On Tue, Nov 26, 2013 at 11:24:47AM -0800, Jarno Rajahalme wrote: > Ofproto always provides a valid pointer to a mask, but the mask length > is zero when megaflows are disabled. Check the mask length before > checking the mask, as the check can fail otherwise. > > Signed-off-by: Jarno Rajahalme <jrajaha...@nicira.com>
I agree that the code is wrong, but I think that this does not go far enough as a fix. If dpif_netdev_flow_mask_from_nlattrs()'s caller provides a nonnull 'mask' then it expects 'mask' to be initialized whether or not it provides a nonempty 'mask_key', but if I am reading it correctly this (and the existing code) leave it indeterminate. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
