Commit Graph

3179 Commits

Author SHA1 Message Date
Campbell Barton
f01c6e185f Cleanup: typos 2015-05-23 22:38:47 +10:00
Campbell Barton
f340595dba Quadric: add common why doubles are used. 2015-05-23 22:37:46 +10:00
Campbell Barton
bc83076355 Math Lib: update API usage info 2015-05-23 22:37:37 +10:00
Campbell Barton
e37c4e5819 Fix T44780: Decimate planar creates concave edges
Float precision was causing problems for decimate,
small faces that create a nearly flat surface were detected as having no 'cost' to collapse.
2015-05-21 21:06:30 +10:00
Campbell Barton
6ee653352b Math Lib: double versions of vector funcs
- add_vn_vn_d
- add_vn_vnvn_d
- mul_vn_db
2015-05-21 21:06:29 +10:00
Campbell Barton
24e1d7f4f4 BMesh: decimate wasn't using face/edge centers
When calculating quadrics, using the first-vertex isn't correct.
2015-05-21 00:00:13 +10:00
Campbell Barton
5d30c23c35 doxygen: corrections/updates
Also add depsgraph & physics
2015-05-20 14:12:22 +10:00
Sergey Sharybin
40091ff83a Correction to early output in the parallel range implementation
The used heuristic of checking the value prior to lock is not totally safe
because assignment is not atomic and check might not give proper result.
2015-05-18 16:40:51 +05:00
Sergey Sharybin
b88597c218 Make switching to threaded malloc safe to be called from threads
For a long time this function was only intended to be used from the main thread,
but since out implementation of parallel range (which is currently only used by
mesh deform modifier) we might want to switch to threaded alloc from object
update thread.

Now we're using spinlock around the check, which makes the code safe to be used
from all over the place.

We might consider using a bit of atomics operations magic there, but it's not so
much important for now, this code is not used in the performance critical code
path.
2015-05-18 16:40:51 +05:00
Campbell Barton
6ba128129d CMake: correct file listing 2015-05-17 16:09:15 +10:00
Campbell Barton
a050d6063c Project Paint: resolve ugly bleed artifacts
Use the bilinear reverse to find the pixel to bleed from.
Was using pixel space which didn't work well.
2015-05-13 18:04:46 +10:00
Campbell Barton
18cf235893 GHash: use const keys when only used for lookups 2015-05-11 12:39:08 +10:00
Campbell Barton
9e2e85a367 GHash: Add BLI_ghash_ensure_p_ex to copy the key
Needed in cases where the memory from each key is owned by the GHash.
2015-05-11 09:27:05 +10:00
Campbell Barton
c246e0c3b6 Prefer name 'program' over 'binary'
binary-search is confusing!
2015-05-06 06:34:19 +10:00
Campbell Barton
e00142bfa7 BLI_path: add PATH search utility functions 2015-05-06 06:21:16 +10:00
Campbell Barton
e5e73ccc90 Math Lib: rename fill_*, to copy_*
matching convention for fixed length api, eg: copy_v3_fl
2015-05-05 17:08:29 +10:00
Campbell Barton
52b9d83417 Fix T44592: Zero scale on an axis, zeros all 2015-05-04 22:39:33 +10:00
Lukas Tönne
c19f3ea1b2 Fixed comment. 2015-05-04 14:31:44 +02:00
Brecht Van Lommel
b6caefdaa9 Fix T43711: dual quaternion deform bug with shearing in deform matrix.
This also increases the tolerances in is_orthogonal / is_orthonormal functions,
which were much too low for practical purposes.
2015-05-01 19:21:41 +02:00
Campbell Barton
3acc1ba49c Add macro BLI_SMALLSTACK_AS_TABLE
Use for edge-split (a little less overhead compare to popping each item).
2015-04-29 23:59:48 +10:00
Campbell Barton
7851534541 disable ARRAY_SIZE fixed length check for Clang 2015-04-29 01:37:45 +10:00
Campbell Barton
1f9fe0626f BLI_utildefines: add MEMSET_STRUCT_OFS macro 2015-04-28 23:15:47 +10:00
Campbell Barton
a4965249ec Math Lib: add range_vn_u 2015-04-28 06:54:50 +10:00
Campbell Barton
84836e8952 Cleanup: use strict flags 2015-04-28 06:24:06 +10:00
Campbell Barton
278ff15c7f BLI_array: add permute utility function 2015-04-28 04:12:48 +10:00
Campbell Barton
9dadc8f599 Math Lib: add mul_v3_mat3_m4v3 2015-04-27 21:52:42 +10:00
Campbell Barton
3917e9f1f2 Macro GENERIC_TYPE_ANY map many types to one value 2015-04-26 17:19:51 +10:00
Campbell Barton
bdf6393c98 Math Lib: pow_i for int power-of 2015-04-24 11:37:48 +10:00
Campbell Barton
372752c739 Math Lib: add count_bits_i utility function 2015-04-24 01:37:12 +10:00
Campbell Barton
e6ceecdf97 Cleanup: pep8, spelling 2015-04-22 16:26:54 +10:00
Campbell Barton
bf69453ae7 BLI_string: add BLI_snprintf_rlen
use when the length of the destination string is needed.
2015-04-22 05:37:22 +10:00
Campbell Barton
c9f9e29538 Math Lib: handling bits handling into own file 2015-04-22 04:44:24 +10:00
Campbell Barton
c5a126ee70 Docs: BLI_rand non-obvious behavior 2015-04-16 16:14:30 +10:00
Campbell Barton
d36429b930 initialize members in-order 2015-04-14 18:41:01 +10:00
Antony Riakiotakis
efe90384b7 Be a bit more specific in the message. 2015-04-13 12:28:12 +02:00
Antony Riakiotakis
8a97d46296 Windows:
Only print backtrace on debug builds, since on release builds there is
only some useless output from a python library and fftw.
2015-04-13 12:21:32 +02:00
Sergey Sharybin
97ae0f22cd Depsgraph debug: Remove hardcoded array of BLENDER_MAX_THREADS elements
Allocate statistics array dynamically, so increasing max number of threads does
not increase sloppyness of the memory usage.

