On Sat, Aug 17, 2013 at 02:14:45PM -0700, Linus Torvalds wrote:
>
> So anybody using "get_cached_acl()" is safe. It gets either a locked
> ref-counted ACL pointer, or an unlocked optimistic NULL pointer.
Ah, I see. But as we start adding the RCU annotations and the calls
to rcu_derference(), then I imagine we'll start triggering warnings
from various other static analysis tools, even if it makes sparse happy.
> So the ACL accesses are this somewhat strange mix of RCU and non-RCU
> use. We probably could make *more* of them use the RCU model, but
> apart from the RCU pathname lookup nothing else has ever been critical
> enough to care.
Ugh. OK, I care mostly because it's adding noise to my make C=1 runs,
but I'm not sure I care enough to fix it up right away. I'll put it
on my todo list, and hope someone beats me to it. :-)
- Ted
--
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
