blender.git
3 years agoFix warnings reported by MSVC
Sergey Sharybin [Tue, 23 Feb 2016 08:44:54 +0000 (09:44 +0100)]
Fix warnings reported by MSVC

Mainly it's related on a bad practice in SDL to force-define __SSE__
and __SSE2__ flags which generates quite some warnings and causes too
much noise.

There are some other warnings fixed. Should be no functional changes.

NeXyon, please check the changes in audaspace :)

3 years agoFix white balance sequencer modifier
Campbell Barton [Tue, 23 Feb 2016 03:50:30 +0000 (14:50 +1100)]
Fix white balance sequencer modifier

- division often visibly clipped channels.
- division by zero caused NAN pixel values.

3 years agoFix T47531: crash with particle roughness curve mapping, after fix for T46382.
Brecht Van Lommel [Tue, 23 Feb 2016 03:22:13 +0000 (04:22 +0100)]
Fix T47531: crash with particle roughness curve mapping, after fix for T46382.

3 years agoFix button display clamping values
Campbell Barton [Mon, 22 Feb 2016 22:55:16 +0000 (09:55 +1100)]
Fix button display clamping values

Displaying a button would clamp the value if the button was outside the range.
This could be OK in some cases,
however it's problematic with object dimensions which would re-scale objects on showing the panel.

Add `ui_but_update_edited` when its OK to modify the value.

3 years agoFix T47529: Selecting paths makes them relative
Campbell Barton [Mon, 22 Feb 2016 19:37:01 +0000 (06:37 +1100)]
Fix T47529: Selecting paths makes them relative

When selecting file-paths from the interface, initialize the 'Relative' setting from existing paths.

3 years agoSculpt: use Ctrl to set grab/sculpt normal-weight
Campbell Barton [Mon, 22 Feb 2016 19:16:53 +0000 (06:16 +1100)]
Sculpt: use Ctrl to set grab/sculpt normal-weight

Having to change this with the slider each time isn't so convenient
and invert in this case does nothing.

3 years agoSculpt: apply grab normal-weight to snake-hook
Campbell Barton [Mon, 22 Feb 2016 19:04:23 +0000 (06:04 +1100)]
Sculpt: apply grab normal-weight to snake-hook

3 years agoSculpt: Enable original-normal for grab brush
Campbell Barton [Mon, 22 Feb 2016 18:23:47 +0000 (05:23 +1100)]
Sculpt: Enable original-normal for grab brush

Without this, grabbing with normal weight will continually select new normals
based on where you move the cursor,
causing the normal location to flicker in a way which isn't controllable in any useful way.

3 years agoSculpt brush can now grab inwards w/ normal weight
Campbell Barton [Mon, 22 Feb 2016 18:04:32 +0000 (05:04 +1100)]
Sculpt brush can now grab inwards w/ normal weight

Previously the grab brush could only move ourwards when normal aligned.

3 years agoDon't draw object center when outside the view
Campbell Barton [Mon, 22 Feb 2016 16:24:30 +0000 (03:24 +1100)]
Don't draw object center when outside the view

3 years agoFix line-width (drawing hair at wrong width)
Campbell Barton [Mon, 22 Feb 2016 15:53:22 +0000 (02:53 +1100)]
Fix line-width (drawing hair at wrong width)

3 years agoFix T47517: "Show Weights" edit-mode fails
Campbell Barton [Mon, 22 Feb 2016 14:22:32 +0000 (01:22 +1100)]
Fix T47517: "Show Weights" edit-mode fails

Don't use materials when showing weight colors.

3 years agoFix memory leak in DM_to_mesh
Campbell Barton [Mon, 22 Feb 2016 12:20:18 +0000 (23:20 +1100)]
Fix memory leak in DM_to_mesh

3 years agoFix T47482: take II, restore 2.76 behavior regarding groups when deleting an object.
Bastien Montagne [Mon, 22 Feb 2016 11:09:36 +0000 (12:09 +0100)]
Fix T47482: take II, restore 2.76 behavior regarding groups when deleting an object.

Also re-reported through IRC by Thomas Beck (@plasmasolutions), thanks.

