Re: [patch V5 03/38] nmi, tracing: Provide nmi_enter/exit_notrace()

[Steven Rostedt]

On Thu, 14 May 2020 21:35:22 -0400
Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:

> On Thu, 14 May 2020 21:32:30 -0400
> Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> 
> > It's accounting only. Now it does use a per-cpu variable (which I see
> > in other patches can be an issue) and calls time_get().  
> 
> time_get() being defined as trace_clock_local(). And this in only
> called if CONFIG_GENERIC_SCHED_CLOCK is not true, as that isn't NMI
> safe.
>

Forget what I said. After reading the next patch, this makes much more sense. ;-)

Acked-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>

-- Steve