
> Delta way too big! 18446743856563626466 ts=18446744054496180323 write stamp = 197932553857
> If you just came from a suspend/resume,
> please switch to the trace global clock:
> echo global > /sys/kernel/debug/tracing/trace_clock
> Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Cc: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
> ---
> v2: Tell the user what's happening and what they can do to correct it.

Thanks, I pulled them in. I'll start testing them before I post to git.

-- Steve