On Sat, 11 Mar 2000, Alan Cox wrote:
> Fix module remove race bug
Infrastructure is in place now and I'm going through the rest of
fs-related stuff. Then it will be a lot of mess with drivers/ldiscs/protocol
families.
> Truncate races (Debian apt shows it nicely) [done ?]
99% done. Remaining 1% is the i_size handling in CODA. We have almost
everything to do it now - it's on the list.
> Protection on isize (sct)
? It's mostly done now. If Stephen is going to return to that stuff -
fine, but then we'ld better go through the code together - I've tweaked it
a lot and will touch it one more time (see above).
> NTFS needs updating/binning or something
> rw semaphores on inodes to fix read/truncate races ?
Probably already fixed, but I'll have to recheck it.
> AFFS doesn't work on current page cache
Other things to consider:
a) NTFS is unmaintained and I'ld just bitbucket the current variant
(sharing codebase between the kernel and userland library is *WHAM* bad
*WHAM* idea *WHACK* *SPLASH*).
b) AFFS... Well, there are different issues with it. OFS (aka "we don't
need no stinkin' powers of 2, 488 bytes of data in sector is just fine")
would be better off in some equivalent of mtools. IOW, in the userland.
Floppies are floppies... Now, the rest is not that hard to update for
pagecache, but handling the directories is going to be hell.
c) HFS is still b0rken. Not too hard to fix, though.
d) iget() abuse in knfsd. Being worked on.
e) UMSDOS. Sigh... I'll have to touch it anyway (dcache stuff), but I
didn't really look into it yet.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:19 EST