Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1620145

Re: [PATCH] x86/irq: Do not check available vectors if current CPU has no irq to migrate

From Chen Yu <yu.c.chen@intel.com>
Newsgroups linux.kernel
Subject Re: [PATCH] x86/irq: Do not check available vectors if current CPU has no irq to migrate
Date 2017-04-10 18:00 +0200
Message-ID <tuJTw-2D4-103@gated-at.bofh.it> (permalink)
References <tulxL-3uX-1@gated-at.bofh.it> <tuFGa-8rl-21@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


Hi Prarit,
On Mon, Apr 10, 2017 at 07:26:31AM -0400, Prarit Bhargava wrote:
> 
> 
> On 04/09/2017 09:53 AM, Chen Yu wrote:
> > This is an optimization to bypass the free vector checking if the current
> > CPU has no irq to migrate. This can especially speed up the CPU offline
> > process when there are many CPUs in the system.
> 
> Chen, OOC how much time does this save?
> 
Previously I was doing some debugging on irq related issue
when check_irq_vectors_for_cpu_disable() printed a warning
on that platform, thus I incidentally look at this code and
did not measure it : )

I've just done a comparison of the function duration of
check_irq_vectors_for_cpu_disable() with/without the patch
(32 online CPUs)
before patch:
[001] d..1   264.017803: check_irq_vectors_for_cpu_disable <-native_cpu_disable
[001] d..1   264.017886: clear_local_APIC <-native_cpu_disable
after patch:
[001] d..1   862.011847: check_irq_vectors_for_cpu_disable <-native_cpu_disable
[001] d..1   862.011850: clear_local_APIC <-native_cpu_disable
About  (83 - 17) = 66 microsecond.

thanks,
Yu
> P.
> 
> > 
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > Cc: Ingo Molnar <mingo@redhat.com>
> > Cc: "H. Peter Anvin" <hpa@zytor.com>
> > Cc: Prarit Bhargava <prarit@redhat.com>
> > Cc: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>
> > Cc: Len Brown <len.brown@intel.com>
> > Cc: x86@kernel.org
> > Cc: linux-kernel@vger.kernel.org
> > Signed-off-by: Chen Yu <yu.c.chen@intel.com>
> > ---
> >  arch/x86/kernel/irq.c | 3 +++
> >  1 file changed, 3 insertions(+)
> > 
> > diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
> > index 4d8183b..b9bdc1e 100644
> > --- a/arch/x86/kernel/irq.c
> > +++ b/arch/x86/kernel/irq.c
> > @@ -394,6 +394,9 @@ int check_irq_vectors_for_cpu_disable(void)
> >  		    !cpumask_subset(&affinity_new, &online_new))
> >  			this_count++;
> >  	}
> > +	/* No need to check any further. */
> > +	if (!this_count)
> > +		return 0;
> >  
> >  	count = 0;
> >  	for_each_online_cpu(cpu) {
> > 

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


Thread

[PATCH] x86/irq: Do not check available vectors if current CPU has no irq to migrate Chen Yu <yu.c.chen@intel.com> - 2017-04-09 16:00 +0200
  Re: [PATCH] x86/irq: Do not check available vectors if current CPU  has no irq to migrate Prarit Bhargava <prarit@redhat.com> - 2017-04-10 13:30 +0200
    Re: [PATCH] x86/irq: Do not check available vectors if current CPU  has no irq to migrate Chen Yu <yu.c.chen@intel.com> - 2017-04-10 18:00 +0200
      Re: [PATCH] x86/irq: Do not check available vectors if current CPU  has no irq to migrate Thomas Gleixner <tglx@linutronix.de> - 2017-04-14 22:40 +0200

csiph-web