For the further cleanups: we can try alloca-ing this array, but it's also not
really safe because we can have quite huge number of threads in the future.
Plus statistics will allocate memory for each individual entry, so using alloca
is not going to give anything beneficial here.
2015-04-13 14:41:02 +05:00
Campbell Barton
e688ba0304 GHash: use unsigned int for ghash_size 2015-04-13 13:45:48 +10:00
Campbell Barton
ccf44c400c BMesh: simplify hashing for dyntopo
Was using pointer hashing when the keys are in fact uint's.
Since they're well distributed from the rangetree,
no need to do bit-shifting tricks. just use int as hash.
Gives ~8% speedup in own tests.
2015-04-11 23:36:37 +10:00
Campbell Barton
b0906bcc5e Cleanup: remove win32 workaround 2015-04-10 22:06:28 +10:00
Campbell Barton
12661de087 Remove invalid assert 2015-04-09 13:00:58 +10:00
Bastien Montagne
aa24704749 Fix T44235: UNC Path Fails in open.
Here again, stat on '\\MYSERVER\foo\..' does not work...

Anyway, we can handle this in a much much simpler way using
BLI_access and BLI_parent_dir...
2015-04-08 21:59:05 +02:00
Bastien Montagne
114d1b23d2 Fix T44113: Some System Folders do not contain go back arrow.
On windows empty dirs are completely empty - no par or current entries
are listed either, in those cases artificially add those.

Furthermore, stat on UNC paths do not support current/parent 'shortcuts'
(i.e. things like '\\SERVER\foo\bar\..' do not work), so we have to hack
around that mess...

This should ensure us we always do have valid parrent entry...
2015-04-08 21:59:05 +02:00
Campbell Barton
01b1eb445c Math Lib: add project_plane_v3_v3v3
Useful for projecting one vector onto another (as a plane).

This is a rather common operation,
doing inline isn't always obvious whats happening.
2015-04-08 23:01:37 +10:00
Joshua Leung
441f68e2b0 Partial fix for T44219
Prevent crashes from out of bounds array access if an invalid rotation mode
is passed to the euler rotation blenlib calculations.
2015-04-08 14:34:07 +12:00
Campbell Barton
9510137d12 Cleanup: brace placement /w structs 2015-04-07 11:25:42 +10:00
Campbell Barton
5217d2bc0e Use BKE_edgehash_ensure_p where possible 2015-04-07 10:53:58 +10:00
Campbell Barton
808de65d91 EdgeHash: ensure function, avoids multiple lookups 2015-04-07 10:53:20 +10:00
Campbell Barton
1b9f1519bc Cleanup 2015-04-07 08:46:48 +10:00
Dalai Felinto
d5f1b9c222 Multi-View and Stereo 3D
Official Documentation:
http://www.blender.org/manual/render/workflows/multiview.html

Implemented Features
====================
Builtin Stereo Camera
* Convergence Mode
* Interocular Distance
* Convergence Distance
* Pivot Mode

Viewport
* Cameras
* Plane
* Volume

Compositor
* View Switch Node
* Image Node Multi-View OpenEXR support

Sequencer
* Image/Movie Strips 'Use Multiview'

UV/Image Editor
* Option to see Multi-View images in Stereo-3D or its individual images
* Save/Open Multi-View (OpenEXR, Stereo3D, individual views) images

I/O
* Save/Open Multi-View (OpenEXR, Stereo3D, individual views) images

Scene Render Views
* Ability to have an arbitrary number of views in the scene

Missing Bits
============
First rule of Multi-View bug report: If something is not working as it should *when Views is off* this is a severe bug, do mention this in the report.

Second rule is, if something works *when Views is off* but doesn't (or crashes) when *Views is on*, this is a important bug. Do mention this in the report.

Everything else is likely small todos, and may wait until we are sure none of the above is happening.

Apart from that there are those known issues:
* Compositor Image Node poorly working for Multi-View OpenEXR
(this was working prefectly before the 'Use Multi-View' functionality)
* Selecting camera from Multi-View when looking from camera is problematic
* Animation Playback (ctrl+F11) doesn't support stereo formats
* Wrong filepath when trying to play back animated scene
* Viewport Rendering doesn't support Multi-View
* Overscan Rendering
* Fullscreen display modes need to warn the user
* Object copy should be aware of views suffix

Acknowledgments
===============
* Francesco Siddi for the help with the original feature specs and design
* Brecht Van Lommel for the original review of the code and design early on
* Blender Foundation for the Development Fund to support the project wrap up

Final patch reviewers:
* Antony Riakiotakis (psy-fi)
* Campbell Barton (ideasman42)
* Julian Eisel (Severin)
* Sergey Sharybin (nazgul)
* Thomas Dinged (dingto)

Code contributors of the original branch in github:
* Alexey Akishin
* Gabriel Caraballo
2015-04-06 10:40:12 -03:00