1
1

Compare commits

..

427 Commits

Author SHA1 Message Date
73ed9afd03 tag active viewer 2021-06-02 15:22:29 +02:00
de040ccb96 Merge branch 'master' into temp-geometry-nodes-viewer-node 2021-06-02 14:34:28 +02:00
6cd64f8caa Add workaround for gcc 11 compiler bug
Differential Revision: https://developer.blender.org/D11462
2021-06-02 14:13:55 +02:00
f92f5d1ac6 Keymap: use D-Key for view-pie menu
Use the D-key to access the view menu instead of the Tilda key,
which isn't accessible on some international layouts.

To resolve the conflicts the following changes have been made.

- `D` (motion) opens the view menu.
- `D` (mouse-button) uses grease-pencil (as it does currently).
- `Tilda` is used to for "Object Mode Transfer" instead of the View menu.

See T88092 for details.

Reviewed By: Severin

Ref D11189
2021-06-02 21:34:30 +10:00
69e15eb1e4 Object: support running transfer mode in any object mode
There is no need to limit this to sculpt mode,
prepare for key short cut changes, see: T88092.
2021-06-02 21:29:15 +10:00
facc62d0d5 Docs: 2.93 release description for Linux appdata 2021-06-02 13:22:16 +02:00
a9ba86c046 Merge branch 'master' into temp-geometry-nodes-viewer-node 2021-06-02 12:17:08 +02:00
46447594de Fix T88567: Cryptomatte only works for the first View Layer.
The view layer was always set to 0. This patch increments it.
2021-06-02 10:05:39 +02:00
507c19c0f7 Docs: formalize naming for generic callbacks in BKE_callbacks.h
Add a doc-string explaining the purpose of each call back and
how they should be used.

Also add a currently unused callback 'POST_FAIL' that is to be used in
cases the action fails - giving script authors, a guarantee that a
call to `pre` will always have a matching `post/post_fail` call.

- D11422: adds a callback that can use 'post_fail'.
- T88696: proposed these conventions.
2021-06-02 17:35:24 +10:00
3b3742c75f Cleanup: trailing commas to avoid right shift
This matches most declarations already in this file.
2021-06-02 17:06:12 +10:00
b5a883fef9 Cleanup: spelling in comments 2021-06-02 17:04:57 +10:00
711ddea60e Fix assert with geometry node output
The previous commit (my own) returned early without providing a value
for the node's output geometry set, which is required.
2021-06-01 23:50:23 -04:00
6b5bbd22d9 Cleanup: Avoid duplicating node input retrieval
Pass the selection name and the invert argument to each component
instead of retrieving them every time.
2021-06-01 17:51:21 -04:00
Wannes Malfait
464797078d Geometry Nodes: Add Delete Geometry Node
This node is similar to the mask modifier, but it deletes the elements
of the geometry corresponding to the selection, which is retrieved as
a boolean attribute. The node currently supports both mesh and point
cloud data. For meshes, which elements are deleted depends on the
domain of the input selection attribute, just like how behavior depends
on the selection mode in mesh edit mode.

In the future this node will support curve data, and ideally volume
data in some way.

Differential Revision: https://developer.blender.org/D10748
2021-06-01 17:32:03 -04:00
3400ba329e VSE: Use own category for metadata panel
Metadata panel was visible in each category. In other editors, this
panel is usually placed in category with other source media properties.
In sequencer, metadata is transfered over while compositing and relation
to particular strip is lost, therefore separate category for metadata
seems to be best option. Since Metadata panel is alone in this category,
it will be open by default.
2021-06-01 23:06:31 +02:00
17b09b509c Geometry Nodes: Skip calculating normals in transform node
This commit skips the eager recalculation of mesh normals in the
transform node. Often another deformation or topology-altering
operation will happen after the transform node, which means the
recalculation was redundant anyway.

In one of my test cases this made the node more than 14x faster.
Though depending on the situation the cost of updating the normals
may just be shifted elsewhere.
2021-06-01 12:03:59 -04:00
5b6e0bad1b Fix T88715: particle size influence texture not working for 'keyed' or 'none' physics types
This was reported for the special case of mapping with "Strand /
Particle" coords, but was not working with other coordinates either.

Dont see a reason for not supporting Size influence textures for these
kinds of particles (and since these types of particles have an "age"
like all others as well, even the "Strand / Particle" coords are
supported here as well)

Maniphest Tasks: T88715

Differential Revision: https://developer.blender.org/D11449
2021-06-01 17:54:20 +02:00
404b946ac0 LibOverride: Fix again infinite loop in resync in some complex/degenerated cases.
Broken in recent refactor of (recursive)resync, reported by studio,
thanks.
2021-06-01 17:37:22 +02:00
842375e1ed test adding shortcut for geo viewer 2021-06-01 16:23:43 +02:00
41bfb64da1 Merge branch 'master' into temp-geometry-nodes-viewer-node 2021-06-01 16:05:30 +02:00
6583fb67c6 Geometry Nodes: add empty material slot to new meshes
This fixes T88455 by adding an empty material slot to newly
generated meshes. This allows the object to overwrite the
"default" material without any extra nodes. Technically,
all polygons reference the material index 0 already, so it
makes sense to add a material slot for this material index.

Differential Revision: https://developer.blender.org/D11439
2021-06-01 15:25:01 +02:00
James Monteath
8a63466ca3 BuildBot: Cleanup
Removing scripts that were placed in the source tree that would drive
the old buildbot. With the new buildbot in place these scripts aren't
being used anymore.

The buildbit is currently driven by
`build_files/config/pipeline_config.json`. In the near future make
update would also use this config.

Overview of the new buildbot: https://wiki.blender.org/wiki/Infrastructure/BuildBot

Work done by James Monteath.
2021-06-01 15:18:11 +02:00
db59a5f9f0 Merge branch 'master' into temp-geometry-nodes-viewer-node 2021-06-01 13:54:07 +02:00
6899dcab53 Fix T88658: Force Fields of curve shape can crash if curve has only one point
`bvhtree_from_mesh_edges_create_tree` can actually leave the BVHTree
NULL (e.g. if no edges are present).

Now dont allocate `BVHTreeFromMesh` on the `SurfaceModifierData` at all
in case the tree would be NULL anyways.
Places like `get_effector_data` check for `SurfaceModifierData`-
>`BVHTreeFromMesh` and dont try to stuff like getting a closest point on
surface, which would crash as soon as BVHNodes would need to be accessed
(from the NULL BVHTree).

Maniphest Tasks: T88658

Differential Revision: https://developer.blender.org/D11443
2021-06-01 13:38:29 +02:00
c078540512 Cleanup: remove unused parameter. 2021-06-01 13:18:41 +02:00
6f1af44695 Cleanup: remove unused parameter. 2021-06-01 13:18:41 +02:00
633b70905a Cleanup: NULL->nullptr. 2021-06-01 13:18:41 +02:00
f0d93a71df Cleanup: API for MeshExtractRunTimeData. 2021-06-01 13:18:41 +02:00
98c6626729 DrawManager: Use CPP for Mesh Extraction Scheduling.
More cleanups will come to make this more CPP-like.
2021-06-01 13:18:41 +02:00
4b57bbac22 Cleanup: LibOverride: rename 'local' group tagging functions to 'overrides'.
Since we use them on linked data now as well, 'local' does not fit them
anymore.
2021-06-01 12:29:56 +02:00
bf8659930f Fix libOverride resync issues in some corner-cases.
This commit fixes two different issues:
* In some cases, when an object was added to a sub-collection and used
  into a different subcollection, and the root common collection would
  not need to be resynced, it would end up creating multiple overrides
  of the new object. This was affecting both normal and recursive
  resync.
* In recurisve resync case, the barrier code to define what is part or
  not of a override group/hierarchy was wrong.

Note that the current solution for the first issue is sub-optimal (it
goes back to the root of the override hierarchy and resync the whole
thing), a better solution is TODO for now.
2021-06-01 12:01:06 +02:00
f25316e97a IDManagement: BKE_libblock_rename: assert we get local ID.
For now at least, linked data should never be renamed that way.
2021-06-01 12:01:06 +02:00
84e16c4992 Fix BLI_libblock_ensure_unique_name not ignoring linked IDs.
This function would considere that there was a name conflict even in
case existing ID would be a linked one.

This is only a (symbolic) perforance improvement and logical fix, since
`BKE_id_new_name_validate` would not do that mistake anyway.
2021-06-01 12:01:06 +02:00
db15c9d1bd ID Management: Allow unique name check for linked IDs too.
This is mandatory for liboverride resync, since this feature may imply
we have to create linked overrides in libraries, and there may be
several copies of those.

This is also a first step to a more general support of IDmanagement-editing
library data.

Note that this commit should have absolutely no effect on current code,
as the only function allowed to check unique names for linked IDs
currently is `BKE_libblock_management_main_add`, which is unused.

This commit also adds some basic testing for `BKE_id_new_name_validate`.
2021-06-01 12:01:06 +02:00
27910ccce6 Cleanup: clang-tidy
* `readability-redundant-member-init`
* `readability-inconsistent-declaration-parameter-name`
* Remove constructor that can be defaulted.
2021-06-01 12:00:16 +02:00
9adfd278f7 Compositor: Full-frame base system
This patch adds the base code needed to make the full-frame system work for both current tiled/per-pixel implementation of operations and full-frame.

Two execution models:
- Tiled: Current implementation. Renders execution groups in tiles from outputs to input. Not all operations are buffered. Runs the tiled/per-pixel implementation.
- FullFrame: All operations are buffered. Fully renders operations from inputs to outputs. Runs full-frame implementation of operations if available otherwise the current tiled/per-pixel. Creates output buffers on first read and free them as soon as all its readers have finished, reducing peak memory usage of complex/long trees. Operations are multi-threaded but do not run in parallel as Tiled (will be done in another patch).

This should allow us to convert operations to full-frame in small steps with the system already working and solve the problem of high memory usage.

FullFrame breaking changes respect Tiled system, mainly:
- Translate, Rotate, Scale, and Transform take effect immediately instead of next buffered operation.
- Any sampling is always done over inputs instead of last buffered operation.

Reviewed By: jbakker

Differential Revision: https://developer.blender.org/D11113
2021-06-01 10:51:53 +02:00
930ad9257d Cleanup: Split draw_cache_extract_mesh into multiple files.
draw_cache_extract_mesh for task scheduling. Will be refactored to draw_cache_extract_mesh_scheduling later on after migrating to CPP.

draw_cache_extract_mesh_render_data extraction of mesh render data from edit mesh/mesh into a more generic structure.

draw_cache_extract_mesh_extractors containing all the extractors. This will be split up further into a single file per extractor.
2021-06-01 09:23:37 +02:00
13deb5088a Cleanup: split face tessellation into inline functions
Prepare for multiple code-paths that recalculate tessellation.
2021-06-01 15:30:53 +10:00
b8d0f28f70 Cleanup: split bmesh tessellation into it's own file
Prepare for further refactoring for these functions.
2021-06-01 14:04:00 +10:00
3a18e304be Cleanup: remove disabled face tessellation logic
This was kept since these blocks are easier to follow.
Remove as the overall result wasn't so readable
(especially with nested ifdef's).

