Messages in this thread | | | From | Keith Owens <> | Subject | Re: [patch] 2.6.6-rc2 Allow architectures to reenable interrupts on contended spinlocks | Date | Tue, 27 Apr 2004 15:19:37 +1000 |
| |
On Mon, 26 Apr 2004 21:49:52 -0700, Andrew Morton <akpm@osdl.org> wrote: >Keith Owens <kaos@sgi.com> wrote: >> >> Enable interrupts while waiting for a disabled spinlock, but only if >> interrupts were enabled before issuing spin_lock_irqsave(). It makes a >> measurable difference to interrupt servicing on large systems. > >Do you know which are the offending locks?
Workload dependent. We already service interrupts while waiting for a non-disabled spinlock. The patch allows a cpu to do some useful work and service interrupts while waiting for disabled spinlocks as well.
>How much difference, and how large are the systems?
From memory (September 2003) 3-5% improvement on an AIM7 run, with 64 processors.
- 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/
| |