On Wed, 2001-09-19 at 21:13, David Lang wrote:
> also how useful is the latency info once swapping starts? if you have
> something that gets swapped out it will have horrible latency and skew the
> results (or at the very least you won't know if the problem is disk IO or
> fixable problems)
Its not that bad because the patch measures latency across points where
preemption was enabled and disabled -- the various kernel locking
mechanisms.
Thus either page fault holds a lock and it is a legimate latency
recorded over that specific lock, or it doesn't in which case its not an
issue.
-- Robert M. Love rml at ufl.edu rml at tech9.net- 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 : Sun Sep 23 2001 - 21:00:35 EST