Replace disabled code with comment on the indices used for quads/tris.
2021-06-01 13:26:26 +10:00
5e7fb77dc4 BMesh: remove checks for tessellating 2 sided faces
2 sided faces aren't supported and will cause problems in many areas
of Blender's code.

Removing (implied) support for faces with fewer than 3 sides
means the total number of triangles is known ahead of time.

This simplifies adding support for multi-threading and partial updates
to an existing tessellation - as the face and loop indices can be used
to access the range of triangles associated with a face.

Also correct outdated comments.
2021-06-01 13:25:00 +10:00
f8ce744c83 Cleanup: correct sculpt quat argument size 2021-06-01 12:50:01 +10:00
f71cf99616 GPU: add 2D projection function
When projecting into screen space Z value isn't always needed.
Add 2D projection functions, renaming them to avoid accidents
happening again.

- Add GPU_matrix_project_2fv
- Add ED_view3d_project_v2
- Rename ED_view3d_project to ED_view3d_project_v3
- Use the 2D versions of these functions when the Z value isn't used.
2021-06-01 12:49:53 +10:00
c145cb7998 Fix buffer overrun in paint_line_strokes_spacing
Error in 87cafe92ce
2021-06-01 12:49:20 +10:00
e7b8a3cb0a Cleanup: spelling in comments 2021-06-01 12:49:18 +10:00
c59d2c739d Cleanup: spelling in comments 2021-06-01 01:45:17 +02:00
f253e59221 Docs: Limit the OCIO env vars that we document
Brecht mentioned that these are a bit obscure and don't make much sense 
to override these.
2021-05-31 19:19:00 -04:00
8180d478e1 Speedup exact boolean by avoiding some mallocs and frees.
This is from patch D11432 from Erik Abrahamsson. He found that
in some mpq3 functions called frequently from loops, passing in
buffers for termporary mpq3 values can save substantial time.
On my machine, his example in that patch went from 9.48s to 7.50s
for the boolean part of the calculation. On his machine, a running
time went from 17s to 10.3s.
2021-05-31 17:03:48 -04:00
73967e2047 Fix undeclared identifiers with 'DEBUG_TIME'
These identifiers were accidentally removed in rB44d2479dc36f.
2021-05-31 16:28:14 -03:00
09fb81f66d Merge branch 'blender-v2.93-release' 2021-05-31 19:23:49 +02:00
25316ef9d7 Cycles: optimize 3D viewport rendering with camera passepartout
If the area outside the camera is fully opaque, don't render it.

Contributed by Kdaf.

Differential Revision: https://developer.blender.org/D11182
2021-05-31 19:23:44 +02:00
875a8a6c79 Cleanup: Replace fseek() calls with BLI_fseek()
The fseek() function on Windows only accepts a 32-bit long offset
argument. Because of this we have our own version, BLI_fseek(), which
will use 64-bit _fseeki64() on Windows. This patch just replaces some
fseek() calls with BLI_fseek().

Differential Revision: https://developer.blender.org/D11430

Reviewed by Brecht Van Lommel
2021-05-31 10:08:58 -07:00
James Monteath
ee54a8ace7 Update all README to clearify intention or usage
Add snap configuration file used by Buildbot snap store steps1
2021-05-31 18:58:42 +02:00
261a10edb0 Display source video fps in the VSE
Now FPS is displayed in the video source for videos to provide easy
access.

Reviewed By: Richard Antalik

Differential Revision: http://developer.blender.org/D11441
2021-05-31 18:22:24 +02:00
d647e730fb Win: Fix warnings as errors being off for bmesh
bf_bmesh historically always build with the /WX flag
on windows making all warnings errors, somewhere along
the way this has broken for msbuild, ninja still exhibits
the expected behaviour.

The flags are still passed to the target, and I've validated
they are there when the add_library call fires, but they
somehow never make it to the generated msbuild project files.

I suspect this is a cmake bug but I'm seemingly unable
to extract a repro case to file a bug upstream.

Setting the same options target_compile_options seems to work,
I'm not happy about the unexplained nature of the breakage
but this will have to do for now.
2021-05-31 09:59:29 -06:00
46a14bd6a3 Fix T88670: Load Previous Settings does not copy symlinks
The same code existed in 2.82 and earlier so this should be safe. Removing the
custom implementation of shutil.copytree in f34d5d9 did not correctly add back
the option to copy symlinks.
2021-05-31 17:30:04 +02:00
b862916eaf VSE: Fix missing cache invalidation
Fixes T88606
2021-05-31 17:28:49 +02:00
Jeroen Bakker
44d2479dc3 Refactor: DRW Mesh Extractor: Join the extractors in a same loop
This patch replaces / redoes the entire MeshExtractors system.
Although they were useful and facilitated the addition of new buffers, they made it difficult to control the threads and added a lot of threading overhead.

Part of the problem was in traversing the same loop type in different threads. The concurrent access of the BMesh Elements slowed the reading.

This patch simplifies the use of threads by merging all the old callbacks from the extracts into a single series of iteration functions.

The type of extraction can be chosen using flags.

This optimized the process by around 34%.

Initial idea and implementation By @mano-wii.
Fine-tuning, cleanup by @atmind.

MASTER:
large_mesh_editing:
- rdata 9ms iter 50ms (frame 155ms)
- Average: 6.462874 FPS

PATCH:
large_mesh_editing:
- rdata 9ms iter 34ms (frame 136ms)
- Average: 7.379491 FPS

Differential Revision: https://developer.blender.org/D11425
2021-05-31 17:11:25 +02:00
aebeb85fe0 Windows: Clean-up win 8/8.1 API use
For 2.93 we bumped the minimum windows requirement
to windows 8.1, but did not do any clean-up of any
win 8/8.1 API usage we dynamically accessed though
LoadLibrary/GetProcAddress.

This patch bumps _WIN32_WINNT to 0x0603 (win 8.1)
and cleans up any API use that was accessed in a
more convoluted way than necessary

Differential Revision: https://developer.blender.org/D11331

Reviewed by: harley, nicholas_rishel
2021-05-31 08:56:57 -06:00
d67c13ca76 Cleanup: else-after-return 2021-05-31 15:48:06 +02:00
d94ba979d8 Fix T88569: UI VSE: Menu-based range change, doesn't update the Timeline scrollbar width
Use the appropriate notifier, listeners were already doing the rest
properly.

Maniphest Tasks: T88569

Differential Revision: https://developer.blender.org/D11436
2021-05-31 15:32:33 +02:00
a25dbf243a remove outdated code 2021-05-31 15:06:11 +02:00
e0a1c3da46 Fix T88666: Cryptomatte: EXR sequence does not update when scrubbing the timeline.
Cause is that initializing the cryptomatte session would reset the
current frame of an image sequence. The solution is to always use the
scene current frame so it resets to the correct frame.

This was a todo that wasn't solved after it landed in master.
Needs to be backported to 2.93.
2021-05-31 14:32:39 +02:00
782b0df9c8 recalc when viewer node changed 2021-05-31 13:49:13 +02:00
8969f4ba23 take viewer node into account when searching for output connection 2021-05-31 13:44:15 +02:00
7cea4dc6f5 compute viewed sockets even when it is not required for the output 2021-05-31 13:35:29 +02:00
1e006e58dc remove toggles from node headers 2021-05-31 13:18:35 +02:00
0702871110 handle viewer node in modifier 2021-05-31 13:15:41 +02:00
5ea958a992 update spreadsheet context when activating viewer 2021-05-31 13:08:20 +02:00
7694a098ef initial viewer node 2021-05-31 12:46:39 +02:00
e9f2f17e85 Fix (unreported): TextureOperation inputs have no resolution
When compositor node tree has a texture node, TextureOperation vector inputs  has always {0, 0} resolution instead of having same resolution as TextureOperation which is the expected behaviour for resolutions propagation.

Current TextureOperation determineResolution implementation doesn't determine inputs resolution, breaking propagation of preferred resolution and that's the reason why they are always 0. Setting scene resolution always would mean it is its own resolution and could make sense, but setting it only when preferred resolution is 0, breaks preferred resolution logic affecting other operations as explained in D10972. In any case scene resolution is already the default preferred resolution on viewer and compositor nodes.

Reviewed By: jbakker

Differential Revision: https://developer.blender.org/D11381
2021-05-31 12:26:46 +02:00
83fe479c7f Merge branch 'blender-v2.93-release' 2021-05-31 11:56:17 +02:00
ce649c7344 Fix T88623, T87044: Make encoded videos play correctly in VLC
The issue was two fold. We didn't properly:

1. Initialize the codec default values which would lead to VLC
   complaining because of garbage/wrong codec settings.

2.Calculate the time base for the video. FFmpeg would happily accept
  this but VLC seems to assume the time base value is at least somewhat
  correct and couldn't properly display the frames as the internal time
  base was huge. We are talking about 90k ticks (tbn) for one second of
  video!

This patch initializes all codecs to use their default values and fixes
the time base calculation so it follows the guidelines from ffmpeg.

Reviewed By: Sergey, Richard Antalik

Differential Revision: http://developer.blender.org/D11426
2021-05-31 11:29:08 +02:00
2161840d07 Fix (studio-reported) crash in collection management code.
Code checking for potential collection loop dependencies can be called
in cases where we cannot guarantee that there is no NULL pointers, so we
need to check those. Was already done for objects.

NOTE: doubled-checked by @jbakker, thanks.
2021-05-31 11:25:08 +02:00
2534609262 Revert "Added v2.93 pipeline config for new buildbot."
This reverts commit 632bfee0a5.
This config is only intended for 2.93. Master will get its own config
file after testing that 2.93 is correct.
2021-05-31 11:16:13 +02:00
86eaddb3ca Merge branch 'blender-v2.93-release' 2021-05-31 11:15:50 +02:00
632bfee0a5 Added v2.93 pipeline config for new buildbot. 2021-05-31 11:15:03 +02:00
26fb7b9474 Geometry Nodes: do not create unnecessary geometry components
Previously, making instances real would always create an (empty)
volume and curve component, even when not necessary.

This also fixes T88653.
2021-05-31 11:12:39 +02:00
421c0b45e5 Fix (studio-reported) crash in collection management code.
Code checking for potential collection loop dependencies can be called
in cases where we cannot guarantee that there is no NULL pointers, so we
need to check those. Was already done for objects.
2021-05-31 10:20:23 +02:00
Jeroen Bakker
5f749a03ca Fix T88456: DrawManager: Keep subset RenderMeshData around when geometry does not change.
Reuse loose geometry during selection (and other operations) from
previous calculation. Loose geometry stays the same, but was
recalculated to determine the size of GPU buffers. This patch would
reuse the previous loose geometry when geometry wasn't changed.

Although not the main bottleneck during selection it is measurable.

