On Tue, Nov 27, 2012 at 1:35 PM, Ansis Atteka <[email protected]> wrote: > On Tue, Nov 27, 2012 at 11:28 AM, Ben Pfaff <[email protected]> wrote: >> On Tue, Nov 27, 2012 at 10:45:42AM -0800, Ansis Atteka wrote: >>> The flow_format() function was incorrectly passing skb_priority >>> to the match_format() function. match_format() function instead >>> expects rule priority. >>> >>> Signed-off-by: Ansis Atteka <[email protected]> >> >> Good catch, thank you. >> >> Will you please cite the commit that added the bad match_format() call >> in the change log, so that it is easy to see what versions are >> affected? > > Sure, skb_priority has been used like this since the very beginning - > Nov 1 2011 (abff858b5ad310a529d5a5ac2a230ee4ac9736db datapath: Convert > kernel priority actions into match/set.). Though, that particular line > of code has been re-factored multiple times since then.
I think it was only when we started using match_format() for this purpose that it became a problem. Previously, skb_priority was printed out in flow_format() but it was the correct one. However, this is particularly confusing because the same patch also didn't print out the non-OpenFlow fields, including skb_priority, so the effects somewhat canceled each other out. Both of these issues came in on 10/22/2012 in aa6c9932f2937fa9a2140ec1737668eb9105b0b5 (Change logging format for flows to that accepted by ofproto/trace.). _______________________________________________ dev mailing list [email protected] http://openvswitch.org/mailman/listinfo/dev
