Re: [PATCH v2 6/6] stm class: ftrace: use different channel accroding to CPU

From: Tingwei Zhang
Date: Mon Jul 27 2020 - 20:28:40 EST


Hi Steve,
On Tue, Jul 28, 2020 at 03:08:12AM +0800, Steven Rostedt wrote:
> On Sun, 26 Jul 2020 10:59:31 +0800
> Tingwei Zhang <tingwei@xxxxxxxxxxxxxx> wrote:
>
> > --- a/drivers/hwtracing/stm/ftrace.c
> > +++ b/drivers/hwtracing/stm/ftrace.c
> > @@ -37,8 +37,9 @@ static void notrace
> > stm_ftrace_write(struct trace_export *export, const void *buf, unsigned
> int len)
> > {
> > struct stm_ftrace *stm = container_of(export, struct stm_ftrace,
> ftrace);
> > + unsigned int cpu = smp_processor_id();
>
> Probably should add a comment to the above stating that this is called
> from the tracing system with preemption disabled.
>

Good point. I'll add that comment in next revision.

Thanks,
Tingwei

> Other than my two comments:
>
> Reviewed-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
>
> -- Steve
>
>
> >
> > - stm_source_write(&stm->data, STM_FTRACE_CHAN, buf, len);
> > + stm_source_write(&stm->data, STM_FTRACE_CHAN + cpu, buf, len);
> > }
> >
> > static int stm_ftrace_link(struct stm_source_data *data)