| To: | Alex Elder <aelder@xxxxxxx> |
|---|---|
| Subject: | Re: [PATCH 1/6] xfs: preallocation transactions do not need to be synchronous |
| From: | Dave Chinner <david@xxxxxxxxxxxxx> |
| Date: | Sat, 26 Mar 2011 09:02:16 +1100 |
| Cc: | xfs@xxxxxxxxxxx |
| In-reply-to: | <1301086836.2537.679.camel@doink> |
| References: | <1300860870-15471-1-git-send-email-david@xxxxxxxxxxxxx> <1300860870-15471-2-git-send-email-david@xxxxxxxxxxxxx> <1301086836.2537.679.camel@doink> |
| User-agent: | Mutt/1.5.20 (2009-06-14) |
On Fri, Mar 25, 2011 at 04:00:36PM -0500, Alex Elder wrote: > On Wed, 2011-03-23 at 17:14 +1100, Dave Chinner wrote: > > From: Dave Chinner <dchinner@xxxxxxxxxx> > > > > Preallocation and hole punch transactions are currently synchronous > > and this is causing performance problems in some cases. The > > transactions don't need to be synchronous as we don't need to > > guarantee the preallocation is persistent on disk until a > > fdatasync, fsync, sync operation occurs. If the file is opened > > O_SYNC or O_DATASYNC, only then should the transaction be issued > > synchronously. > > There's a minor (but important) bug in this. Other than that > this looks good. > > Reviewed-by: Alex Elder <aelder@xxxxxxx> Already fixed and reposted. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx |
| Previous by Date: | Re: [PATCH 6/6] xfs: make discard operations asynchronous, Alex Elder |
|---|---|
| Next by Date: | Re: [PATCH 6/6] xfs: stop using the page cache to back the buffer cache, Dave Chinner |
| Previous by Thread: | Re: [PATCH 1/6] xfs: preallocation transactions do not need to be synchronous, Alex Elder |
| Next by Thread: | [PATCH 6/6] xfs: stop using the page cache to back the buffer cache, Dave Chinner |
| Indexes: | [Date] [Thread] [Top] [All Lists] |