Re: [RFC] reserved inode fields to use for i_blocks_high

On Jun 05, 2006  13:11 +0100, Stephen C. Tweedie wrote:
> On Fri, 2006-06-02 at 16:30 -0600, Andreas Dilger wrote:
> > If we have 2 bits of epoch extension with the nsecond timestamps,
> > then we can hope everyone has updated away from the 128-byte inodes
> > in the next 270 years, and they can get larger timestamps still...
> 
> I don't understand --- if we couple epoch extension with the nanosecond
> timestamps, then it will *not* be available to 128-byte inodes and so we
> only have 32 years left.

Ah, true.  When I was thinking "128-byte" inodes I was really thinking
"large inodes with 128-byte+i_extra_size fixed inode", instead of the
(highly theoretical right now) full-64-bit inode (256-bytes+ fixed size).

The current "large inode" support is existing since 2.5.44 kernels (Oct
2002) and 2.4.24 kernels (Jan 2004) so changing e2fsprogs to format
filesystems with 256-byte inodes by default would be quite widely
compatible. It becomes less backward compatible if you start storing
EAs in the large inode space, 2.6.10 = Nov 2004, not sure about 2.4.

Using Ted's new "/etc/mke2fs.opt" (or whatever) file if we don't want
to change it in the default mke2fs would be one way to do this for a
distro without enabling it for all e2fsprogs users.

Cheers, Andreas
--
Andreas Dilger
Principal Software Engineer
Cluster File Systems, Inc.