Though it's not ideal in theory, we have quite poor handling of object datablock currently
from user PoV - before this commit, it was not easily possible to get fully rid of an object
anymore if you did not removed it from all its groups before deleting it.

So for now, restore 2.76 behavior (namely, unlink an object from avaerything in Blender
once it is no more used by any scene).

Better handling of all this is TODO for later (also related to much more heavy changes
done in id-remap branch regarding sanitizing our ID deletion process).

3 years agoCompositor: fix memory leak in filter node operation.
Kévin Dietrich [Mon, 22 Feb 2016 09:58:49 +0000 (10:58 +0100)]
Compositor: fix memory leak in filter node operation.

Avoid allocating the (tiny) array on the heap in the first place.

Reviewers: sergey, lukastoenne

Differential Revision: https://developer.blender.org/D1815

3 years agoFix T43388, fix T40369: Cycles baking not antialiasing textures.
Lukas Stockner [Sun, 21 Feb 2016 13:14:52 +0000 (14:14 +0100)]
Fix T43388, fix T40369: Cycles baking not antialiasing textures.

This re-enables the AA jittering, but with proper clamping so that u >= 0,
v >= 0 and u+v <= 1.

Differential Revision: https://developer.blender.org/D1254

3 years agoBake API: add assert to verify we have valid UVs.
Brecht Van Lommel [Sun, 21 Feb 2016 13:30:37 +0000 (14:30 +0100)]
Bake API: add assert to verify we have valid UVs.

3 years agoFix crash baking selected to active after recent fix for uninitialized variables.
Brecht Van Lommel [Sun, 21 Feb 2016 13:37:47 +0000 (14:37 +0100)]
Fix crash baking selected to active after recent fix for uninitialized variables.

3 years agoAttempt to fix GLog compilaton on FreeBSD
Sergey Sharybin [Sun, 21 Feb 2016 12:40:22 +0000 (13:40 +0100)]
Attempt to fix GLog compilaton on FreeBSD

Copied some config variables from an original config file.

3 years agoFix compositor using edge repeat policy when it shouldn't
Sergey Sharybin [Sun, 21 Feb 2016 12:16:00 +0000 (13:16 +0100)]
Fix compositor using edge repeat policy when it shouldn't

Was easy to notice when alpha-overing smaller image with blur on
the bigger frame.

3 years agoFix uninitialized var use initializing bake data
Campbell Barton [Sun, 21 Feb 2016 05:19:10 +0000 (16:19 +1100)]
Fix uninitialized var use initializing bake data

3 years agoCorrect incomplete doxgen file paths in WM
Julian Eisel [Sun, 21 Feb 2016 00:24:17 +0000 (01:24 +0100)]
Correct incomplete doxgen file paths in WM

3 years agofix T47484: Added fallback solution (copy exported file when rename is not supported)
Gaia Clary [Sat, 20 Feb 2016 22:26:33 +0000 (23:26 +0100)]
fix T47484: Added fallback solution (copy exported file when rename is not supported)

3 years agofix T47484: Added proper error reporting for Collada Exporter
Gaia Clary [Sat, 20 Feb 2016 20:42:12 +0000 (21:42 +0100)]
fix T47484: Added proper error reporting for Collada Exporter

3 years agofix T47484: Added final status of exported file to Blender console
Gaia Clary [Sat, 20 Feb 2016 19:33:29 +0000 (20:33 +0100)]
fix T47484: Added final status of exported file to Blender console

3 years agoFix unnecessary separators in image editor Image menu.
Brecht Van Lommel [Sat, 20 Feb 2016 19:18:58 +0000 (20:18 +0100)]
Fix unnecessary separators in image editor Image menu.

3 years agoFix crash enabling Use Multiview on images with an empty filepath.
Brecht Van Lommel [Sat, 20 Feb 2016 16:43:31 +0000 (17:43 +0100)]
Fix crash enabling Use Multiview on images with an empty filepath.

3 years agoFix memory leak when saving OpenEXR half file fails.
Brecht Van Lommel [Sat, 20 Feb 2016 14:48:15 +0000 (15:48 +0100)]
Fix memory leak when saving OpenEXR half file fails.

