On Sat, Sep 12, 2020 at 08:45:10AM +0100, Dan Scally wrote: > Hello Heikki > > On 09/09/2020 14:40, Heikki Krogerus wrote: > > I'm almost certain that my graph patch is broken. My tests did not > > cover nearly as much that is needed to properly test the patch. I > > think at least the refcount handling is totally broken in > > software_node_graph_get_next_endpoint(), so that function at least > > needs to be rewritten. > > > > Unfortunately I do not have time to work on that patch right now. > > > > thanks, > > I managed to get to the bottom of the remaining issue (which was the cause > of the problem I originally posted here about - that's now all resolved). > In addition to the refcount handling problems, > software_node_graph_get_next_endpoint() was occasionally passing an invalid > combination of parameters to software_node_get_next_child(); sometimes it > would pass an existing endpoint as old in combination with a port which was > not the endpoint's parent. Applying this on top of your patch seems to stop > both of those errors: Can you send a formal patch where you will be a co-developer to Heikki? Though I don't remember if Heikki signed off it. -- With Best Regards, Andy Shevchenko
