They were caused by not having a free_windowmanager_cb set and by not having registered SpaceTypes, which meant data allocated for thosse SpaceTypes could not be freed. These were solved by defining a free_windowmanager_cb for the player that just frees wmWindows, and by making sure we only allocate memory for registered SpaceTypes.
with recent outliner optimizations (revision 58855, 3 days ago), in any Blender
version before this commit. On current Blender it would give a double free
warning in the console.
The problem is that it creates a temporary TreeStore on file save. However if you
have multiple outliners this memory block would always be at the same memory
address making it no longer unique. That then meant old memory address lookups on
file read were failing. Solution now is to postpone freeing these temporary memory
blocks until the end so that they are at unique addresses.
- performance of outliner was low because of unoptimal data structures.
- now it uses BLI_mempool instead of custom mempool and GHash to make searches for duplicates faster.
- also fix undesired behaviour of BLI_mempool_as_arrayN
thanks to Campbell Barton and Lukas Tönne for helping me get a better fix put together.
* Change overlay alphas to 33 if not initialized. This should have been
done for 2.67 but better do it now to avoid frustration with overly
transparent overlays. For users that have set this low manually this
will reset the setting to 33 but I think it's less irritating than users
who unwillingly had the setting to 1 and saw nothing when activating the
overlay.
* Allow overlay alpha to be zero as well
* Reset old, now obsolete BRUSH_FIXED flag for brushes that still use
this. I am doing this here to avoid patching hell when the paint brush
is merged, since it's not possible to know the merged version
beforehand, and this flag will be used.
existing code was very stupid.
- all ID pointers for clipboard strips are handled uniformly.
- clipboard stores a duplicate ID pointer which are restored on paste.
- restoring pointers...
-- use ID's that are still in the database (copy&paste within the same file).
-- fallback to name lookup.
-- fallback to loading them from the original filepath (movie-clip and sound only).
also fix bug pasting where initialing the sound wasn't done if there was no frame-offset.
the way Curve.len is used at the moment is really stupid, calculate string size on save for now, but should really store the length in bytes and total number of characters.
custom pynodes tree edited directly as library data would get a 0 user count on reload and subsequently not be saved. Now the node space follows the same pattern as the image space: the node tree(s) user
count gets incremented on file load and opening in the editor ensures a real user. This leads to 1 "unreal" user for the editor (dropped on reload), but seems to be the only viable solution atm.
* [#35724] Backdrop zoom can be set to a very small value, making the backdrop disapear.
There were checks in the drawnode that needed to be placed in the readfile.
The checks checked if the zoomlevel was 0.0, then it was defaulted to 1.0, but the zoomvalue had a minimum limit of 0.01, hence it did not work.
Moved the check to the readfile and checked for all values smaller then 0.02. These values are then reset to 1.0
Jeroen & Monique
- At Mind -
Previously it was nearly impossible to have fast moving objects emitting smoke or they would just leave behind a row of smoke poofs instead of continious stream of smoke. Now it's possible to set number of subframes for each smoke flow.
Another new thing is ability to set size of smoke flow particles instead of using closest smoke cell. This also works with my earlier "full sample" commit, so no more blocky particles either. :)
For more info check my blog post: http://www.miikahweb.com/en/blog/2013/05/17/blender-smoke-subframes
This commit also includes couple of fixes I spotted while testing:
* Fix: dissolve was applied at different time for low res and high res simulations.
* Fix: full sample setting didn't get copied with domain.
This is hopefully the ultimate solution against smoke blockiness near emitter.
Previously high resolution flow/emitter voxels were generated based on the low resolution ones. So if you had 32 resolution and 4 division high resolution, it still used smoke flow generated from those 32 resolution voxels. Now I introduced a new sampling method called "Full Sample" that generates full resolution flow for for high resolution domain as well.
Read more about it in my blog post: https://www.miikahweb.com/en/blog/2013/05/10/getting-rid-of-smoke-blockiness
Also changed "quick smoke" operator default voxel data interpolation mode to "Cubic B-Spline" to smoothen out it even more.
Issue was caused by ntreeUpdateTree calling for a ntree
which is not in G.main.
This lead to issues in ntreeVerifyNodes (which is called
from ntreeUpdateTree).
Made is so ntreeUpdateTree now accepts main as an argument.
Will work for the release, later we could either solve the
TODO mentioned in ntreeUpdateTree which will eliminate need
in main there or make it so context's main is used from all
over where ntreeUpdateTree is called (currently there're
still some usages of G.main).
Also removed unneeded image buffer scaling, it was only needed
for "early output" if there was no rotation. That is no longer
supported since it used to pixelate result a lot and interpolation
is always used now.
Saves quite a few of memory and CPU cycles.
Initial fix was not fully working, because faulty
edge shared the same address as screen area. This
lead to screen freeing issues -- double free (which
was supressed by guarded allocation actually) but
also freed memory access when iterating via edges
to free them.
Solved by a small hack which removes edges with bad
vertices from edges list. This prevents double-free
and freed memory access cased by corrupted files we're
currently fixing. In other cases this tweak is likely
be harmless -- in worst case scenario it'll lead to
small memory leak, which is not as much lethal as
freed memory access.
Added provision for saved corrupt blend files - caused by a startup.blend
addressing > 16 GB space, which was read in 32 bits.
Now an invalid screen will get removed immediate after read. Might give
a memory-not-in-memlist print, but that's quite safe.
Added a mutex lock for smoke data access. The render was already working with a
copy of the volume data, so it's just a short lock to copy things and should not
block the UI much.
Issue was cause dby ima->ibufs being stored in a ma on
undo/redo and ima->rr not. In case of multilayer image
ibufs hares pointer with render result, so current
undo/redo policy lead to a dead pointer stored in ibuf.
Made it so ima->rr also stores in the map and restores
on redo/undo keeping all the pointers fine.
