| To: | "David Chinner" <dgc@xxxxxxx> |
|---|---|
| Subject: | Re: [PATCH 4 of 4] Per Allocation Group inode radix trees |
| From: | nscott@xxxxxxxxxx |
| Date: | Thu, 9 Aug 2007 10:14:40 +1000 (EST) |
| Cc: | "xfs-dev" <xfs-dev@xxxxxxx>, "xfs-oss" <xfs@xxxxxxxxxxx> |
| Importance: | Normal |
| In-reply-to: | <20070808231310.GS12413810@xxxxxxx> |
| References: | <20070808231310.GS12413810@xxxxxxx> |
| Sender: | xfs-bounce@xxxxxxxxxxx |
| User-agent: | SquirrelMail/1.4.8-4.el4.centos |
> ... > operations by keeping structures and operations within AGs. This > removes the need for the ihashsize mount parameter altogether > as the inode caches should never need size hints anymore. >... > --- 2.6.x-xfs-new.orig/fs/xfs/xfs_vfsops.c 2007-08-01 15:43:30.383433897 > +1000 > +++ 2.6.x-xfs-new/fs/xfs/xfs_vfsops.c 2007-08-01 15:43:34.150945479 +1000 > ... > -#define MNTOPT_IHASHSIZE "ihashsize" /* size of inode hash table */ You probably want to keep this string & parsing for awhile (forever?), but move it to the end of the list with the other deprecated options - just so as to not fail mounts on kernel upgrades... and surprise/annoy sysadmins the world over. cheers. -- Nathan |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Impact of atime updates on XFS, David Chinner |
|---|---|
| Next by Date: | Re: Review: Factor some common freelist checks, Christoph Hellwig |
| Previous by Thread: | [PATCH 4 of 4] Per Allocation Group inode radix trees, David Chinner |
| Next by Thread: | Review: Factor some common freelist checks, David Chinner |
| Indexes: | [Date] [Thread] [Top] [All Lists] |