[lkml]   [2010]   [Dec]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [RFC PATCH 2/3] sched: add yield_to function
On 12/03/2010 04:23 PM, Peter Zijlstra wrote:
> On Fri, 2010-12-03 at 19:40 +0530, Srivatsa Vaddagiri wrote:
>> On Fri, Dec 03, 2010 at 07:36:07PM +0530, Srivatsa Vaddagiri wrote:
>>> On Fri, Dec 03, 2010 at 03:03:30PM +0100, Peter Zijlstra wrote:
>>>> No, because they do receive service (they spend some time spinning
>>>> before being interrupted), so the respective vruntimes will increase, at
>>>> some point they'll pass B0 and it'll get scheduled.
>>> Is that sufficient to ensure that B0 receives its fair share (1/3 cpu in this
>>> case)?
>> Hmm perhaps yes, althought at cost of tons of context switches, which would be
>> nice to minimize on?
> Don't care, as long as the guys calling yield_to() pay for that time its
> their problem.

Also, the context switches are cheaper than spinning
entire time slices on spinlocks we're not going to get
(because the VCPU holding the lock is not running).

All rights reversed

 \ /
  Last update: 2010-12-04 14:05    [from the cache]
©2003-2014 Jasper Spaans. hosted at Digital Ocean