Master.
`rdata 46ms iter 55ms (frame 410ms)`

This patch.
`rdata 5ms iter 52ms (frame 342ms)`

Reviewed By: mano-wii

Differential Revision: https://developer.blender.org/D11339
2021-05-31 09:33:31 +02:00
a1556fa05c Boolean: applying patch D11431 to speed up hole-tolerant raycast.
This patch from Erik Abrahamsson uses a parallel_for to speed up
the case where the input is not manifold and the "hole_tolerant"
option is set.
In a test case on a 24 core (48 thread) machine, this sped up a
the boolean part on an object with 221k triangles from 12.06s to 0.46s.
2021-05-30 16:37:49 -04:00
e6a69f7653 Docs: Capitalize first word of sentence 2021-05-30 11:09:01 -04:00
7b5796dcaa Docs: clarify description and usage of 'bpy.app.version_file'
Fixes T88669
2021-05-30 11:07:38 -04:00
f5d14e36e8 PyDoc: Use em dash instead of comma for enum items 2021-05-29 12:16:13 -04:00
870aaf3fb6 Docs: Add documentation for 'material_index'
Fixes T88485
2021-05-29 11:22:58 -04:00
6f86d50b92 UI: Match tooltip with interface name 2021-05-29 11:22:58 -04:00
30f7acfffa Docs: Add relevant OCIO envvars to Blender's help message 2021-05-29 11:22:58 -04:00
115547991a Merge branch 'blender-v2.93-release' 2021-05-29 15:29:42 +02:00
3c02e648f3 GPencil: Fix unreported random rotation for single point with texture
When using ``Path`` alignment, if the stroke has one point the texture rotates randomly when move the viewport. This was because with one point is impossible to calculate a path.

Now, if the stroke has only one point, the texture for this stroke is aligned to Object.
2021-05-29 15:29:08 +02:00
James Monteath
ffe7a41540 Fix typos 2021-05-28 20:50:52 +02:00
James Monteath
74c7e21f6c Add and update README.md files for CI script removal 2021-05-28 19:46:53 +02:00
9225fe933a Make encoded video fps correct with ffmpeg < 4.4
Before the FFmpeg commit: github.com/FFmpeg/FFmpeg/commit/1c0885334dda9ee8652e60c586fa2e3674056586
FFmpeg would use deprecated variables to calculate the video fps.

We don't use these deprecated variables anymore, so ensure that the
duration is correct in ffmpeg versions without this fix.

Reviewed By: Sergey, Richard Antalik

Differential Revision: http://developer.blender.org/D11417
2021-05-28 18:37:36 +02:00
3311350670 Fix T87932: Failure to build movie strip proxy
We didn't initialize the scaled proxy frame properly.
This would lead to issues in ffmpeg 4.4 as they are more strict that the API is properly used.

Now we initialize the size and format of the frame.
2021-05-28 18:35:26 +02:00
5b8a41d387 Merge branch 'blender-v2.93-release' 2021-05-28 18:19:07 +02:00
653bbaa246 Geometry Nodes: Polish switch node UI
Based on the task T88006, there are a few simple changes
to make to improve the switch node:
- Change the label to "False" / "True" for clarity
- Change default to geometry, as it's the basic data container in
  geometry nodes.
- Change node class to `NODE_CLASS_CONVERTOR`, which was an oversight
  in the original patch.

I will add the new socket types (material and texture) in a separate commit.
Thanks to @EitanSomething for the original patch.

Differential Revision: https://developer.blender.org/D11165
2021-05-28 12:17:04 -04:00
c369382977 EEVEE: Fix NaN caused by ensure_valid_reflection()
This was caused by unsafe sqrt calls.

Fixes T86578 white artifacts in EEVEE

Reviewed By: brecht, dfelinto

Differential Revision: https://developer.blender.org/D11428
2021-05-28 18:16:56 +02:00
c0ce7fce89 Merge branch 'blender-v2.93-release' 2021-05-28 12:09:08 -04:00
20e250dae3 Fix T88601: Attribute Compare boolean doesn't expose socket
While we could make this node work for boolean inputs in the future,
currently it's really just designed to compare "float-like" inputs.
Many comparison modes don't even make sense for boolean inputs.
Therefore, the simplest fix for this bug is just to disable the
boolean attribute input modes for this node.

Differential Revision: https://developer.blender.org/D11427
2021-05-28 12:04:52 -04:00
James Monteath
fc2b56e68c Add branch based pipeline config file used by buildbot and make_update.py script. 2021-05-28 17:53:02 +02:00
a4700549e0 GPencil: Fix unreported random rotation for single point with texture
When using ``Path`` alignment, if the stroke has one point the texture rotates randomly when move the viewport. This was because with one point is impossible to calculate a path.

Now, if the stroke has only one point, the texture for this stroke is aligned to Object.
2021-05-28 17:15:44 +02:00
c65f4b3d76 DrawManager: Early return for buffer cache creation
No real functional changes.

This is useful for benchmark cases when `cache->uv_cage` is passed but
has no buffers are requested.
2021-05-28 11:51:13 -03:00
11e32332dd Geometry Nodes: Add Mesh to Curve Node
This node creates poly curve splines from mesh edges. A selection
attribute input allows only using some of the edges from the mesh.
The node builds cyclic splines from branchless groups of edges where
possible, but when there is a three-way intersection, the spline stops.

The node also transfers all attributes from the mesh to the resulting
control points. In the future we could add a way to limit that to a
subset of the attributes to improve performance.

The algorithm is from Animation Nodes, written by @OmarSquircleArt.
I added the ability to use a selection, attribute transferring, and
used different variable names, etc, but other than that the algorithm
is the same.

Differential Revision: https://developer.blender.org/D11265
2021-05-28 10:42:22 -04:00
080623b8ac Fix incorrect Denoise node SSE 4.1 warning on macOS Intel 2021-05-28 16:02:43 +02:00
418888f1c9 MSVC: Fix build error with 16.10/11
Not entirely sure why this was not an issue for 16.9
but TBB includes the Windows.h header which by default
will define min and max macro's

These collide with the stl versions in <algorithm>

This patch requests Windows.h not to define the
problematic macro's, resolving the conflict.
2021-05-28 07:57:21 -06:00
bdac47f8d4 Merge branch 'blender-v2.93-release' 2021-05-28 15:47:38 +02:00
adafd7257d Fix T88635: VSE: Select Linked gives unpredictable results
Caused by {rB66923031e6f2}.

Code would process unselected sequences and skip selected, needs to be
the other way around.

Maniphest Tasks: T88635

Differential Revision: https://developer.blender.org/D11424
2021-05-28 15:41:56 +02:00
97ccd592ce Fix crash in liboverride resync.
Reported by studio (@andy), thanks.
2021-05-28 15:33:04 +02:00
James Monteath
51bbdfbef3 Moved to new git repo 2021-05-28 12:16:45 +02:00
63e50cf265 Fix T88499: Copy data path operator does not consider library affiliation
When using the operator `ui.copy_data_path_button(full_path=True)` ({key
ctrl shift Alt C} on hover) the copied path does not consider the
library origin. That means that when there is a name clash the data path
is not accurate and refers to the local item instead.

This patch adds the library (if the ID is linked) of the returned string
from RNA_path_full_ID_py.

bpy.data.objects["Cube", "//library.blend"] instead of
bpy.data.objects["Cube"]

note: parsing this happens in
pyrna_prop_collection_subscript_str_lib_pair_ptr

Maniphest Tasks: T88499

Differential Revision: https://developer.blender.org/D11412
2021-05-28 11:06:55 +02:00
James Monteath
5baf1dddd5 Buildbot related files have been moved to own repository 2021-05-28 10:16:39 +02:00
3ec57ce6c7 Tests: add utility to generate interactive user actions
A utility that supports passing in actions as command line arguments for
writing reproducible interactions, benchmarking, profiling and testing.

Unlike regular scripts this is able to control model operators usefully.

Typical ways of controlling Blender using this utility are via
operator id's, menu search and explicit events.
Others methods can be added as needed.

See the doc-string for example usage.
2021-05-28 16:59:38 +10:00
3bee77bb7c Cleanup: use static set syntax 2021-05-28 16:58:50 +10:00
6b03621c01 DrawManager: Use Compute Shader to Update Hair.
This patch will use compute shaders to create the VBO for hair.
The previous implementation uses transform feedback.

Timings before: between 0.000069s and 0.000362s.
Timings after:  between 0.000032s and 0.000092s.

Speedup isn't noticeable by end-users. The patch is used to test
the new compute shader pipeline and integrate it with the draw
manager. Allowing EEVEE, Workbench and other draw engines to
use compute shaders with the introduction of `DRW_shgroup_call_compute`
and `DRW_shgroup_vertex_buffer`.

Future improvements are possible by generating the index buffer
of hair directly on the GPU.

NOTE: that compute shaders aren't supported by Apple and still use
the transform feedback workaround.

Reviewed By: fclem

Differential Revision: https://developer.blender.org/D11057
2021-05-28 08:16:26 +02:00
4a1ba155d5 Merge branch 'blender-v2.93-release' 2021-05-27 21:27:48 -04:00
02395fead7 Docs: Update RNA to User Manual mappings 2021-05-27 21:25:31 -04:00
0a83f32d79 Fix T86465: Annotation Tool is missing in VSE Preview toolbar
Added missing topbar in VSE.

Also added the Stabilizer options to Topbar for all modes.

Reviewed By: mendio, pepeland

Maniphest Tasks: T86465

Differential Revision: https://developer.blender.org/D11347
2021-05-27 19:53:23 +02:00
758c115103 Fix own crash in today's rBf68288a8746f. 2021-05-27 19:32:35 +02:00
311ffd967b LibOverride: refactor recursive resync.
We need to re-evaluate what needs to be resynced after each step of
processing overrides from a given 'indirect level' of libraries.
Otherwise, recusrive overrides (overrides of linked overrides) won't
work.

Note that this should not change too much in practice currently, since
there are other issues with recursive overrides yet.

Also, checks (CLOG errors) added show that some ID (node trees) seem to
be detected as needing resynced even after beig just resynced, this
needs further investigation still. Could be though that it is due to
limit currently set on nodetrees, those are always complicated
snowflakes to deal with...
2021-05-27 19:32:35 +02:00
530f2994e9 Fix T88614: Mixdown crashes Blender 2.92.0 and 3.0.0 Alpha
The problem is caused by the most recent ffmpeg version (4.4) which
needs channels to be set when submitting a frame for encoding.
2021-05-27 19:05:17 +02:00
dfa3dcbab9 Fix T88625: Multiobject UV hiding/unhiding does not work with UV_SYNC_SELECTION
Oversight in {rB470f17f21c06}.

Hiding was only done for the first mesh, then the operator finished (in
case of UV_SYNC_SELECTION).
Now just continue to the next.

Maniphest Tasks: T88625

Differential Revision: https://developer.blender.org/D11413
2021-05-27 17:27:38 +02:00
8590cb26a9 Merge branch 'blender-v2.93-release' 2021-05-27 17:15:39 +02:00
ba4228bcf7 Revert "EEVEE: Ensure Reflection: Use new implementation"
Both before and after can have artifacts with some normal maps, but this seems to give
worse artifacts on average which are not worth the minor performance increase.

