Re: [PATCH RFC for-next] net/mlx4_core: Fix racy flow in the driver CQ completion handler

On Tue, Sep 4, 2012 at 11:12 AM, Max Matveev [off-list ref] wrote:

On Thu, Aug 30, 2012 at 22:35 PM Roland Dreier wrote:

quoted

On Thu, Aug 30, 2012 at 3:17 PM, Or Gerlitz [off-list ref] wrote:

quoted

Roland Dreier [off-list ref]  wrote:

quoted

Can you be explicit about the race you're worried about?

few

1. on the  time CQ A is deleted an interrupt that relates to CQ B
takes place and a radix
tree lookup is running while an element is being deleted from the
tree, looking on the radix tree API, I don't see that this is
allowed.

I don't think this is a real problem; the radix tree code is
explicitly designed for RCU use, and the data structure is pretty
clearly safe for looking up one slot while another slot is being
cleared.  In fact it's hard to see how this could screw up.

What about races between radix_tree_extend and radix_tree_lookup?

As far as I can see (even on newer kernels with rcu support) there is
nothing which protects changes to node->height and it is used to
decide how deep lookup must go to find the actual data.

Max,

I'm putting your response on linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org as it was
wrongly sent to linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org ...

Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html