Messages in this thread | | | Subject | Re: [PATCH -mm 1/2] irq_work, Use llist in irq_work | From | Peter Zijlstra <> | Date | Fri, 02 Sep 2011 15:26:12 +0200 |
| |
On Fri, 2011-09-02 at 09:18 +0800, Huang Ying wrote: > On 09/01/2011 06:00 PM, Peter Zijlstra wrote: > > On Thu, 2011-09-01 at 16:44 +0800, Huang Ying wrote: > > > >> Because llist is in library, it may be used in highly contended case and > >> light/un-contended loads. So maybe code as above is best choice for llist. > > > > Well the thing is, if you're heavily contended you should probably be > > doing something else.. > > So which solution is preferable? > > 1) no cpu_relax > 2) cpu_relax after first cmpxchg > > Personally, I prefer 2). It should have acceptable overhead in > ligh/un-contended loads. Do you agree.
I guess so, we can always poke at it again when someone finds something to measure.
| |