3 years agoFix T47482: Own mistake in new handling of 'userone' ID usages when decrementing...
Bastien Montagne [Sat, 20 Feb 2016 15:47:43 +0000 (16:47 +0100)]
Fix T47482: Own mistake in new handling of 'userone' ID usages when decrementing usercount.

Ended up not handling at all 'userone' case, now it increments usercount correctly when needed.

3 years agoMake Blender ready for C++11
Sergey Sharybin [Sat, 20 Feb 2016 15:43:46 +0000 (16:43 +0100)]
Make Blender ready for C++11

Did a full compile of debug build with C++11 enabled, it all passed compilation
apart from some deprecated type used in GE's Video Texture. Solved it inside of
ifdef block now.

In the future we should uncomment the MSVC part of it, it should all be safe and
correct (MSVC2013 does not define new C++ version but supports C++11). The reason
it is commented is to have absolutely no effect on the upcoming release.

3 years agoAdd GHash/GSet pop() feature.
Bastien Montagne [Sat, 20 Feb 2016 14:17:40 +0000 (15:17 +0100)]
Add GHash/GSet pop() feature.

Behavior is similar to python's set.pop(), it removes and returns a 'random' entry from the hash.

Notes:
* Popping will return items in same order as ghash/gset iterators (i.e. increasing
  order in internal buckets-based storage), unless ghash/gset is modified in between.
* We are keeping a track of the latest bucket we popped out (through a 'state' parameter),
  this allows for similar performances to iterators when iteratively popping a whole hash
  (without it, we are roughly O(n!), with it we are roughly O(n)...).

Reviewers: campbellbarton

Differential Revision: https://developer.blender.org/D1808

3 years agoUserpreferences Panel (for the 3DView)
Campbell Barton [Sat, 20 Feb 2016 12:46:53 +0000 (13:46 +0100)]
Userpreferences Panel (for the 3DView)

This is an attempt to improve the User preferences panel for the 3DView. I made 2 changes:

- I reordered the sequence of properties by grouping them into more logical groups as it made sense to me. Please indicate where to rearrange the order if necessary.

- Then i added some changes in the code to get the groups better arranged visually. I am pretty sure that this can be done much better, more clever, more generic, whatever. This is just what i could figure out on my own so far.

Reviewers: aligorith, sergey, gaiaclary

Subscribers: sergey

Projects: #user_interface

Maniphest Tasks: T47295

Differential Revision: https://developer.blender.org/D1757

3 years agoOCIO: Disable the warnings related on C4251
Sergey Sharybin [Sat, 20 Feb 2016 08:45:01 +0000 (13:45 +0500)]
OCIO: Disable the warnings related on C4251

Those warnings are trigerred by stl classes in OCIO's public interface.

To quote MSDN: "C4251 can be ignored if you are deriving from a type in
the Standard C++ Library"

This is the only instance where those warnings hunts us, so for now we
can keep it all local in intern/opencolorio but this might be changed
in the future.

3 years agoGHost: Avoid macro re-definition by undefining the macro first
Sergey Sharybin [Sat, 20 Feb 2016 08:18:01 +0000 (13:18 +0500)]
GHost: Avoid macro re-definition by undefining the macro first

Should b totally harmless since the define was overriten anyway.

3 years agoCarve: Silence 32/64 bit shit warning
Sergey Sharybin [Sat, 20 Feb 2016 08:15:51 +0000 (13:15 +0500)]
Carve: Silence 32/64 bit shit warning

There's no need to mix ints and size_t here at all
because all the values fits into integer.

It's unlikely we'll be re-bundling Carve, so didn't
bother with the patchset.

3 years agoCMake: Ignore paranoid MSVC linker warnings
Sergey Sharybin [Sat, 20 Feb 2016 07:45:29 +0000 (12:45 +0500)]
CMake: Ignore paranoid MSVC linker warnings

3 years agoFix for NULL used instead of false
Sergey Sharybin [Sat, 20 Feb 2016 06:50:58 +0000 (11:50 +0500)]
Fix for NULL used instead of false

3 years agoCleanup, avoid passing const pointer to MEM_freeN()
Sergey Sharybin [Sat, 20 Feb 2016 06:47:51 +0000 (11:47 +0500)]
Cleanup, avoid passing const pointer to MEM_freeN()

This generates warnings with MSVC. Similar typecast was
already done in other cases, so think it's all fine.

