See below for my in-line responses... On Tue, Mar 12, 2013 at 4:50 PM, J. Liles <[email protected]> wrote:
> > > On Tue, Mar 12, 2013 at 1:39 PM, Douglas Regehr > <[email protected]>wrote: > >> Hi, >> >> This is similar to the problem I was having a few weeks ago. My session >> fails to load. The progress bar gets stuck at 78% and then non-daw >> eventually crashes. I've attached my history file, a gdb trace, and the >> console output from non-daw. >> > > You're getting pretty good at this. > PalmOS used to have an app called "Gremlins" that injected an enormous number of UI events in short succession to test the stability of an app. Just think of me as Gremlins for Non. > > >> The problem happened while I was "comping" together a number of takes. I >> do this by creating a new take, and ctrl-dragging regions from other takes >> into the new take, resizing them and crossfading them to suit. I think I >> was resizing a region when the session suddenly stopped responding to >> input, there was a big spike in cpu usage during which I struggle and >> eventually succeeded in clicking "close" in NSM, and then eventually >> non-daw crashed. Since then I have not been able to re-load the session. >> > > I take it that you were doing this with the transport stopped? > > Yes the transport was stopped > >> I tried building and running the latest version (after modifying the >> wscript file to let me build with cairo 1.10...is that bad?) to see if that >> made a difference but it didn't. >> > > I'm just surprised it worked. I thought cairo_region_t was introduced in > 1.12. I'll lower the requirement. > > >> >> I was able to work-around the problem by renaming the snapshot file. So >> I guess the snapshot was corrupt but the history seems ok... >> >> Doug >> > > > The backtrace... Is this from attempting to load the crashed session > later? Was it hung at the time or is this from a core dump? > This was from attempting to load the crashed session later. I attached gdb before attempting to load the session. Then, when the progress bar in non-daw reached 78% gdb caught a SIGABRT and I did the backtrace.
