On Wed, 2010-06-02 at 12:38 -0400, Cliff Brake wrote: > I ran into some issues with srctree that are fixed by the following. > Not sure if this is the correct, or valid fix, but perhaps someone can > review: > > http://cgit.bec-systems.com/cgit.cgi/bitbake/commit/?h=cbrake/1.10-srctree-fixes&id=c6fe73d986d596dca7f76bd277cf279cd5c21308
This is just a bandaid over a bigger problem. The question is why the task isn't in that list. It should be either returning success or failure once and that code should work. So it sounds like a reference counting type issue and the question is why srctree is breaking the refcounting? I'd really like to get to the underlying problem. Do you have a complete event log for a case when this happens? Are the number of starts/succeeds/fails balanced? I know you uncommented the "print event" debug line and this info should help understand whats going on. Cheers, Richard _______________________________________________ Bitbake-dev mailing list [email protected] https://lists.berlios.de/mailman/listinfo/bitbake-dev