This reverts commit 5c4d24e1fd.

Ref T88368, D10084
2021-05-27 17:12:23 +02:00
3025c34825 Geometry Nodes: Expose texture and material inputs to modifier
This allows choosing material and texture sockets for the group input
node in the modifier. Note that currently grease pencil materials are
displayed in the list, even though grease pencil data is not supported
yet by geometry nodes. That is more complicated to fix in this case,
since we use IDProperties to store the dynamic exposed inputs.

Differential Revision: https://developer.blender.org/D11393
2021-05-27 11:06:08 -04:00
24b2482ad9 Cleanup: Fix forward declaring class with "struct" 2021-05-27 11:00:00 -04:00
a3edf4a381 Fix build error: Make CurveEval a struct
We need a pointer to this in DNA, which means it cannot be a class.
2021-05-27 10:50:43 -04:00
e8ca635e43 Cleanup: rename blender-launcher source file.
blender-laucher.c was not an ideal name for this file
since it's not directly clear it is windows only.

This change renames it to blender_launcher_win32.c
to be more in line with other win32 specific files
we have.
2021-05-27 08:10:31 -06:00
ac833108db Geometry Nodes: Draw curve data in the viewport
This patch adds relatively small changes to the curve draw
cache implementation in order to draw the curve data in the
viewport. The dependency graph iterator is also modified
so that it iterates over the curve geometry component, which
is presented to users as `Curve` data with a pointer to the
`CurveEval`

The idea with the spline data type in geometry nodes is that
curve data itself is only the control points, and any evaluated
data with faces is a mesh. That is mostly expected elsewhere in
Blender anyway. This means it's only necessary to implement
wire edge drawing of `CurveEval` data.

Adding a `CurveEval` pointer to `Curve` is in line with changes
I'd like to make in the future like using `CurveEval` in more places
such as edit mode.

An alternate solution involves converting the curve wire data
to a mesh, however, that requires copying all of the data, and
since avoiding it is rather simple and is in-line with future plans
anyway, I think doing it this way is better.

Differential Revision: https://developer.blender.org/D11351
2021-05-27 10:08:40 -04:00
5621a8ed7f Fix T88452: Point Separate crash on curve component
The point separate node should create a point cloud from control points
in this case, but for now disable the node on curves to avoid the crash.
2021-05-27 09:37:19 -04:00
a12fd5c3ad Cleanup: Use consistent variable names 2021-05-27 09:27:08 -04:00
5721c89ba8 Cleanup: rename BKE_main_id_{clear_newpoins => newptr_and_tag_clear}
It wasn't obvious this function cleared the tag as well.
2021-05-27 22:44:02 +10:00
Erik Abrahamsson
9200e73136 Cleanup: remove duplicate LIB_TAG_NEW untag code
This patch removes unnecessary calls to `BKE_main_id_tag_all` where the
same job is done by `BKE_main_id_clear_newpoins` on the following line.

Reviewed By: campbellbarton, mont29

Ref D11379
2021-05-27 22:42:41 +10:00
6ad4b8b764 LineArt: List Optimization for tile linked data.
Use array instead of ListBase for line art
bounding area linked triangles and edges.

Reviewed By: Sebastian Parborg (zeddb)

Differential Revision: https://developer.blender.org/D11302
2021-05-27 20:33:02 +08:00
f45270b4fb Cleanup: Line art variable naming.
Change `reln` to `eln`.

Reviewed By: Sebastian Parborg (zeddb)

Differential Revision: https://developer.blender.org/D11411
2021-05-27 18:47:13 +08:00
42fba7f0d2 LibOverride: Add heuristic protection against infinite loop due to libraries inter-dependencies.
This is not supposed to happen, but better be safe than sorry, and
assume it is beyond unlikely that someone would use chains of over 10k
linked libraries.
2021-05-27 11:55:31 +02:00
ac2266fe57 Cleanup: Unused include in Cycles 2021-05-27 11:36:30 +02:00
2414a5787d Refactor: Move display pass to Cycles viewport parameters
Allows to centralize storage and modification checks in a single place,
avoiding duplication in the synchronization code.

Ideally we would somehow be able to more granularly modify Cycles side
objects. Leaving this for a future decision, because it might be better
to implement it as a graph on the sync side.
2021-05-27 11:36:30 +02:00
7b5f4c8837 Cleanup: Redundnat member init in Cycles viewport parameters 2021-05-27 11:36:30 +02:00
616b071c4f Refactor: Naming in Cycles viewport methods
Makes it more explicit they operate on shading/light.

Gives room to move more viewport related settings into this class and
cover with specific or generic modification checks.
2021-05-27 11:36:30 +02:00
15df83e9c9 Cleanup: Use logical OR in Cycles background shader 2021-05-27 11:36:30 +02:00
8f43e31a71 Cleanup: const qualifier of return type 2021-05-27 11:36:30 +02:00
4ea7742973 Refactor: Remove friend class from Cycles viewport parameters
Such pattern should only be used when it is really needed. Otherwise
just stick to a more regular design, without worrying who is the user
of the class. Otherwise it will be annoying to subclass or unit test.
2021-05-27 11:36:30 +02:00
bf7a67c5c6 Refactor: Rename pass accessor in viewport parameters
No need to state that it is a viewport display pass, since the method
is within viewport parameters it is implied that parameters do belong
to the viewport.

Brings this code closer to the Cycles-X branch.
2021-05-27 11:36:30 +02:00
06d48367fa Fix (studio-reported) infinite loop in resync code.
Very stupid mistake in libraries indirect-level building code, was not
skipping 'loop-back' ID pointers.

Note that we also need some level of checks for the case where there
would be an actual dependency loop between libraries, this is not
supposed to be possible, but better be safe than sorry. Will add in next
commit.
2021-05-27 10:49:31 +02:00
46e1ad711c Cleanup: inconsistent parameter name 2021-05-27 10:46:43 +02:00
ffa2a1771e Cleanup: inconsistent parameter name 2021-05-27 10:41:40 +02:00
fd0bb24e4a Cleanup: simplify logic for copying vector button as text
An arbitrary size offsets was used in float_array_to_string,
simplify the loop, use exact size limits.

Also rename variables so it's clear which array the length apply to.
2021-05-27 17:59:21 +10:00
41f2ea4045 Fix incorrect BLI_snprintf usage
Event though in practice this wasn't causing problems as the fixed size
buffers are generally large enough not to truncate text.

Using the result from `snprint` or `BLI_snprintf` to step over a fixed
size buffer allows for buffer overruns as the returned value is the size
needed to copy the entire string, not the number of bytes copied.

Building strings using this convention with multiple calls:

    ofs += BLI_snprintf(str + ofs, str_len_max - ofs);

.. caused the size argument to become negative,
wrapping it to a large value when cast to the unsigned argument.
2021-05-27 17:59:21 +10:00
1276d0024f Cleanup: specify array sizes, remove warnings in comments 2021-05-27 17:59:21 +10:00
f0342065b1 Cleanup: spelling 2021-05-27 17:59:21 +10:00
2ad3a1c318 Nodes: fix material node copied over when socket is copied
This was missing from rB207472930834a2916cf18bbdff51bcd77c6dd0c0.
2021-05-27 09:58:45 +02:00
223c6e1ead Geometry Nodes: disable multi-threading in evaluator for now
A deadlock could happen under certain circumstances when
geometry nodes is used on multiple objects.
Once T88598 is resolved, multi-threading can be enabled again.

Differential Revision: https://developer.blender.org/D11405
2021-05-27 09:43:11 +02:00
7d20cf92dd Cleanup: use UndoMesh as links instead of allocating LinkData
While the advantage isn't large,
it's simpler to skip the intermediate link.

Also remove unused next and previous struct members
from MeshUndoStep_Elem.
2021-05-27 16:44:21 +10:00
deb71cef38 Undo: resolve inefficient edit-mesh memory use with multiple objects
When editing more than 1 object at a time, complete copies of each mesh
were being stored. Now the most recent undo-data for each mesh is used
(when available).
2021-05-27 16:44:21 +10:00
a10e41b02e Cleanup: Remove completed "TODO" comment
Since rBb67fe05d4bea, the dependency graph supports relations
on collection geometry, and the nodes modifier uses that.
2021-05-26 22:32:53 -04:00
b132dd042e Cleanup: Simplify spline point attribute materialize functions
- Iterate over the mask directly instead of using an index.
- Use Span slice and copy_from instead of a lower level function.
2021-05-26 22:22:09 -04:00
c97b6215a3 Geometry Nodes: Support interpolation between curve domains
This commit adds interpolation from the point domain to the spline
domain and the other way around. Before this, spline domain attributes
were basically useless, but now they are quite helpful as a way to use
a shared value in a contiguous group of points.

I implementented a special virtual array for the spline to points
conversion, so that conversion should be close to the ideal performance
level, but there are a few ways we could optimize the point to spline
conversion in the future:
 - Use a function virtual array to mix the point values for each spline
   on demand.
 - Implement a special case for when the input virtual array is one of
   the virtual arrays from the spline point attributes. In other words,
   decrease curve attribute access overhead.

Differential Revision: https://developer.blender.org/D11376
2021-05-26 22:14:59 -04:00
f3944cf503 Win: Add launcher to hide the console window flash
This patch fixes a long-standing complaint from users:
the console window shortly flashing when they start
blender.

This is done by adding a new executable called
blender-launcher.exe which starts blender.exe while
hiding the console.

Any command line parameters given to blender-launcher
will be passed on to blender.exe so it'll be a drop
in replacement.

Starting blender.exe on its own will still function as
a proper console app so no changes required here for
users that use blender for batch processing.

Notable changes:

Registering blender (-R switch) will now register
blender-launcher as the preferred executable.

This patch updates the installer and updates the
shortcuts to start blender-launcher.exe rather
than blender.exe

Differential Revision: https://developer.blender.org/D11094

Reviewed by: brecht, harley
2021-05-26 20:02:35 -06:00
Germano Cavalcante
6fc9ec9257 Cleanup: Specify amount of buffers through preprocessor directives 2021-05-26 22:52:59 -03:00
e5b51cb511 Fix T88603: Crash with spline attributes after curve resample
The output curve's spline attribute domain custom data needs to be
reallocated with the correct length after adding the splines.
2021-05-26 21:12:38 -04:00
2c607ec2f6 Revert "DrawManager: Use Compute Shader to Update Hair."
This reverts commit 8f9599d17e.

Mac seems to have an error with this change.
```
                 ERROR: /Users/blender/git/blender-vdev/blender.git/source/blender/draw/intern/draw_hair.c:115:44: error: use of undeclared identifier 'shader_src'
                 ERROR: /Users/blender/git/blender-vdev/blender.git/source/blender/draw/intern/draw_hair.c:123:13: error: use of undeclared identifier 'shader_src'
                 ERROR: make[2]: *** [source/blender/draw/CMakeFiles/bf_draw.dir/intern/draw_hair.c.o] Error 1
                 ERROR: make[1]: *** [source/blender/draw/CMakeFiles/bf_draw.dir/all] Error 2
                 ERROR: make: *** [all] Error 2

```
2021-05-26 20:32:05 +02:00
7438f0c6c0 Cleanup: array-parameter warning with GCC 11
Pass the string size as this is less error prone in general.
2021-05-27 02:32:44 +10:00
c07c7957c6 Revert "Cycles: optimize ensure_valid_reflection(), reduces render time by about 1%"
Both before and after can have artifacts with some normal maps, but this seems to give
worse artifacts on average which are not worth the minor performance increase.

