lkml.org 
[lkml]   [2013]   [Feb]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH -v4 4/5] x86,smp: keep spinlock delay values per hashed spinlock address
From
On Wed, Feb 6, 2013 at 12:10 PM, Rik van Riel <riel@redhat.com> wrote:
> On 01/27/2013 08:04 AM, Michel Lespinasse wrote:
>>
>> On Fri, Jan 25, 2013 at 11:18 AM, Rik van Riel <riel@redhat.com> wrote:
>>>
>>> + u32 delay = (ent->hash == hash) ? ent->delay :
>>> MIN_SPINLOCK_DELAY;
>>
>> I still don't like the reseting of delay to MIN_SPINLOCK_DELAY when
>> there is a hash collision.
>
> I've been spending some time looking at this, because I am
> not a fan either.
>
> However, it seems to work and I failed to come up with
> anything better. Therefore, I have left it as is in the
> -v5 patch series.

Does that mean you know of workloads that would regress if you didn't
reset the delay to MIN_SPINLOCK_DELAY when detecting a hash collision
?

I have not seen any public reports of that...

--
Michel "Walken" Lespinasse
A program is never fully debugged until the last user dies.


\
 
 \ /
  Last update: 2013-02-10 01:21    [W:0.061 / U:0.072 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site