> > added/enabled by the new pagecache: now file writes can happen in
> > parallel, even at once (if on an SMP box).
> >
> > Until 2.3.7 all file writes were serialized by the 'inode semaphore', ie.
> > one big write excluded all other writers until the big write finished.
>
> Caution: side effect of that change was not pretty. truncate() used to be
> serialized wrt write(). Now it isn't. We might contract some races here.
check out block_*_write(), we get the inode semaphore when we allocate new
blocks. So most races should be taken care of.
-- mingo
-
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/