Miklos Szeredi wrote:
And the things kick_inodes() does without any sort of locking look
even more dangerous.
It should be the other way round: first make sure nothing is
referencing the inode, and _then_ start cleaning it up with
appropriate locks held. See prune_icache().
kick_inodes() only works on inodes that first have undergone
get_inodes() where we establish a refcount under inode_lock(). The final
cleanup in kick_inodes() is done under iprune_mutex. You are looking at
the loop that does writeback and invalidates attached dentries. This can
fail for various reasons.
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
