Rebooted, verified, looks good.
Btw, Ingo, I really like how the ext2 changes are turning out: especially
the last set of changes were a real cleanup. The buffer head could still
be passed down some more, and then we could get rid of the "*new" thing
altogether, and make ext2 internally use the same much cleaner setup. Nice
work, Ingo.
The other comment I have is that I think that the _filesystem_ is the one
that should set the "bh->b_dev" thing, and conceptually what we should
have is that a unmapped buffer has neither a device nor a blocknr, and
then when we ask the low-level FS to map it for us, we could in theory
have filesystems that span multiple devices (so the VFS layer should _not_
set bh->b_dev blindly to inode->i_dev). I've done that in both pre-4 and
pre-5, I just wanted to point out that it's by design.
(Our ll_rw_block() cannot handle multi-device requests, but we'll see if
we ever need them before I would fix that up).
Linus
-
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/