Blender Git Loki

Fix broken "force setting multisamples only once", introduced in b5b359b48f7f

Fix metaball->mesh removing all metas in the scene

Assumed the entire scene used the one motherball.

Fix access freed memory when doing constraints ID counter on main free

Basically just made constraints free function aware of possible do_id_users
argument, same as we've got for objects, object data and so on.

Fix T43905: Crash happens when press Create Plane Track

It was only happening on 32bit platforms because of alignment
differences when allocating class.

Now got rid of copy of eigen matricies stored by value in the
residual block which solves aligment issues and should also
give some unmeasurable speedup.

PyAPI: clarify fallback docstring

Fix T43904: calc_vert_angle returns wrong value

This was mis-named, rename to `calc_edge_angle`
and allow a fallback value in the case when the vert doesn't have 2-edges.

BMesh: Add extended BM_vert_calc_edge_angle that takes a fallback value

Cleanup: quiet minilzo warnings

Cleanup: unused warnings

Cleanup: unused macro warning

Fix typo in "Add" and "Subtract" mode of Vector Math node.

Differential Revision: https://developer.blender.org/D1003

Correction to previous commit

Calling ensure_lookup_table for each face is stupid! :/
(Noted by Sergey - thx)

Fix T43900: "Operator Mesh Add" template fails
Another script that was missing the lookup_table call.

Revert "COLLADA - support for shadeless material (SHADER_CONSTANT)"

This reverts commit 04b0a9f4b885e8e3b0b3207f3b3cda74b936df3e.

Attempted fix for OpenAL synchronization.

This hit animators already, basically when using sound sync we can
hit negative frames. This happens because we always subtracted the full
range of the triple buffer size from the timing, even when buffers were
flushed.

Also, old code read the offset from the offset of the reader. The
problem here, is that due to threading there is a time offset between
the when this offset is set and when it is offloaded to the buffers,
which means that we could get quite some variance between time
reporting.

Now sounds keep a private byte offset which is incremented right before
invalidating old buffers. This should make the combination of OpenAL
time report + byte offset more accurate. For even more accuracy we might
spinlock while updating those values but for now left it as is for fear
of the lock interfering with frame update performance. We can try to be
smarter here, storing old value while update is happening, and use
trylock and the old values if it fails but for now commit the simple
version.

Merge branch 'master' into temp-ghash-experiments

Making hash storage optional.

This required quite a bit of work, but storing hashes for 'simple' very quick
hashing/comparison functions is stupid (loss of space, and even introduces a
very small slow down), while with e.g. strings (which comparison and hasing
is quite expansive), it gives huge speedup (twice quicker insertion, about
33% quicker on lookup).

This makes lower-level code a bit less clear, since we add one 'child type' of Entry
for each case (GHash/GSet, and storing hash or not), but this also has the benefit
of being stricter when accessing entries' members...

Merge branch 'master' into temp_motionpaths

crash fix: forgot to "walk" over the cutter_group inside the modifier, effectively forgetting to set references to the helper group, thus resulting in a crash at loading time

fix: greasepencil by edges did not take object transformation into account properly