Re: [PATCH] ftrace, sched: Add TRACE_FLAG_PREEMPT_RESCHED

[Peter Zijlstra <peterz@infradead.org>]

On Wed, Nov 06, 2013 at 11:37:04AM -0500, Steven Rostedt wrote:
> On Fri, 27 Sep 2013 17:29:08 +0200
> Peter Zijlstra <peterz@infradead.org> wrote:
> 
> > Subject: ftrace, sched: Add TRACE_FLAG_PREEMPT_RESCHED
> > From: Peter Zijlstra <peterz@infradead.org>
> > Date: Fri Sep 27 17:11:00 CEST 2013
> > 
> > Since we now have two need_resched states; trace the two so we can
> > observe discrepancies.
> 
> I see this is dependent on the addition of tif_need_resched() and
> friends.

This also wasn't the last version of the patch, I distinctly remember
you making me update some documentation crap.

> > +
> > +	if ((entry->flags & TRACE_FLAG_NEED_RESCHED) &&
> > +	    (entry->flags & TRACE_FLAG_PREEMPT_RESCHED))
> > +		need_resched = 'N';
> > +	else if (entry->flags & TRACE_FLAG_NEED_RESCHED)
> > +		need_resched = 'n';
> > +	else if (entry->flags & TRACE_FLAG_PREEMPT_RESCHED)
> > +		need_resched = 'p';
> > +	else
> > +		need_resched = '.';
> 
> Perhaps we should make this a switch statement?
> 
> 	switch (entry->flags & (TRACE_FLAG_NEED_RESCHED |
> 				TRACE_FLAG_PREEMPT_RESCHED)) {
> 	case TRACE_FLAG_NEED_RESCHED | TRACE_FLAG_PREEMPT_RESCHED:
> 		need_resched = 'N';
> 		break;
> 	case TRACE_FLAG_NEED_RESCHED:
> 		need_resched = 'n';
> 		break;
> 	case TRACE_FLAG_PREEMPT_RESCHED:
> 		need_resched = 'p';
> 		break;
> 	default:
> 		need_resched = '.';
> 		break;
> 	}
> 
> Simply because I find switch statements easier to read than else if
> statements.

Sure whatever your code ;-), but says he who wrote:

	hardsoft_irq =
		(hardirq && softirq) ? 'H' :
		hardirq ? 'h' :
		softirq ? 's' :
		'.';

Just apply the later patch and change it however you like.