Re: [STATUS 2.5] May 1, 2002

[ebiederm@xmission.com (Eric W. Biederman)]

Stephen Lord <lord@sgi.com> writes:

> On Wed, 2002-05-01 at 16:38, Andrew Morton wrote:
> > Mike Fedyk wrote:
> > > 
> > > On Wed, May 01, 2002 at 09:53:37AM -0400, Guillaume Boissiere wrote:
> > > > new framebuffer layer, as well as some more delayed disk block
> > > > allocation bits.
> > > 
> > > Actually Andrews work on address_space based writeback is related somewhat,
> > > but really it's a rewrite/cleanup of the buffer layer.  Delayed block
> > > alocation is helped alot by this, and almost depends on it IIRC.
> > > 
> > > One vote for a seperate listing in the status for "Address Space based
> > > Writeback / Buffer layer cleanup".
> > 
> > Well the next major step here is going direct
> > pagecache<->BIO, bypassing the intermediate submit_bh
> > for most I/O.
> > 
> > Probably that will make most of the performance benefits
> > of delayed-allocate go away.
> 
> Most of the performance benefits of delayed allocate are that
> you do not the hard work of allocating the disk space in each
> write call, you get to do it once, in potentially larger chunks,
> and often not in the user's context.

Except for moving the work out of the users context, ext2 gets
a similar benefit by reserving disk space ahead of time.  So it isn't
clear that you need to have a delayed allocation to achieve this.

Eric