[lkml]   [2019]   [Mar]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [LKP] [page cache] eb797a8ee0: vm-scalability.throughput -16.5% regression
Linus Torvalds <> writes:

> On Wed, Feb 27, 2019 at 5:19 PM Huang, Ying <> wrote:
>> So I think in the heavily contended situation, we should put the fields
>> accessed by rwsem holder in a different cache line of rwsem. But in
>> un-contended situation, we should put the fields accessed in rwsem
>> holder and rwsem in the same cache line to reduce the cache footprint.
>> The requirement of un-contended and heavily contended situation is
>> contradicted.
> Generally, we should strive to optimize for the uncontended state.
> The performance profile of a contended state tends to be very
> different, and the actual solution tends to be to try really hard to
> just avoid contention to begin with.
> I think we've gotten to the point where we have very few real loads
> that show lock contention on a kernel level. And when people do find
> loads that cause contention, we should try really hard to fix the
> locking rather than try to then treat the symptom of contention.
> So on the whole, aim to make the uncontended case go fast, at least to
> a first approximation.

Sounds reasonable! Thanks for clarification!

Best Regards,
Huang, Ying

> Linus

 \ /
  Last update: 2019-03-02 09:27    [W:0.060 / U:1.548 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site