On Fri, 8 Feb 2002, Andrew Morton wrote:
> I'd be interested in hearing more details on the regression which Ingo
> has seen due to the introduction of i_sem locking in llseek. [...]
i saw heavy scheduling during dbench runs (even if just running 6 threads
on an 8 CPU box), and checked out the source of the scheduling storm -
most of it was due to llseek()'s down().
i also wrote a dbench-alike load simulator for pagecache scalability,
there i saw this in an even more prominent way, 200k/sec reschedules in a
situation when there should be none.
i'd suggest 64-bit update instructions on x86 as well, they do exist.
spinlock only for the truly hopeless cases like SMP boxes composed of
i486's. We really want llseek() to scale ...
Ingo
-
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 : Fri Feb 15 2002 - 21:00:20 EST