Re: [PATCH v2 0/4] /proc/stat: Reduce irqs counting performance overhead

[Waiman Long]

On 01/09/2019 12:44 PM, Matthew Wilcox wrote:
> On Wed, Jan 09, 2019 at 12:23:44PM -0500, Waiman Long wrote:
>> This v2 patch optimizes the way the IRQ counts are retrieved and computed
>> and getting rid of the sysctl parameter altogether to achieve a performance
>> gain that is close to the v1 patch. This is based on the idea that while
>> many IRQs can be supported by a system, only a handful of them are actually
>> being used in most cases. We can save a lot of time by focusing on those
>> active IRQs only and ignore the rests.
> So your reaction to being told "Make this the same as every other thing
> we have to sum across all CPUs" is to make it _even more different_ and
> special-cased?  I'm done.  NAK.

The paragraph above may be a bit misleading. This v2 patch actually
touches very little on percpu accounting aspect of the IRQ counts. See
patches 2 and 3 for the relevant changes which is just a few line of new
codes. Please review the individual patches before Nak'ing.

I could theoretically generalize them into a new set of percpu counting
helpers, but the idea behind it is quite different from the use cases of
percpu counter. So it may not be a good idea of adding it to there.

Cheers,
Longman