Messages in this thread | | | Date | Wed, 18 May 2005 16:38:38 +0200 | From | Carsten Otte <> | Subject | Re: [RFC/PATCH 4/5] loop: execute in place (V2) |
| |
Christoph Hellwig wrote:
>On Wed, May 18, 2005 at 03:53:52PM +0200, Carsten Otte wrote: > > >>[RFC/PATCH 4/5] loop: execute in place (V2) >> >> >> > >This should be ifdef'ed to avoid bloat for non-XIP builds. Or just be dropped >completely. How much difference does it make over read/write and where does >loop performance matter? > > I don't think loop on xip is performance critical. For page cache lookup I see a performance difference of factor 2 on our platform because we have decent memory bandwidth and lock contention slows things down with many CPUs. Given that even without this patch we don't do page cache lookups, I don't think there's much difference. Initially this patch was written for the old loop driver that won't work without this patch... Guess that dropping it is a good idea.
- 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/
| |