On Wed, Mar 06, 2002 at 12:50:20AM +0000, Alan Cox wrote:
> > The patch is against 2.4.19-pre2. The patch significantly lowers BKL contention (50%) on a 2-way PII-300 running dbench 4. Dbench throughput is not significantly affected, but that is probably a function of my puny little machine more than the effectiveness of the patch. I'll have some results on a much more beefy 8-way PIII tomorrow. Earlier versions of the patch reduced BKL contention during dbench by 60% on the 8-way. CPU utilization spinning on the BKL has been as high as 40%.
>
> I certainly am not interested in it. 2.4 locking changes for very big boxes
> strike me as a little dangerous.
Also, there seems to be a plan already in progress to do this on 2.5. BLK
has been moved around in several filesystems to ease transition to another
lock.
-
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/
This archive was generated by hypermail 2b29 : Thu Mar 07 2002 - 21:00:51 EST