lkml.org 
[lkml]   [2001]   [Sep]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: linux-2.4.10-pre5
On Sep 08, 2001  10:30 -0700, Linus Torvalds wrote:
> I'll merge the blkdev in pagecache very early in 2.5.x, but I'm a bit
> nervous about merging it in 2.4.x.
>
> That said, if you'll give a description of how you fixed the aliasing
> issues etc, maybe I'd be less nervous. Putting it in the page cache is
> 100% the right thing to do, so in theory I'd really like to merge it
> earlier rather than later, but...

I think this may have bad interactions with the filesystems, which still
use buffer cache for metadata. If the block devices move to page cache,
so should the filesystems.

For example, the "tune2fs" program will modify parts of the superblock
from user space (fields that are read-only from the kernel, e.g. label,
reserved blocks count, etc), because it knows that the data read/written
on /dev/hda1 is coherent with that in the kernel for the filesystem
on /dev/hda1. The same is true with e2fsck - the metadata should be
kept coherent from user-space and kernel-space or bad things happen.

Cheers, Andreas
--
Andreas Dilger \ "If a man ate a pound of pasta and a pound of antipasto,
\ would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/ -- Dogbert

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:03    [W:0.126 / U:2.308 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site