That code in bmesh_interp.c you are asserting near, that frees the "oldpool" was recently added (by me) to fix a memory leak. It seemed right, but if you're having crashes from it, maybe not?
In knifetool.c, knife_input_ray_cast... whats the point of the local var "imat"? I think it's dead now... On Tue, Oct 18, 2011 at 1:39 AM, Campbell Barton <ideasma...@gmail.com>wrote: > Revision: 41103 > > http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=41103 > Author: campbellbarton > Date: 2011-10-18 08:39:26 +0000 (Tue, 18 Oct 2011) > Log Message: > ----------- > - Knife tool would switch the knife line to 0,0,0 when not over any > geometry, this felt glitchy and while drawing in free space isn't supported > 0,0,0 can easily be free space too. So draw a line to the mouse cursor. > > - knife tool was missing calls to BLI_array_free() > > - add assert for bug [#28960] where dissolve can cause invalid bmesh. > > _______________________________________________ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
