Re: [RFC 0/13] extents and 48bit ext3

On 6/9/06, Joel Becker <[email protected]> wrote:
> On Fri, Jun 09, 2006 at 04:50:36PM -0400, Dave Jones wrote:
> > On Fri, Jun 09, 2006 at 01:38:03PM -0700, Joel Becker wrote:
> >  > that the older code cannot read.  Alex claims people just shouldn't use
> >  > "-o extents", but the fact is their distro will choose it for them.
> >
> > .. on partitions over a certain size, which couldn't be read with
> > older ext3 filesystems _anyway_
>
>         Certainly that would be fine.  Is that what will actually
> happen?  Experience says no.  Even if you get it right in your distro,
> not all distros will.  Heck, can you promise me that your distro will
> provide e2fsprogs updates to its older releases so that multiboot will
> continue to work?

If the kernel were bound by all the stakeholders' ability to _always_
"do the right thing" very little innovation would be possible.  These
tenuous arguments of hypothetical (ab)users are tiresome.

If the distro vendor did default to ext3+extents and it screwed your
hypothetical extents-naive user (booting a non-vendor kernel isn't
something your mom is going to do) then they strayed too far from
their Linux comfort-zone. If worst came to worst _THE UPDATED
EXT3UTILS WOULD PREVENT MOUNTING AN EXT3 FS WITH AN INCOMPATIBLE
FEATURE_.  God forbid the naive-user get an error when they try
something they shouldn't.