3 years agoImBuf: Cleanup, don't use `using namespace`, it's a bad practice
Sergey Sharybin [Sat, 20 Feb 2016 06:42:16 +0000 (11:42 +0500)]
ImBuf: Cleanup, don't use `using namespace`, it's a bad practice

3 years agoImBuf: Solve re-definition warnings
Sergey Sharybin [Sat, 20 Feb 2016 06:40:25 +0000 (11:40 +0500)]
ImBuf: Solve re-definition warnings

The idea now is to have FFmpeg/OIIO headers listed after
the system ones. This is because FFmpeg/OIIO might define
some constants with the same name as the ones from math.h.

FFmpeg/OIIO has ifdef around defines, but math.h doesn't
check whether constants were already defined or not, which
causes some noisy warnings.

3 years agoCycles: Avoid re-definition of math cnstants with MSVC
Sergey Sharybin [Sat, 20 Feb 2016 06:37:24 +0000 (11:37 +0500)]
Cycles: Avoid re-definition of math cnstants with MSVC

3 years agoFix some warnings in GLog/GFlags when buildingwith MSVC2013
Sergey Sharybin [Sat, 20 Feb 2016 06:31:58 +0000 (11:31 +0500)]
Fix some warnings in GLog/GFlags when buildingwith MSVC2013

Those warnings were only addressed for MSVC2015 so far.

3 years agoCMake: Don't pass SSE2 flags in compositor for 64bit MSVC
Sergey Sharybin [Sat, 20 Feb 2016 05:51:27 +0000 (10:51 +0500)]
CMake: Don't pass SSE2 flags in compositor for 64bit MSVC

ALl 64 bit platforms supports SSE2, hence the flag is ignored
and warning was generated.

3 years agoFix (unreported) crash when ungrouping a nodegroup with some animated node.
Bastien Montagne [Sat, 20 Feb 2016 12:15:40 +0000 (13:15 +0100)]
Fix (unreported) crash when ungrouping a nodegroup with some animated node.

Was accessing already freed action in temp nodegroup animation data...

3 years agoFix T44453: Exporting Keying Sets referencing node tree properties generates invalid...
Joshua Leung [Sat, 20 Feb 2016 04:53:32 +0000 (17:53 +1300)]
Fix T44453: Exporting Keying Sets referencing node tree properties generates invalid Python code/paths

The problem is that node trees (such as the Material, Lamp, and Compositor node trees)
are stored as "nested node trees" on the affected datablocks. They are particularly
troublesome to deal with, as the are not easily identified, and also cannot be easily
mapped back to the ID's which actually own them. As a result, the usual automated
methods do not work when dealing with these!

3 years agoFix T47433: Driven shapekey values (from 2.4x) are not indicated as having drivers
Joshua Leung [Sat, 20 Feb 2016 04:25:04 +0000 (17:25 +1300)]
Fix T47433: Driven shapekey values (from 2.4x) are not indicated as having drivers

Only Driver FCurves with named shapekeys (instead of shapekey indices) was
getting picked up by the UI code for testing whether a property had drivers
or not. So, while this version patching code worked when it was initially
written for the 2.4x -> 2.5 transition, some subsequent changes ended up
breaking this. As this stuff is not used often, the breakage wasn't noticed.

3 years agoFix T47492: Name is not displayed for newly renamed FCurve, when the path was broken
Joshua Leung [Sat, 20 Feb 2016 04:01:27 +0000 (17:01 +1300)]
Fix T47492: Name is not displayed for newly renamed FCurve, when the path was broken

Thanks Alexander Romanov (a.romanov) for the fix!

4 years agoOnly use outlines around faces for surface objects
Campbell Barton [Sat, 20 Feb 2016 02:44:17 +0000 (13:44 +1100)]
Only use outlines around faces for surface objects

Was drawing wire when mixed with non-wire nurbs.

4 years agoFix nurbs surface drawing using material for lines
Campbell Barton [Sat, 20 Feb 2016 02:39:33 +0000 (13:39 +1100)]
Fix nurbs surface drawing using material for lines

Also reduce context switching

4 years agoMinor adjustments to keymap editor layouts
Julian Eisel [Fri, 19 Feb 2016 23:42:07 +0000 (00:42 +0100)]
Minor adjustments to keymap editor layouts

* Remove unnecessary big margins after triangle icons (esp. for keymap items they where waaay to big)
* Remove ugly margin after "Foo (Global)" entries
* Add little margin after expanded key maps

4 years agoFix wrong indentation of items in keymap editor
Julian Eisel [Fri, 19 Feb 2016 23:10:28 +0000 (00:10 +0100)]
Fix wrong indentation of items in keymap editor

Actually, 3 issues here:
* Keymap items of "Foo (Global)" maps weren't indented enough
* Items after a opened "Foo (Global)" map were indented too much (col was overridden then).

4 years agoFix compilation error with FFmpeg and Ubuntu Trusty
Sergey Sharybin [Fri, 19 Feb 2016 21:04:07 +0000 (22:04 +0100)]
Fix compilation error with FFmpeg and Ubuntu Trusty

Seems this is a confusion between Libav which was silently used
by Ubuntu instead of FFmpeg.

4 years agoSet line-width, point size when running callbacks
Campbell Barton [Fri, 19 Feb 2016 19:55:16 +0000 (06:55 +1100)]
Set line-width, point size when running callbacks

Since 2.76 Blender no longer resets these after use,
this may break add-ons, so reset before calling region callbacks.

4 years agoCleanup: remove unused environment files.
Campbell Barton [Fri, 19 Feb 2016 16:47:38 +0000 (03:47 +1100)]
Cleanup: remove unused environment files.

4 years agoFix implicit double to float conversion caused by use of double precision M_PI
Sergey Sharybin [Fri, 19 Feb 2016 16:40:45 +0000 (17:40 +0100)]
Fix implicit double to float conversion caused by use of double precision M_PI

4 years agofix T47484: replaced tempnam() by simple name 'untitled'
Gaia Clary [Fri, 19 Feb 2016 15:23:11 +0000 (16:23 +0100)]
fix T47484: replaced tempnam() by simple name 'untitled'

4 years agoFix T47472: Border select in the Action Editor does not respect NLA scaling on Summar...
Joshua Leung [Fri, 19 Feb 2016 13:53:08 +0000 (02:53 +1300)]
Fix T47472: Border select in the Action Editor does not respect NLA scaling on Summary Tracks

4 years agoMac OS X Openvdb, add blosc support.
Martijn Berger [Fri, 19 Feb 2016 10:24:07 +0000 (02:24 -0800)]
Mac OS X Openvdb, add blosc support.
even if WITH_OPENVDB_BLOSC is off openvdb still contains 1 reference to snappy so we link with it always

4 years agoCMake: New dependency graph requires either Boost or C++11
Sergey Sharybin [Fri, 19 Feb 2016 09:18:38 +0000 (10:18 +0100)]
CMake: New dependency graph requires either Boost or C++11

4 years agoFix T47454: Line width display error
Campbell Barton [Fri, 19 Feb 2016 01:45:01 +0000 (12:45 +1100)]
Fix T47454: Line width display error

Note, there are still cases where nurbs surfaces display double line width,
but this isn't a regression from previous releases.

4 years agoWorkaround for T47213: branched path sampling issues with CUDA 7.5.
Brecht Van Lommel [Thu, 18 Feb 2016 23:39:00 +0000 (00:39 +0100)]
Workaround for T47213: branched path sampling issues with CUDA 7.5.

4 years agoFix T47477: Transform allows 'inf' input
Campbell Barton [Thu, 18 Feb 2016 23:19:19 +0000 (10:19 +1100)]
Fix T47477: Transform allows 'inf' input

Numeric input wasn't checking numbers were finite,
could crash transforming with skin modifier.

4 years agoFix crash drawing fonts at size 1
Campbell Barton [Thu, 18 Feb 2016 22:40:26 +0000 (09:40 +1100)]
Fix crash drawing fonts at size 1

4 years agoFix frame node text ignoring blank lines
Campbell Barton [Thu, 18 Feb 2016 22:22:26 +0000 (09:22 +1100)]
Fix frame node text ignoring blank lines

4 years agoFix T47451: Blender internal not rendering meshes with zero vertices in the original...
Brecht Van Lommel [Thu, 18 Feb 2016 04:29:05 +0000 (05:29 +0100)]
Fix T47451: Blender internal not rendering meshes with zero vertices in the original mesh.

4 years agoFix T45343: incorrect Cycles baking of bump maps from selected to active.
Brecht Van Lommel [Thu, 18 Feb 2016 00:35:52 +0000 (01:35 +0100)]
Fix T45343: incorrect Cycles baking of bump maps from selected to active.

The differentials were incorrect, now they are properly transferred from the
low to the high poly mesh.

4 years agoFix related to T45343: UV unwrap giving poor results for some n-gons.
Brecht Van Lommel [Thu, 18 Feb 2016 03:16:19 +0000 (04:16 +0100)]
Fix related to T45343: UV unwrap giving poor results for some n-gons.

4 years agoCycles: Fix wrong sample all direct/indirect settings on sm_50 and higher
Sergey Sharybin [Thu, 18 Feb 2016 15:25:51 +0000 (16:25 +0100)]
Cycles: Fix wrong sample all direct/indirect settings on sm_50 and higher

Seems to be some compiler fault which leads to a wrong flag being used,
making it so wrong number of samples is used for the background.

This should in theory fix issue reported in T47213.

4 years agoBGE: Allow access to original texture openGL Bind code/Id/Number
Ulysse Martin [Thu, 18 Feb 2016 10:56:25 +0000 (11:56 +0100)]
BGE: Allow access to original texture openGL Bind code/Id/Number

This patch adds a python method to get openGL bind code of material's texture according to the texture slot.

Example:
import bge

cont = bge.logic.getCurrentController()
own = cont.owner

bindId = own.meshes[0].materials[0].getTextureBindcode(0)
Test file: http://www.pasteall.org/blend/40679

This can be used to play with texture in openGL, for example, remove mipmap on the texture or play with all wrapping or filtering options.
And this can be used to learn openGL with Blender.

Reviewers: TwisterGE, kupoman, moguri, panzergame

Reviewed By: TwisterGE, kupoman, moguri, panzergame

Projects: #game_engine

Differential Revision: https://developer.blender.org/D1804

4 years agoActually only remove sqlite dll
Martijn Berger [Thu, 18 Feb 2016 09:53:45 +0000 (10:53 +0100)]
Actually only remove sqlite dll

4 years agoRevert "Fix: T46526, Do not try to install sqlite3.dll anymore as we link it statical...
Martijn Berger [Thu, 18 Feb 2016 09:52:36 +0000 (10:52 +0100)]
Revert "Fix: T46526, Do not try to install sqlite3.dll anymore as we link it statically into pythons _sqlite"

This reverts commit a8e6b633c9f54f0948a0b9a47534bc38ac95889f.

4 years agoCleanup: add missing struct qualifier
Campbell Barton [Thu, 18 Feb 2016 08:32:33 +0000 (19:32 +1100)]
Cleanup: add missing struct qualifier

4 years agoFix: T46526, Do not try to install sqlite3.dll anymore as we link it statically into...
Martijn Berger [Thu, 18 Feb 2016 07:19:14 +0000 (08:19 +0100)]
Fix: T46526, Do not try to install sqlite3.dll anymore as we link it statically into pythons _sqlite

4 years agoFix T47461: Different results on CPU and GPU when using Branched Path Tracing
Lukas Stockner [Thu, 18 Feb 2016 00:13:07 +0000 (01:13 +0100)]
Fix T47461: Different results on CPU and GPU when using Branched Path Tracing

The issue here was actually somewhere else - the attached scene from the report used a light falloff node in a sunlamp (aka distant light).
However, since distant lamps set the ray length to FLT_MAX and the light falloff node squares this value, it overflows and produces a NaN
weight, which propagates and leads to a NaN intensity, which is then clamped to zero and produces the black pixels.

To fix that issue, the smoothing part of the light falloff is just ignored if the smoothing term isn't finite (which makes sense since
the term should converge to 1 as the distance increases).
The reason for the different results on CPUs and GPUs is not perfectly clear, but probably can be explained with different handling of
Inf/NaN edge cases.

Also, to notice issues like these faster in the future, kernel_asserts were added that evaluate as false as soon as a non-finite intensity is produced.

4 years agoFix crash on exit with Blender Internal baking and envmap render.
Brecht Van Lommel [Wed, 17 Feb 2016 21:18:45 +0000 (22:18 +0100)]
Fix crash on exit with Blender Internal baking and envmap render.

4 years agoFix armature relationship lines missing w/ hi-dpi
Campbell Barton [Wed, 17 Feb 2016 19:37:12 +0000 (06:37 +1100)]
Fix armature relationship lines missing w/ hi-dpi

4 years agoFix line width for armature w/ custom bones
Campbell Barton [Wed, 17 Feb 2016 18:46:14 +0000 (05:46 +1100)]
Fix line width for armature w/ custom bones

4 years agoFix T47452: Translate-node seams w/ subpixel offset
Campbell Barton [Wed, 17 Feb 2016 14:28:41 +0000 (01:28 +1100)]
Fix T47452: Translate-node seams w/ subpixel offset

4 years agoBuildbot: Remove CMake cache on all platforms
Sergey Sharybin [Wed, 17 Feb 2016 14:29:13 +0000 (15:29 +0100)]
Buildbot: Remove CMake cache on all platforms

This way we are always sure that we are using latest ever configuration
without need to manually poke buildbot after doing changes in the building
environment.

Also uncomment code back, was a left-over from some debug.

4 years agoCycles: Make CUDA 7.5 officially recommended
Sergey Sharybin [Wed, 17 Feb 2016 14:10:02 +0000 (15:10 +0100)]
Cycles: Make CUDA 7.5 officially recommended

This was a hard decision, because going newer CUDA toolkit makes
rendering up to 5% slower. But on another hand, it solves major
speed regressions (up to 30%) with branched path tracing on a
top level cards.

Neither of those regressions have a meaningful and sane workaround
from the code itself.

Toolkit 6.5 could still be used, but it's no longer recommended one.

4 years agoFix/workaround static object initialization in gflags
Sergey Sharybin [Wed, 17 Feb 2016 14:00:12 +0000 (15:00 +0100)]
Fix/workaround static object initialization in gflags

4 years agoinstall_deps: final CMake instructions: always update all libs controlled by this...
Bastien Montagne [Wed, 17 Feb 2016 13:47:49 +0000 (14:47 +0100)]
install_deps: final CMake instructions: always update all libs controlled by this script.

Otherwise, switch from local built to system installed libs (once a valid package becomes available)
would be missed...

4 years agoinstall_deps: OSD building: disable CUDA/OpenCL backends.
Bastien Montagne [Wed, 17 Feb 2016 13:14:41 +0000 (14:14 +0100)]
install_deps: OSD building: disable CUDA/OpenCL backends.

Those are can of worms to support properly (versions mismatch, etc.),
and not used in Blender currently anyway.

4 years agoCycles: Don't silence unused macro, remove the macro instead
Sergey Sharybin [Wed, 17 Feb 2016 11:40:56 +0000 (12:40 +0100)]
Cycles: Don't silence unused macro, remove the macro instead

It's not really handy to silence something unused hoping for it'll be
used in the future. We can end up with quite some silencing then.

Also made this flag which i find rather useless to NOT cause -Werror
in Cycles code.

4 years agoCleanup: unused define warning
Campbell Barton [Wed, 17 Feb 2016 10:37:44 +0000 (21:37 +1100)]
Cleanup: unused define warning

4 years agoCleanup: OSL style
Campbell Barton [Wed, 17 Feb 2016 10:36:54 +0000 (21:36 +1100)]
Cleanup: OSL style

4 years agoFix T47447: Cycles crash when reaching texture limit with 3D textures
Sergey Sharybin [Wed, 17 Feb 2016 10:43:03 +0000 (11:43 +0100)]
Fix T47447: Cycles crash when reaching texture limit with 3D textures

