> interrupt changes
I would remove the interrupt fix from the list.
>with their brain firmly wrapped around the page cache/vfs verify this change
>that was made is absolutely safe
It's absolutely safe. read_swap_cache and lookup_swap_cache can't return a
not swap cache page. See read_swap_cache_async and lookup_swap_cache(). If
the patch made a difference it means that there's a race somewhere (but
I really don't think it's the case).
I never had corruption here so my guess it that the corruption isn't
related to the kernel proper (or that I fixed the bug by mistake in my
2.2.x VM patches but I don't think so). I had a fast review of the
2.2.7->2.2.9 changes too last week and I noticed nothing of dubious in the
kernel proper/memory-management (I didn't looked into the other things
though).
I'll did an & between your list with the things that I don't use
intensively here, the remaining suspects are:
o Quota - which has big 2.2.7->2.2.9 changes.
o The small scsi changes (dubious)
o TCP changes
o IRDA
o NFS
BTW, do you have all .config of people who reported fs corruption? An &
between all config options would be interesting :).
(2.2.10 has also the forgotten-write fix in threading from Stephen, it
looks perfectly ok to me though)
Andrea Arcangeli
-
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/