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

[Jeff Garzik <jeff@garzik.org>]

Matthew Wilcox wrote:
> On Fri, Jun 09, 2006 at 11:40:03AM -0400, Jeff Garzik wrote:
>> Users are now forced to remember that, if they write to their filesystem 
>> after using either $mmver or $korgver kernels, they are locked out of 
>> using older kernels.
>>
>> From the user's perspective, ext3 has no clear "metadata version 1", 
>> "metadata version 2" division.  Thus they are now forced to keep a 
>> matrix of kernel versions and ext3 feature flag support, to know which 
>> kernels are usable with which data.  It is a support nightmare.
> 
> Hang on, you're going too far.  You have to enable extents with the
> extent mount option.  Otherwise you don't get to use them.  The user
> does, in fact, have a clear division, although maybe the blinky signs
> aren't quite luminous enough.

...and how are distros going to deploy this?  They are going to turn on 
extents by default.

And do we honestly think that is a scalable option _anyway_?  That will 
slowly bloat fstab and mount command lines with an ever-increasing list 
of options.

It's IMO better experience for the user, and gives the developers more 
freedom.Look, I _really_ want extents.  I am a big fan.  But I think 
that extents are good time to make a clean break, and let ext3 live as 
it is.  And it will let ext3 stabilize.

	Jeff