Re: in_irq_or_nmi() and RFC patch

From	Mel Gorman <mgorman@techsingularity.net>
Newsgroups	linux.kernel
Subject	Re: in_irq_or_nmi() and RFC patch
Date	2017-04-05 11:00 +0200
Message-ID	<tsOXg-nu-21@gated-at.bofh.it> (permalink)
References	(5 earlier) <tqrip-2Ue-1@gated-at.bofh.it> <tqrLs-34R-17@gated-at.bofh.it> <tqHZT-6On-1@gated-at.bofh.it> <tqJS1-8b4-5@gated-at.bofh.it> <ts8Y3-6GK-57@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Mon, Apr 03, 2017 at 01:05:06PM +0100, Mel Gorman wrote:
> > Started performance benchmarking:
> >  163 cycles = current state
> >  183 cycles = with BH disable + in_irq
> >  218 cycles = with BH disable + in_irq + irqs_disabled
> > 
> > Thus, the performance numbers unfortunately looks bad, once we add the
> > test for irqs_disabled().  The slowdown by replacing preempt_disable
> > with BH-disable is still a win (we saved 29 cycles before, and loose
> > 20, I was expecting regression to be only 10 cycles).
> > 
> 
> This surprises me because I'm not seeing the same severity of problems
> with irqs_disabled. Your path is slower than what's currently upstream
> but it's still far better than a revert. The softirq column in the
> middle is your patch versus a full revert which is the last columnm
> 

Any objection to resending the local_bh_enable/disable patch with the
in_interrupt() check based on this data or should I post the revert and
go back to the drawing board?

-- 
Mel Gorman
SUSE Labs

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Re: in_irq_or_nmi() Peter Zijlstra <peterz@infradead.org> - 2017-03-29 10:20 +0200
  Re: in_irq_or_nmi() Jesper Dangaard Brouer <brouer@redhat.com> - 2017-03-29 11:10 +0200
    Re: in_irq_or_nmi() Peter Zijlstra <peterz@infradead.org> - 2017-03-29 11:30 +0200
      Re: in_irq_or_nmi() Matthew Wilcox <willy@infradead.org> - 2017-03-29 20:20 +0200
        Re: in_irq_or_nmi() Jesper Dangaard Brouer <brouer@redhat.com> - 2017-03-29 21:20 +0200
          Re: in_irq_or_nmi() and RFC patch Jesper Dangaard Brouer <brouer@redhat.com> - 2017-03-29 21:50 +0200
            Re: in_irq_or_nmi() and RFC patch Peter Zijlstra <peterz@infradead.org> - 2017-03-30 09:00 +0200
              Re: in_irq_or_nmi() and RFC patch Jesper Dangaard Brouer <brouer@redhat.com> - 2017-03-30 09:20 +0200
                Re: in_irq_or_nmi() and RFC patch Peter Zijlstra <peterz@infradead.org> - 2017-03-30 09:40 +0200
                Re: in_irq_or_nmi() and RFC patch Jesper Dangaard Brouer <brouer@redhat.com> - 2017-03-30 11:50 +0200
            Re: in_irq_or_nmi() and RFC patch Mel Gorman <mgorman@techsingularity.net> - 2017-03-30 15:10 +0200
              Re: in_irq_or_nmi() and RFC patch Jesper Dangaard Brouer <brouer@redhat.com> - 2017-03-30 17:10 +0200
                Re: in_irq_or_nmi() and RFC patch Mel Gorman <mgorman@techsingularity.net> - 2017-04-03 14:10 +0200
                Re: in_irq_or_nmi() and RFC patch Mel Gorman <mgorman@techsingularity.net> - 2017-04-05 11:00 +0200

csiph-web