Re: [STATUS 2.5] May 1, 2002

[Andrew Morton <akpm@zip.com.au>]

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.

There are other reasons for implementing delalloc
(XFS, improved layout, ...).  So it ain't dead yet.


At 48 bytes, 2.5's buffer_head is now precisely half the
size of 2.4's.  I'm hoping to be able to shed another eight
bytes yet.

With the pagecache<->BIO change, the buffer_head will most
definitely become "per-page metadata which describes the state
of sub-page segments" and not "something which is used for
performing I/O".



-