4 years agoSome more tweaks to make linking to a shared FFmpeg-3.0 happy
Sergey Sharybin [Wed, 17 Feb 2016 08:25:26 +0000 (09:25 +0100)]
Some more tweaks to make linking to a shared FFmpeg-3.0 happy

4 years agoAdd missing glLineWidth call
Campbell Barton [Wed, 17 Feb 2016 09:35:47 +0000 (20:35 +1100)]
Add missing glLineWidth call

4 years agoUse high precision for scene scale
Campbell Barton [Wed, 17 Feb 2016 09:31:11 +0000 (20:31 +1100)]
Use high precision for scene scale

Without this its possible to accidentally truncate the value.

4 years agoFix T47411: Cycles portals luminance artifacts
Lukas Stockner [Tue, 16 Feb 2016 23:29:21 +0000 (00:29 +0100)]
Fix T47411: Cycles portals luminance artifacts

When using multiple portals, scene areas behind one of the portals were rendered darker than they should.
The reason for that is a pretty stupid mistake: Since portals are only used at positions that aren't behind them,
only portals that are used should be accounted for in the PDF calculation. That was actually the case, but the final
divide incorrectly divided by the total amount of portals, not the amount of visible ones.

Another issue with areas behind portals was the PDF evaluation function.
The new evaluation code is shorter, simpler and fixes this issue.

Also, the threshold for the distance check was increased to avoid artifacts where portals touch a surface.

4 years agoFix OpenCL kernel build errors after recent 3D texture changes.
Brecht Van Lommel [Wed, 17 Feb 2016 00:20:46 +0000 (01:20 +0100)]
Fix OpenCL kernel build errors after recent 3D texture changes.

4 years agoFix (harmless) assert in BVH spatial splits with Visual Studio debug builds.
Brecht Van Lommel [Tue, 16 Feb 2016 21:35:18 +0000 (22:35 +0100)]
Fix (harmless) assert in BVH spatial splits with Visual Studio debug builds.

4 years agoFix broken ghash performance gtest.
Bastien Montagne [Tue, 16 Feb 2016 21:18:49 +0000 (22:18 +0100)]
Fix broken ghash performance gtest.

Regression from rB2dba2b3d71d9781bce45. Do not understand why MSVC needs this convoluted
allocation (looks like broken compiler crap?), but at least let's do it correctly!

4 years agoCleanup: avoid -Wnarrowing warning
Campbell Barton [Tue, 16 Feb 2016 18:17:03 +0000 (05:17 +1100)]
Cleanup: avoid -Wnarrowing warning

4 years agoCycles: Remove back some tweaks for inline policies
Sergey Sharybin [Tue, 16 Feb 2016 16:25:58 +0000 (17:25 +0100)]
Cycles: Remove back some tweaks for inline policies

While kernel compiles fine, it seems generated binary is not really correct.

4 years agoMake game engine ready for FFmpeg-3.0 as well
Sergey Sharybin [Tue, 16 Feb 2016 15:29:47 +0000 (16:29 +0100)]
Make game engine ready for FFmpeg-3.0 as well

4 years agoCorrect label for simple deform modifier
Campbell Barton [Tue, 16 Feb 2016 14:55:40 +0000 (01:55 +1100)]
Correct label for simple deform modifier

4 years agoCycles: Some cleanup, should be no functional changes
Sergey Sharybin [Tue, 16 Feb 2016 14:32:26 +0000 (15:32 +0100)]
Cycles: Some cleanup, should be no functional changes

Addressing meaningful feedback from coverity.

4 years agoFix items in mode pie changing position by supporting more than 8 items in operator...
Julian Eisel [Tue, 16 Feb 2016 13:50:26 +0000 (14:50 +0100)]
Fix items in mode pie changing position by supporting more than 8 items in operator-enum pies

Now a 'More' item is added to the pie when there are too many items. It opens a sub-pie that contains the remaining items.
Note that this only touches operator-enum pies (like the object mode pie is), it is not a complete support for pies with more than 8 items. For this further design and code work would be needed, but this is too urgent to wait for that.

This is a better fix for T46973, should definitely be applied for 2.77 release.

Patch D1800 by myself with some edits by @campbellbarton, thanks!