This reverts commit 21bc1a99ba.

Ref T88368, D10084
2021-05-26 18:27:21 +02:00
f12e2ec088 Fix buildbot CUDA/OptiX warnings on macOS
Explicitly disable these, rather than relying on them not being found.
Also, don't duplicates the architectures list.
2021-05-26 18:27:17 +02:00
de3d54eb9a GPencil: Cleanup - Conform with RNA naming scheme
The newly added `disable_masks_viewlayer` RNA property did not conform
with the RNA naming scheme. This renames it to `use_viewlayer_masks`.
2021-05-26 18:23:31 +02:00
df32b50344 Fix T88111: Skin modifier assets within invalid face normals
The skin modifier was moving vertices without updating normals for the
connected faces, this happened when smoothing and welding vertices.

Reviewed By: mont29

Ref D11397
2021-05-27 01:52:45 +10:00
6b00df1105 Cleanup: shadow warning
Move reproject_type into an extern, to avoid declaring multiple times.
2021-05-27 01:52:45 +10:00
72d660443b LibOverride: add recursive resync.
Recursive resync means also resyncing overrides that are linked from
other library files into current working file.

Note that this allows to get 'working' files even when their
dependencies are out of sync. However, since linked data is never
written/saved, this has to be re-done every time the working file is
loaded, until said dependencies are updated properly.

NOTE: This is still missing the 'report' side of things, which is part
of a larger task to enhance reports regarding both linking, and
liboverrides (see T88393).

----------

Technical notes:

Implementing this proved to be slightly more challenging than expected,
mainly because one of the key aspects of the feature was never done in
Blender before: manipulating, re-creating linked data.

This ended up moving the whole resync code to use temp IDs out of bmain,
which is better in the long run anyway (and more aligned with what we
generally want to do when manipulating temp ID data). It should also
give a marginal improvement in performances for regular resync.

This commit also had to carefully 'sort' libraries by level of indirect
usage, as we want to resync first the libraries that are the least directly
used, i.e. libraries that are most used by other libraries.
2021-05-26 17:05:01 +02:00
51ca9833d9 LibOverride: add helper to retrieve override data from an ID.
Embedded IDs do not own their own override data, but rather use the one
from their owner.
2021-05-26 17:05:01 +02:00
daf39af70a IDManagement: Shapekey: add a owner_get callback.
Even though shepkeys are not strictly speaking an embedded data, they
share quiet a few points with those, and from liboverride perspective
they are embedded, so...
2021-05-26 17:05:01 +02:00
653d39cec3 LibOverride: Do not try to generate override data of linked data.
This is obviously not saved, and should never be editable, so was only a
waste of time.
2021-05-26 17:05:01 +02:00
6cbe5dd1c3 ID management: remapping: add flag to enforce refcounting handling.
While indeally we should only skip refcounting when relevant tag is set,
doing this in remapping code is too risky for now.

Related to previous commit and T88555.
2021-05-26 17:05:01 +02:00
ee849ca0f8 ID management: Do not assume that NO_MAIN means NO_USER_REFCOUNT
While this is still very fuzzy in current code, this old behavior makes
it close to impossible to efficiently use out-of-main temp data, as it
implies that we'd need to update refcounts everytime we add something
back into BMain (an 'un-refcount' ID usages when removing from BMain).

Now that we have two separate flags/tags for those two different things,
let's not merge them anymore.

Note that this is somewhat on-going process, still needs more checks and
cleanup. Related to T88555.
2021-05-26 17:05:00 +02:00
Jeroen Bakker
8f9599d17e DrawManager: Use Compute Shader to Update Hair.
This patch will use compute shaders to create the VBO for hair.
The previous implementation uses tranform feedback.

Timings master (transform feedback with GPU_USAGE_STATIC between 0.000069s and 0.000362s
Timings transform feedback with GPU_USAGE_DEVICE_ONLY. between 0.000057s and 0.000122s
Timings compute shader between 0.000032 and 0.000092s

Future improvements:
* Generate hair Index buffer using compute shaders: currently done single threaded on CPU, easy to add as compute shader.

Reviewed By: fclem

Differential Revision: https://developer.blender.org/D11057
2021-05-26 17:03:37 +02:00
Jeroen Bakker
87055dc71b GPU: Compute Pipeline.
With the compute pipeline calculation can be offloaded to the GPU.
This patch only adds the framework for compute. So no changes for users at
this moment.

NOTE: As this is an OpenGL4.3 feature it must always have a fallback.

Use `GPU_compute_shader_support` to check if compute pipeline can be used.
Check `gpu_shader_compute*` test cases for usage.

This patch also adds support for shader storage buffer objects and device only
vertex/index buffers.

An alternative that had been discussed was adding this to the `GPUBatch`, this
was eventually not chosen as it would lead to more code when used as part of a
shading group. The idea is that we add an `eDRWCommandType` in the near
future.

Reviewed By: fclem

Differential Revision: https://developer.blender.org/D10913
2021-05-26 16:49:30 +02:00
e459a25e6c GPencil: Add option to disable masks in view layer
This patch adds an option in the Layers > Relations panel called "Disable Masks in Render".
When checked, no masks on this layer are included in the render.

Example:
| {F10087680} | {F10087681} |

See T88202 for why this is needed.

Reviewed By: antoniov

Maniphest Tasks: T88202

Differential Revision: https://developer.blender.org/D11234
2021-05-26 16:46:20 +02:00
a6f3bb36df Merge branch 'blender-v2.93-release' 2021-05-26 16:34:03 +02:00
d5a5575685 Fix: GPencil mask shows in view layer render
Currently when rendering the view layer of a grease pencil layer that has
a mask layer attached, the mask layer would show in the rendered image.
This is inconsistent with the default behaviour with no mask on the
grease pencil layer, because it would only render what's on that
particular layer and not anything from any other layer.

This patch makes the masks invisible in the render.

Note: This might seem like not the best solution, but because masks are
just regular grease pencil layers, it's tricky to pass this edge-case to the
drawing code. The way it is handled right now is the best I could come
up with, without making changes that could affect something else.

Reviewed By: antoniov

Maniphest Tasks: T88202

Differential Revision: https://developer.blender.org/D11403
2021-05-26 16:32:58 +02:00
3f27efd31e Merge branch 'blender-v2.93-release' 2021-05-26 16:20:24 +02:00
ba5b4d1bd6 Fix T88250: crash when instancing object in disabled collection
This issue was that `BKE_object_eval_uber_data` was not called for
the text object, because its geometry was not dependent upon
and its `is_directly_visible` tag was `false`. The crash happens in
rendering code, because the evaluated data is missing.

This not only affects text objects, but all object types that have a
geometry component that geometry nodes does not support yet.

The solution is to just add the missing dependencies.

Differential Revision: https://developer.blender.org/D11385
2021-05-26 16:06:01 +02:00
06f86dd4d9 GPencil: Bake GPencil object transforms into a new GPencil object
This operator is a common request of animators to convert the transformation (inluding modifiers) of one grease pencil object, into a new object, generating strokes.

Reviewed By: pepeland

Maniphest Tasks: T87424

Differential Revision: https://developer.blender.org/D11014
2021-05-26 15:43:40 +02:00
b67423f806 Nodes: fix threading issues with node ui storage
Calling BKE_nodetree_attribute_hint_add from multiple threads still
was not safe before..
One issue was that context_map embedded its values directly. So
when context_map grows, all NodeUIStorage would move as well.
I could patch around that by using std::unique_ptr in a few places,
but that just becomes too complex for now.
Instead I simplified the locking a bit by adding just locking a mutex
in NodeTreeUIStorage all the time while an attribute hint is added.

Differential Revision: https://developer.blender.org/D11399
2021-05-26 14:19:01 +02:00
afec66c024 Fix T88588: crash when muting node with multi input socket
The bug existed before the new evaluator already, but the new evaluator
is more sensitive to this kind of error.
2021-05-26 12:25:48 +02:00
12a06292af Cycles: optimize attributes device updates
When an `AttributeSet` is tagged as modified, which happens after the addition or
removal of an `Attribute` from the set, during the following GeometryManager device
update, we update and repack the kernel data for all attribute types. However, if we
only add or remove a `float` attribute, `float2` or `float3` attributes should not
be repacked for efficiency.

This patch adds some mechanisms to detect which attribute types are modified from
the AttributeSet.

Firstly, this adds an `AttrKernelDataType` to map the data type of the Attribute to
the one used in the kernel as there is no one to one match between the two since e.g.
`Transform` or `float4` data are stored as `float3s` in the kernel.

Then, this replaces the `AttributeSet.modified` boolean with a set of flags to detect
which types have been modified. There is no specific flag type (e.g.
`enum ModifiedType`), rather the flags used derive simply from the
`AttrKernelDataType` enumeration, to keep things synchronized.

The logic to remove an `Attribute` from the `AttributeSet` and tag the latter as modified
is centralized in a new `AttributeSet.remove` method taking an iterator as input.

Lastly, as some attributes like standard normals are not stored in the various
kernel attribute arrays (`DeviceScene::attribute_*`), the modified flags are only
set if the associated standard corresponds to an attribute which will be stored
in the kernel's attribute arrays. This makes it so adding or removing such attributes
does not trigger an unnecessary update of other type-related attributes.

Reviewed By: brecht

Differential Revision: https://developer.blender.org/D11373
2021-05-26 12:18:28 +02:00
2a09634d41 Fix particlesystem not duplicating their pointcache in NO_MAIN case.
Sharing data between duplicated IDs should be restricted to depsgraph
(CoW) cases, not all NO_MAIN ones...

While this was probably not an issue currently, we aim at using more and
more out-of-main IDs for temp data processing.

NOTE: Somewhat related to T88555, and similar issue as the one fixed in
rBdfb963c70df5.
2021-05-26 11:25:51 +02:00
534fcab994 Fix T88552: Cycles changing Render Passes in viewport does not work 2021-05-26 11:16:47 +02:00
a72a580948 Cleanup: Simplify Cycles viewport parameters
Use early output and access shading RNA object only once.
2021-05-26 11:10:56 +02:00
e6c0e6c2a9 Compositor: Use BLI_color in convert alpha node.
Recently the CPP colors module landed in master. This patch will use the
new module in the convert alpha node.
2021-05-26 10:46:46 +02:00
1e6b028580 Cleanup: Remove unused argument in Cycles sync
Makes it easier to see where exactly the viewport is used.
2021-05-26 10:45:27 +02:00
09e77f904d Fix T88534: Unable to add a Geometry Node Tree on Volume object
Volumes are supported, poll corrected.

