Messages in this thread | | | From | Andi Kleen <> | Subject | Re: [Patch 2/3] fast VMA recycling | Date | Thu, 23 Feb 2006 12:57:46 +0100 |
| |
On Thursday 23 February 2006 12:22, Arjan van de Ven wrote: > Andi Kleen wrote: > >> see voluntary preempt. > > > > Only when its time slice is used up > > or if some other thread gets a higher dynamic prio > > but then it would sleep a bit later > > in user space. > > ... but that is without the semaphore held! (and that is the entire > point of this patch, move the sleep moments to outside the lock holding > area as much as possible, to reduce lock hold times)
And you verified this happens often in your workload?
Anyways, how about adding a down_no_preempt() or similar instead that won't voluntarily preempt?
-Andi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |