Re: vfat BKL/lock_super regression in v2.6.26-rc3-g8f59342

[Bart Trojanowski <bart@jukie.net>]

* Linus Torvalds <torvalds@linux-foundation.org> [080819 20:56]:
> So one thing we could perhaps consider is to make FAT in particular 
> consider "sync" mounts to be about open/close consistency, not about 
> per-write-system-call consistency. So the "close()" wouldn't return until 
> the file is on disk, but we wouldn't force a synchronous rewrite the inode 
> or the file allocation table thousands of times just because the file was 
> big.

I was reading the vfat code, and it turns out that vfat has a "flush"
mount option.  Which is documented in the code (not in the manpage) as:

    struct fat_mount_options {
            ...
            unsigned
            ...
                     flush:1,         /* write things quickly */

Since that was very informative I looked at the usage.  It's used
in fat_file_release() to do almost what you describe.  But it seems to
be a best effort thing.  If my data doesn't hit the disk (or flash) in
HZ/10, then all bets are off.

-Bart

-- 
				WebSig: http://www.jukie.net/~bart/sig/