Maniphest Tasks: T88534

Differential Revision: https://developer.blender.org/D11378
2021-05-26 10:28:16 +02:00
03c0fa1cdb Fix T88531: Mantaflow problem with geometry nodes
Objects modified by geometry nodes modifiers were not caught as being
"dynamic".

Now add this modifier type to the list of modifiers making them "dynamic"
in the eyes of mantaflow.

(noticed by @sebbas in chat)

Maniphest Tasks: T88531

Differential Revision: https://developer.blender.org/D11389
2021-05-26 10:21:39 +02:00
b813007a0f Fix T88566: Mantaflow inflow with shapekeys is not working anymore
(regression)

Code was actually checking for shapekeys, but these were not detected
properly (some effects like shape keys are added as virtual modifiers
before the user created modifiers)

Now go over virtual modifiers as well.

Maniphest Tasks: T88566

Differential Revision: https://developer.blender.org/D11388
2021-05-26 10:17:55 +02:00
ff35332610 Merge branch 'blender-v2.93-release' 2021-05-26 09:55:47 +02:00
ebde6e1852 Fix T88251: "Operator Cheat Sheet" Crash
Caused by {rB919558854d62}.

Same fix as in {rBdc8a43c8755a} -- let RNA enum item callbacks check
for NULL context.

The NULL context is used to extract items for document generation.

Maniphest Tasks: T88251

Differential Revision: https://developer.blender.org/D11391
2021-05-26 09:51:55 +02:00
c0bb7d9cb7 Cleanup: Fix short comparison with bool warning
For some reason the hide status is stored in a short and a char
(we cannot have bools in DNA).
2021-05-25 14:37:58 -04:00
95690dd362 Bump FFmpeg version from 4.2.3 to 4.4
Bump FFmpeg version to 4.4 to fix a problem where it would write the
wrong frame rate. Their old API was deprecated and Blender moved to the
new one in rB8d6264ea12bfac0912c7249f00af2ac8e3409ed1. The new one
produced files with the wrong frame rate, which was fixed in FFmpeg 4.4.

Manifest Task: T88568

Reviewed By: LazyDodo, zeddb

Differential Revision: https://developer.blender.org/D11392
2021-05-25 18:58:28 +02:00
490dd279cc deps: Fix broken boost link 2021-05-25 10:27:48 -06:00
45b28c0f88 GPencil: Add a use_light option when creating object.
This option is default off when creating line art objects
because line art seldom use lighting and the normal data
would be all over the place anyway.

Reviewed By: Antonio Vazquez (antoniov)

Differential Revision: https://developer.blender.org/D11372
2021-05-25 23:45:47 +08:00
9f60188cd8 Cleanup: Use ListBase in various places in line art.
This clarifies the data structures for storing edges
for different calculation stages.

Reviewed By: Sebastian Parborg (zeddb)

Differential Revision: https://developer.blender.org/D11386
2021-05-25 23:32:04 +08:00
Jeroen Bakker
cb8a6814fd Blenlib: Explicit Colors.
Colors are often thought of as being 4 values that make up that can make any color.
But that is of course too limited. In C we didn’t spend time to annotate what we meant
when using colors.

Recently `BLI_color.hh` was made to facilitate color structures in CPP. CPP has possibilities to
enforce annotating structures during compilation and can adds conversions between them using
function overloading and explicit constructors.

The storage structs can hold 4 channels (r, g, b and a).

Usage:

Convert a theme byte color to a linearrgb premultiplied.
```
ColorTheme4b theme_color;
ColorSceneLinear4f<eAlpha::Premultiplied> linearrgb_color =
    BLI_color_convert_to_scene_linear(theme_color).premultiply_alpha();
```

The API is structured to make most use of inlining. Most notable are space
conversions done via `BLI_color_convert_to*` functions.

- Conversions between spaces (theme <=> scene linear) should always be done by
  invoking the `BLI_color_convert_to*` methods.
- Encoding colors (compressing to store colors inside a less precision storage)
  should be done by invoking the `encode` and `decode` methods.
- Changing alpha association should be done by invoking `premultiply_alpha` or
  `unpremultiply_alpha` methods.

# Encoding.

Color encoding is used to store colors with less precision as in using `uint8_t` in
stead of `float`. This encoding is supported for `eSpace::SceneLinear`.
To make this clear to the developer the `eSpace::SceneLinearByteEncoded`
space is added.

# Precision

Colors can be stored using `uint8_t` or `float` colors. The conversion
between the two precisions are available as methods. (`to_4b` and
`to_4f`).

# Alpha conversion

Alpha conversion is only supported in SceneLinear space.

Extending:
- This file can be extended with `ColorHex/Hsl/Hsv` for different representations
  of rgb based colors. `ColorHsl4f<eSpace::SceneLinear, eAlpha::Premultiplied>`
- Add non RGB spaces/storages ColorXyz.

Reviewed By: JacquesLucke, brecht

Differential Revision: https://developer.blender.org/D10978
2021-05-25 17:16:54 +02:00
00955cd31e Revert "Blenlib: Explicit Colors."
This reverts commit fd94e03344.
does not compile against latest master.
2021-05-25 17:03:54 +02:00
Jeroen Bakker
fd94e03344 Blenlib: Explicit Colors.
Colors are often thought of as being 4 values that make up that can make any color.
But that is of course too limited. In C we didn’t spend time to annotate what we meant
when using colors.

Recently `BLI_color.hh` was made to facilitate color structures in CPP. CPP has possibilities to
enforce annotating structures during compilation and can adds conversions between them using
function overloading and explicit constructors.

The storage structs can hold 4 channels (r, g, b and a).

Usage:

Convert a theme byte color to a linearrgb premultiplied.
```
ColorTheme4b theme_color;
ColorSceneLinear4f<eAlpha::Premultiplied> linearrgb_color =
    BLI_color_convert_to_scene_linear(theme_color).premultiply_alpha();
```

The API is structured to make most use of inlining. Most notable are space
conversions done via `BLI_color_convert_to*` functions.

- Conversions between spaces (theme <=> scene linear) should always be done by
  invoking the `BLI_color_convert_to*` methods.
- Encoding colors (compressing to store colors inside a less precision storage)
  should be done by invoking the `encode` and `decode` methods.
- Changing alpha association should be done by invoking `premultiply_alpha` or
  `unpremultiply_alpha` methods.

# Encoding.

Color encoding is used to store colors with less precision as in using `uint8_t` in
stead of `float`. This encoding is supported for `eSpace::SceneLinear`.
To make this clear to the developer the `eSpace::SceneLinearByteEncoded`
space is added.

# Precision

Colors can be stored using `uint8_t` or `float` colors. The conversion
between the two precisions are available as methods. (`to_4b` and
`to_4f`).

# Alpha conversion

Alpha conversion is only supported in SceneLinear space.

Extending:
- This file can be extended with `ColorHex/Hsl/Hsv` for different representations
  of rgb based colors. `ColorHsl4f<eSpace::SceneLinear, eAlpha::Premultiplied>`
- Add non RGB spaces/storages ColorXyz.

Reviewed By: JacquesLucke, brecht

Differential Revision: https://developer.blender.org/D10978
2021-05-25 17:01:26 +02:00
b046bc536b Fix T88096: Baking with OptiX and displacement fails
Using displacement runs the shader eval kernel, but since OptiX modules are not loaded when
baking is active, those were not available and therefore failed to launch. This fixes that by falling
back to the CUDA kernels.
2021-05-25 16:56:16 +02:00
8cd506639a Geometry Nodes: Add Shader Curve Nodes
Convert curve vec and curve rgb shader nodes to geometry nodes, based on node_shader_valToRgb.cc implementation.
2021-05-25 15:43:51 +01:00
54ae7baa4c Cleanup: Line art naming changes.
Make variable naming consistent with struct names.

Reviewed By: Sebastian Parborg (zeddb)

Differential Revision: https://developer.blender.org/D11382
2021-05-25 22:11:06 +08:00
c3c576c8c4 Cleanup: Convert to static type directly from CPPType 2021-05-25 09:52:08 -04:00
a20ef4207d Fix T86956: VSE shading mode ignores Grease Pencil Vertex colors.
Issue is that due to the strange definition of render in grease pencil
(meaning should be rendered similar to rendering). This included normal
viewport rendering in OB_RENDER and OpenGL render in OB_RENDER.

For other rendering modes the overlay vertex opacity would be used. This
patch sets this value to 1 when rendering via a scene strip override.

NOTE: that this isn't a good solution as I expect that users want to use
the opacity of the Grease pencil object. Perhaps the GPencil team has a
better solution for it.
2021-05-25 15:18:06 +02:00
1bdd5becc7 Unreported fix: vertex colors overlay not set for new 3d views.
Found during researching {T86956}.
2021-05-25 15:18:06 +02:00
5fddb5ab62 Merge branch 'blender-v2.93-release' 2021-05-25 15:09:54 +02:00
0f7becece7 Fix T88551: Crash accessing uninitialized tool settings
Tool settings for sequencer were not initialized, which caused crash
when adding strips.

There was fix for same issue in versioning rB0f81dafe6cec, but
subversion was not bumped, so files with uninitialized tool settings
may still exist.

Add `SEQ_tool_settings_get()` accessor function that will initialize
tool settings if they are missing. Change operator code to use
`SEQ_tool_settings_fit_method_get()` function instead of accessing
tool settings directly

Reviewed By: campbellbarton

Differential Revision: https://developer.blender.org/D11383
2021-05-25 15:08:32 +02:00
cccfcca33d Cleanup: use nullptr 2021-05-25 13:37:50 +02:00
067587e329 Cleanup: remove unused function 2021-05-25 13:12:54 +02:00
405534c756 Cleanup: remove unnecessary lambda capture 2021-05-25 13:10:44 +02:00
768d4c6cfe Fix T88549: ID sorting tests.
Forgot to initialize the ID types array...

Weird though that this only failed on Windows!

Thanks a lot to @deadpin for helping investigating this.
2021-05-25 12:16:24 +02:00
0a7bd3b6d2 Fix T88464: Incorrect baking with camera markers.
This will update active camera based on the maker for each frame.

Reviewed By: Sebastian Parborg (zeddb), Antonio Vazquez (antoniov)

Differential Revision: https://developer.blender.org/D11358
2021-05-25 16:51:26 +08:00
e3faef686d Fix/Refactor RNA ID preview getter creating preview data.
Same as with forcefields, accessors should never generate data.
2021-05-25 10:38:56 +02:00
ad447705c0 Cleanup: spelling 2021-05-25 18:25:55 +10:00
8c7766dc03 Cleanup: clang-format 2021-05-25 18:25:44 +10:00
Pratik Borhade
0606a37e1a Fix T88500: Constrain window top position (Win32)
Do not allow a window to be created that has a top position that
can obscure all or part of title bar. Right and Left edges can
still be specified slightly outside of monitor bounds, but top
edge must be clamped to monitor top.

see D11371 for more details.

https://developer.blender.org/D11371

Reviewed by Ray Molenkamp
2021-05-24 11:46:35 -07:00
9c53690b30 Merge branch 'blender-v2.93-release' 2021-05-24 12:30:24 -03:00
1a69d491e5 Fix T88478: Fallback to dolly not working when moving the camera in camera view
This fallback is an old hack.

