I've been thinking recently that we should re-enable DIR_INDEX in mke2fs by default. When it first came out, we had done this and were bitten by a few bugs in the code. However, this code has been in heavy use for several thousand filesystem years in Lustre, if not elsewhere, and I'm inclined to think it is pretty safe these days. Likewise, RHEL/FC have had RESIZE_INODE as a standard feature for a good time now, and this should probably be merged into stock e2fsprogs. Comments? Cheers, Andreas -- Andreas Dilger Principal Software Engineer Cluster File Systems, Inc.
Stephen C. Tweedie
2006-Mar-17 22:26 UTC
[RFC] mke2fs with DIR_INDEX, RESIZE_INODE by default
Hi, On Fri, 2006-03-17 at 00:53 -0700, Andreas Dilger wrote:> I've been thinking recently that we should re-enable DIR_INDEX in mke2fs > by default. When it first came out, we had done this and were bitten by > a few bugs in the code. However, this code has been in heavy use for > several thousand filesystem years in Lustre, if not elsewhere, and I'm > inclined to think it is pretty safe these days.I reckon they are safe enough for general use. The only question mark in my mind is over the change in behaviour for people who dual-boot or swap data between newer and older distros. One way around that that I've been wondering about would be to wait until we have accumulated enough new features (extent maps/64-bit, increase the default inode size etc.) and give the new feature set its own explicit flag in mke2fs. It might be something we could call ext4 (ie. enable it if mke2fs is called as "mke4fs"); we might just add a separate flag. Whatever way we chose, it would want to be something that would stand out as an obviously non-backwards-compatible formatting option. If we do want to do that, then there's less reason to want to enable small bits of that feature set in a piece-meal fashion. --Stephen