On Thu, Dec 16, 2010 at 04:30:15PM +0100, Peter Zijlstra wrote:
quoted text
> On Thu, 2010-12-16 at 16:27 +0100, Peter Zijlstra wrote:
> > On Thu, 2010-12-16 at 16:23 +0100, Frederic Weisbecker wrote:
> > 
> > > > -	TP_printk("comm=%s pid=%d prio=%d success=%d target_cpu=%03d",
> > > > +	TP_printk("comm=%s pid=%d prio=%d target_cpu=%03d",
> > > >  		  __entry->comm, __entry->pid, __entry->prio,
> > > > -		  __entry->success, __entry->target_cpu)
> > > > +		  __entry->target_cpu)
> > > 
> > > Note we'll need to fix some perf scripts after that. And also perf sched,
> > > probably perf timechart and so on...
> > 
> > Do any of those actually use the success parameter? If not, then me
> > removing it shouldn't break those tools since they're supposed to parse
> > the format stuff and not notice it missing ;-)
> 
> Alternatively we could always call trace_sched_wakeup() and make success
> reflect actual wakeup success.

So, that would work very well for sched-migration.py

builtin-sched.c would continue to work fine but would notice that as
a bug if a wake up occur on a task already running.

But it will only trigger a warning, besides that it will just continue
to work normally.

It's ok I think, if that old tool triggers a spurious warning. I suspect
very few people use it anyway.
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/