It is difficult to have an orientation convention when several random
factors determine which one should be used.

In this case, to "fix" the problem, a new behavior had to be implemented.
Now the redo when moving the camera in `Camera View` has the default
orientation as `View`.
2021-05-24 12:29:37 -03:00
2265104f0b Fix (unreported): Automatic transform contraint not being disabled
It does not make sense to maintain automatic constraint selection when
you manually set the constraint.
2021-05-24 11:39:50 -03:00
197fa644ee Cleanup: Rename transform enum items
Make them more descriptive.
2021-05-24 11:35:33 -03:00
1318cb26fe Cleanup: Refactor PlaneTrack and PlaneDistort operations
Deduplicates code by introducing a PlaneDirtortBaseOperation for common logic.

Reviewed By: #compositing, jbakker

Differential Revision: https://developer.blender.org/D11273
2021-05-24 14:31:55 +02:00
Charlie Jolly
6c2c16b5f8 Nodes: move shader curves node to C++
Prepare node for conversion to Geometry Nodes.

There should be no functional changes.

Reviewed By: JacquesLucke, LazyDodo

Differential Revision: https://developer.blender.org/D11226
2021-05-24 12:02:49 +01:00
24004e74fa Fix T88524: GPencil PDF does not take into account the marker camera
The camera was not checked before doing the export.
2021-05-24 12:36:13 +02:00
43153e2324 Fix T88466: Sound strips prevent strip rendering
When sound strip is above another strip such as movie strip, it prevents
from rendering movie strip.

This bug was introduced in 0b7744f4da. Function `must_render_strip()`
checks if there is any strip with `SEQ_BLEND_REPLACE` blending and
considers this strip as lowest strip in stack. Sound strips do have this
blend mode set, which caused the bug.

Remove all sound strips and muted strips from stack collection before
checking with `must_render_strip()` function
2021-05-24 05:57:53 +02:00
7e841c797f UI: Use title case for labels 2021-05-23 15:31:54 -04:00
9dcbc195ad Docs: Add readme for mikktspace 2021-05-23 13:57:22 -04:00
f7b22fc3d2 Cleanup: remove redundant property for transfer_mode key-map item
As the property isn't saved and defaults to disabled,
there is no need to set it.
2021-05-22 14:53:41 +10:00
7adbe62a09 Cleanup: tweaks for Object Non-modal keymap
Simplify adding non-legacy keymap items.
2021-05-22 13:14:28 +10:00
2175cbe2ce Cleanup: Move curve draw cache implementation to C++
I'd like to use this file to draw curves from geometry nodes, which
would otherwise require implementing a C API. The changes in this
commit are minimal, mostly just casts and changing to nullptr.

Differential Revision: https://developer.blender.org/D11350
2021-05-21 17:28:29 -04:00
2521e17a58 Merge branch 'blender-v2.93-release' 2021-05-21 14:44:17 -04:00
d5c3bff6e7 Fix T87357: Missing update after removing socket
The CoW copy of the node group was not updated correctly after it
changed. Arguably, tagging the node tree could also be part of
`ntreeUpdateTree` (which needs to be called after changes to the
node tree anyway). However, there are many instances where the
depsgraph is tagged explicitly after `ntreeUpdateTree` is called,
so it should be fine here as well. This is similar to what is done
in `snode_dag_update`.

Differential Revision: https://developer.blender.org/D11342
2021-05-21 14:43:26 -04:00
Wayde Moss
9c4ecef62f Fix T88375: Bone Size Small After V3D.View_All
The wrong matrix function was used and overwrote the custom bone shape
scale instead of reading from it.

Reviewed By: sybren

Differential Revision: https://developer.blender.org/D11330
2021-05-21 13:24:55 -04:00
4fb052f08d Fix T88384: Improved Win32 Window Sizing and Positioning
When creating Win32 windows, the sizes and placements can be out by a
small amount, mostly noticeable near monitor edges. This is because
Windows 10 includes a thin invisible border (typically 7 pixels) when
determining position. Therefore the correct values can sometimes be
just outside the monitor bounds, but we clamp them at those bounds.

This patch fixes this by first clamping the requested values to monitor
bounds, adjusting for window chrome with AdjustWindowRectEx(), and then
using those adjusted values in CreateWindowExW().

see D11314 for more details.

Differential Revision: https://developer.blender.org/D11314

Reviewed by Ray Molenkamp
2021-05-21 09:34:58 -07:00
24deeee09e Fix some RNA physics accessors creating data.
Accessing data through RNA should never implicitely create it. Objects'
and particles' forcefields and collision settings were doing this.

Note that UI code also had to be tweaked to properly handle `None`
(NULL) cases.

Differential Revision: https://developer.blender.org/D11341
2021-05-21 17:22:43 +02:00
adf2f146d8 Fix particle ID init not creating particle deflect data.
This data (the force fields) are expected to always be there, but they
are currently created on the fly by RNA accessors (typically from UI
draw code), which is extremely wrong way to do it.

Differential Revision: https://developer.blender.org/D11341
2021-05-21 17:22:43 +02:00
305b08b521 GPencil: Speed up Occlude Eraser
This is an initial change to speed up the calculation of the Occlude eraser. In the future, we can add more optimizations, but at least this increase speed.

Intead to check always the 3 points, the check is skipped if it's not required.

Base in a solution by Philipp Oeser.

This is related to T88412
2021-05-21 17:00:33 +02:00
5158d1b42d Cleanup: use nullptr 2021-05-21 15:09:22 +02:00
Leon Zandman
865d1889da Cleanup: spelling
Includes fixes to misspelled function names.

Ref D11280
2021-05-21 22:23:07 +10:00
cf42586737 Fix T85752: Collection Instance Crash when instancing collections with disabled subcollections
Root of the issue was actually hidden deep in depsgraph itself: it would
not properly update all of its COW IDs using a datablock when depsgraph
decides to evaluate or un-evaluate it.

This would lead to evaluated IDs pointing to either:
 - orig IDs when there was an evaluated version of those (annoying bug,
   but not a crashing one).
 - old address of previously evaluated IDs that no longer exists in the
   depsgraph (causing the crash from the report e.g.).

This commit adds an extra step at the end of nodes building, that goes
over all of already existing IDs in the depsgraph to check whether they
do one of the two things above, and tag them for COW update if so.

NOTE: This only affects depsgraph (re-)building, not its evaluation.
This remains consistent with the fact that operations that may change
the depsgraph content (like Collection exclusion etc.) need to trigger a
rebuild.

NOTE: Performances: Worst case scenarii, like (un-)excluding a whole
character collection in a production file, lead to 5% to 10% extra
processing time in depsgraph building. Most of it comming from extra COW
processing (in depsgraph's update in `build_step_finalize`), the detection
loop itself only accounts for 1% to 2% of the whole building time.

Maniphest Tasks: T85752

Differential Revision: https://developer.blender.org/D10907
2021-05-21 12:43:32 +02:00
ff51c2e89a Cleanup: Use named unused arguments in Cycles Device 2021-05-21 11:19:33 +02:00
c154b078b5 Geometry Nodes: new Material Replace node
This node can change all faces that use a specific material to use a
different material. Using this node is significantly more efficient
than creating a selection from all faces with a specific material
index and then using the Material Assign node.

Ref T88055.

Differential Revision: https://developer.blender.org/D11325
2021-05-21 09:43:55 +02:00
aa6f0f3d1f Depsgraph: remove mesh edit-mode pointer duplication
Share the pointer with the original mesh instead, this matches behavior
of all other objects edit-mode data.

Duplicating the edit-mesh pointer makes updates to edit-mesh require
a COPY_ON_WRITE update, which is currently an expensive operation
(copying the entire mesh).

Notes:

- This change is from 802027f3f8
  so the edit-meshes object pointer `BMEditMesh.ob` referenced the COW
  version of the object. This pointer has since been removed, so the
  copy is no longer needed.

- Having a separate edit-mesh pointer could be used so linked duplicates
  could have their own generated meshes. For this to be supported,
  many other changes would be needed: see D10920.
2021-05-21 16:23:42 +10:00
12a91f7f5d Fix T88227: Eevee not working on AMD 535 cards.
Enabled HQ normals workaround for this specific configuration.
2021-05-21 07:46:20 +02:00
2b640622ff Transform: use ID_RECALC_GEOMETRY flag when updating object data
While this doesn't provide any noticeable benefits at the moment,
it allows for geometry updates skipping copy-on-write in edit-mode
in the future.
2021-05-21 15:45:25 +10:00
b13953b1f2 Cleanup: quiet -Warray-parameter warnings from GCC11
Some warnings remain that require larger changes.
2021-05-21 15:45:25 +10:00
6c6551c398 Cleanup: conversion warning 2021-05-21 15:45:23 +10:00
9bbee7dd92 Fix T88227: Eevee not working on AMD 535 cards.
Enabled HQ normals workaround for this specific configuration.
2021-05-21 07:44:25 +02:00
Iyad Ahmed
a43c7538b8 Eevee Wavelength Node Support
This patch adds wavelength node support to Eevee, similar to how
Eevee Blackbody node works, thus it is a little off from Cycles.

Reviewed By: #eevee_viewport, fclem, brecht
Differential Revision: https://developer.blender.org/D11326
2021-05-21 10:25:47 +05:30
b18d5dac53 Transform: remove ID_RECALC_SELECT for edit-mode armatures
This was added in 9516921c05
so overlays would redraw, as far as I can see it's no longer needed.

Reviewed By: fclem

Ref D11322
2021-05-21 10:58:07 +10:00
ba9cf91736 Cleanup: spelling 2021-05-21 01:06:20 +02:00
695fa3a4a1 Merge branch 'blender-v2.93-release' 2021-05-20 20:48:44 +02:00
6a0906c09a Fix T87854: Add clamp option to Path Animation
Previously, the "follow path constraint" and "follow parented curve"
were clamped.  This restriction was lifted in rBcf2baa585cc8

Add back an option to get the old behavior in the "Path animation" settings.

Reviewed By: Sybren

Differential Revision: http://developer.blender.org/D11263
2021-05-20 20:41:10 +02:00
1400fdf558 Fix petty comment.
I'm embarrassed this was in our codebase for 18 years
2021-05-20 12:38:01 -06:00
ef04fd9571 GPencil: Fix missing annotations in VSE strip
Since version 2.80, the annotations of the Scene strip were not displayed in VSE. Also, the UI panel was`Grease Pencil` and must be `Annotation`

The problem was the offscreen render hasn't evil_CTX and the section of the annotation was never called.

