| To: | Christoph Hellwig <hch@xxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH 2/2] xfs: don't truncate prealloc from frequently accessed inodes |
| From: | Dave Chinner <david@xxxxxxxxxxxxx> |
| Date: | Wed, 1 Dec 2010 09:00:38 +1100 |
| Cc: | xfs@xxxxxxxxxxx |
| In-reply-to: | <20101130170301.GA5712@xxxxxxxxxxxxx> |
| References: | <1290991431-20519-1-git-send-email-david@xxxxxxxxxxxxx> <1290991431-20519-3-git-send-email-david@xxxxxxxxxxxxx> <20101130170301.GA5712@xxxxxxxxxxxxx> |
| User-agent: | Mutt/1.5.20 (2009-06-14) |
On Tue, Nov 30, 2010 at 12:03:01PM -0500, Christoph Hellwig wrote: > Did any problems show up with just trying to use an inode flag instead > of the counter? I'd really hate to bloat the inode without reason. None that I've noticed in local testing, but I haven't been focussing on this aspect so I hadn't changed it. I'll change it to a flag, and we can go back to a counter if necessary. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | [PATCH] Xfstests: make _scratch_mkfs_sized work for btrfs, Josef Bacik |
|---|---|
| Next by Date: | Re: [PATCH] Xfstest 245: accept ENOTEMPTY as a valid error for the test, Christoph Hellwig |
| Previous by Thread: | Re: [PATCH 2/2] xfs: don't truncate prealloc from frequently accessed inodes, Christoph Hellwig |
| Next by Thread: | [PATCH 1/2] xfs: dynamic speculative EOF preallocation, Dave Chinner |
| Indexes: | [Date] [Thread] [Top] [All Lists] |