[PATCH 0/5] reduce exclusive ilock hold times V2

[Dave Chinner]

On Tue, Mar 27, 2012 at 10:34:45AM -0400, Christoph Hellwig wrote:
> This series tries to reduce the amount we hold the ilock exclusively,
> especially during direct I/O writes where they currently hurt us.
> 
> Dave showed that his earlier version which is less aggressive than this
> one can already provide magnitudes of better throughput and iops for
> parallel direct I/O workloads, and this one should be even better.
> 
> Changes from V1:
>  - do not mark xfs_qm_need_dqattach as inline
>  - various comment and commit message updates

Consider the whole series:

Reviewed-by: Dave Chinner <dchinner at redhat.com>

> I specificly did not add the delalloc assert in the direct I/O write path
> in this series, as it triggers with or without this patch.  I will look into
> that issue next.

I'm getting that assert (and other delalloc block asserts) when
fsstress is running quite often these days. I suspect that he lack
of IOLOCK synchronisation in .page_mkwrite is biting us here, but
I'm interested to know what you find...

Cheers,

Dave.
-- 
Dave Chinner
david at fromorbit.com