Differential Revision: https://developer.blender.org/D11329
2021-05-20 20:16:59 +02:00
677909cdc3 Merge branch 'blender-v2.93-release' into master 2021-05-20 21:58:34 +05:30
2e7d3ad7b5 macOS build: set minimum deployment target correctly
Make `CMAKE_OSX_DEPLOYMENT_TARGET` independent of buildbot settings and
always set to `OSX_MIN_DEPLOYMENT_TARGET`. That fixes the launch error
on OS older than buildbot's.
Remove unused `MACOSX_DEPLOYMENT_TARGET`.

Fix T88419
Diff D11323
2021-05-20 21:55:58 +05:30
9fef934530 Merge branch 'blender-v2.93-release' 2021-05-20 12:20:05 -04:00
991a1cdf5d Fix T87316: No undo push for clicking on attribute name
Since there is a special callback for assigning the name from the
search box, and the callback doesn't call an operator or something
else that would do an undo push, I think the solution is to do it
manually here.

Another option would be adding the button flag "UI_BUT_UNDO", which
isn't set by default for search buttons, but that gives us ugly names for
the undo steps, so an explicit push is better in this case.

Differential Revision: https://developer.blender.org/D11190
2021-05-20 12:17:54 -04:00
cd16123761 GPencil: Adding length modifier.
Reviewed By: Antonio Vazquez (antoniov)

Differential Revision: https://developer.blender.org/D8264
2021-05-20 23:35:53 +08:00
ee5bfde9e6 Fix T87947: Trasnform: Keyboard input uses view orientation
When activated in modal, `translate`, `resize`, `rotate`, `shear` and
`edge_rotate_normal` use a different orientation than the set in scene.

This orientation needed to match since some of these modes can be switched
during operation.

The default orientation for these modes was `V3D_ORIENT_VIEW`.

And this changed when finishing the `translate` and `resize` to
`V3D_ORIENT_GLOBAL`.

But this could cause inconsistencies when inputting values from the
keyboard.

The solution now is to change the orientation when you change the mode.

---
Note: Although the user can expect the value entered to reflect the
orientation set in the scene, it would require a lot of changes and would
not be really useful.
2021-05-20 11:05:44 -03:00
47f4f3c932 Cleanup: Use enum to indicate the current orientation
Improves readability.
2021-05-20 11:05:28 -03:00
4bb2a5b5cb Fix T88345: Blender crash on GPUFrameBuffer creation
Misuse of indexes.

`color_attachements` has only color ones while `config` has color and depth.
2021-05-20 11:04:04 -03:00
92178c7e0e Fix T88365: GPUTexture.read returning a buffer with wrong size
The pixel components were not being considered.
2021-05-20 11:03:47 -03:00
729c579030 Fix T88313: GPencil reproject operator projecting on the wrong surface
`ray_start` must start at the position of the gpencil point.
2021-05-20 11:03:24 -03:00
933999db75 Cleanup: rename incorrectly named experimental feature flag
The `WITH_GEOMETRY_NODES` flag does not make sense anymore,
it is just protecting the `Simulation` data block that is not used
currently.
2021-05-20 14:39:30 +02:00
0745afeddb Merge remote-tracking branch 'origin/blender-v2.93-release' 2021-05-20 13:00:07 +02:00
dreamertooth
a1954e3807 Fix T86193: Bake Action, wrong rotation order for bones
Use bone rotation order to compute the baked rotation. This fixes a bug
introduced in rB0e85d701c654, where the object rotation order was
applied to the bone.

Maniphest Tasks: T88359, T86193

Reviewed By: sybren, GuiltyGhost, #animation_rigging

Differential Revision: https://developer.blender.org/D11282
2021-05-20 12:57:05 +02:00
3e3ecc329c Geometry Nodes: new Material input node
This node is similar to the Value and Vector node.
It just provides a way to use the same material in multiple nodes
without exposing it outside of a node group.

Differential Revision: https://developer.blender.org/D11305
2021-05-20 12:29:06 +02:00
f41a753e75 Geometry Nodes: new Material Assign node
This adds a new Material Assign node. It can be used to change the
material used by an existing mesh or to assign a material to a mesh
that has been generated from scratch.

Differential Revision: https://developer.blender.org/D11155
2021-05-20 12:22:13 +02:00
f3a0267016 Materials: support materials when applying modifier
This fixes the `Apply Modifier` and `Visual Geometry to Mesh` operator
when a modifier changed materials on the evaluated geometry.

This is necessary since rB1a81d268a19f2f1402f408ad1dadf92c7a399607.

Differential Revision: https://developer.blender.org/D11303
2021-05-20 12:15:57 +02:00
b51bd859fd Fix: wrong rna pointer for material slot
This was missing from rB1a81d268a19f2f1402f408ad1dadf92c7a399607.
2021-05-20 12:14:31 +02:00
b084b57fbf Geometry Nodes: new geometry nodes evaluator
The old geometry nodes evaluator was quite basic and missed many features.
It was useful to get the geometry nodes project started. However, nowadays
we run into its limitations from time to time.

The new evaluator is more complex, but comes with new capabilities.
The two most important capabilities are that it can now execute nodes in
parallel and it supports lazy evaluation.

The performance improvement by multi-threading depends a lot on the specific
node tree. In our demo files, the speedup is measurable but not huge. This
is mainly because they are bottlenecked by one or two nodes that have to be
executed one after the other (often the Boolean or Attribute Proximity nodes)
or because the bottleneck is multi-threaded already (often openvdb nodes).

Lazy evaluation of inputs is only supported by the Switch node for now.
Previously, geometry nodes would always compute both inputs and then just
discard the one that is not used. Now, only the input that is required
is computed.

For some more details read D11191, T87620 and the in-code documentation.

Differential Revision: https://developer.blender.org/D11191
2021-05-20 11:35:13 +02:00
44e7192745 Cleanup: spelling 2021-05-20 17:55:35 +10:00
a5ac062c51 Cleanup: inconsistent parameter names 2021-05-20 09:00:42 +02:00
7655cc45eb VSE: Fix select strips at current frame feature
This feature of `select_side_of_frame` was disabled by removing option
from operator property enum but functional code was never removed.

Add back option to use this feature.

Feature was disabled due to keymap issue. Currently this feature doesn't
have keymap assigned.
2021-05-20 05:06:32 +02:00
acba8f617b Cleanup: Remove unused define
References to `SEQ_CACHE_COST_MAX` were removed in 38b77ef8b2.
2021-05-20 04:25:56 +02:00
47e88345a1 VSE: Fix animation duplication in split operator
Due to misunderstanding of how strip duplication works, animation data
was duplicated on all strips when any strip was split.

`SEQ_sequence_base_dupli_recursive()` duplicated data on strip that was
being split, and `SEQ_ensure_unique_name()` duplicated animation on all
strips.

Only duplication should be done with `SEQ_ensure_unique_name()` and only
on right side split strips, because only these strips are duplicated.
2021-05-20 01:27:12 +02:00
d1c9a99c07 Splines: Optimize interpolation in special case virtual array
When the input data is a virtual array for a single value, we don't need
to run any of the interpolation, instead just copy the input data.
2021-05-19 17:17:16 -04:00
0b7744f4da VSE: Fix rendering inconsistency
Fix issue described in T87678, which was partially a bug and partially
change in intended(at least as far as I can tell) behaior.

Function `evaluate_seq_frame_gen` that was partially responsible for
filtering strips in stack for rendering wasn't working correctly.
Intended functionality seems to be removing all effect inputs from stack
as it is unlikely that user would want these to be blended in. However
there was logic to exclude effects placed into same input, which because
of weak implementation caused, that any effect input, that is effect as
well will be considered to be part of stack to be blended in.
This bug was apparently used to produce effects like glow over original
image.

Even though this is originally unintended, I have kept this logic, but
I have made it explicit.

Another change is request made in T87678 to make it possible to keep
effect inputs as part of stack when they are placed above the effect,
which would imply that blending is intended. This change is again
explicitly defined.

Whole implementation has been refactored, so logic is consolidated
and code should be as explicit as possible and more readable.
`must_render_strip function` may be still quite hard to read, not sure
if I can make it nicer.

Last change is for remove gaps feature code - it used same rendering
code, which may be reason why its logic was split in first place.
Now it uses sequencer iterator, which will definitely be faster than
original code, but I could have used `LISTBASE_FOREACH` in this case.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D11301
2021-05-19 22:59:33 +02:00
97cf2a9fb1 VSE: Refactor sorting functions
Recently `SEQ_sort()` function was split so functionality is provided
on per-seqbase basis. After discussion about this split, it turned out,
that per-seqbase operation is only that should be provided, because
RNA API functions need to be able to access arbitrary seqbase

Remove recently introduced function `seq_sort_seqbase` and change
`SEQ_sort` function to operate on seqbase.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D11297
2021-05-19 22:52:57 +02:00
58893eaef8 Cleanup: Use consistent order for functions in node files
After this commit, all geometry node "init" and "update" functions are
at the top of each file, right below the "layout" function. This means
you can always scroll to the bottom of the file to see the entry point,
and the boring boilerplate code is grouped in one section.
2021-05-19 16:13:05 -04:00
3d604ba867 Geometry Nodes: Execute point rotate node on curve component
When adding attributes on the curve component, I missed calling this
node's execute function on the curve component, like the other nodes.
2021-05-19 15:58:34 -04:00
627f357127 Geometry Nodes: Support for dynamic attributes on curve splines
With this patch you will be able to add and remove attributes from curve
data inside of geometry nodes. The following is currently implemented:
* Adding attributes with any data type to splines or spline points.
* Support for working with multiple splines at the same time.
* Interaction with the three builtin point attributes.
* Resampling attributes in the resample node.

The following is not implemented in this patch:
* Joining attributes when joining splines with the join geometry node.
* Domain interpolation between spline and point domains.
* More efficient ways to call attribute operations once per spline.

Differential Revision: https://developer.blender.org/D11251
2021-05-19 13:22:09 -04:00
192a3f1a05 Cleanup: Use copy constructor for CurveEval
There is no need for a special "copy" method with a copy constructor,
which will be necessary to explicitly copy attributes anyway.
2021-05-19 13:02:53 -04:00
0456223cde Fix T87793: Cycles OptiX crash hiding objects in viewport render 2021-05-19 18:30:43 +02:00
3e472d87a8 Cycles OpenCL: disable AO preview kernels
These seem to be causing some stability issues, and really are just not that
useful in practice. Compiling them is slow already, so it does not improve
the user experience much to show an AO preview if it's not nearly instant.
2021-05-19 18:30:43 +02:00
Severin
330fecc9b7 Assets: Use textures (and shadows) for automatic object previews
Textures may be important to be able to identify an object. They are also a way
to make many objects look more like when rendered with an advanced render
engine, without being that expensive.
So this seems like a simple way to increase usefulness of the automatic
previews.
2021-05-19 17:57:51 +02:00
857f39a3d7 GPencil: Apply NanoSVG fix
This fix is ported from official NanoSVG git
2021-05-19 17:56:26 +02:00
23b642cbb4 IDManagement: Enable ID sorting test for linked IDs. 2021-05-19 17:50:22 +02:00
ca74ebc96e IDManagement: Do not prevent sorting of linked IDs in BKE_id_new_name_validate.
While this function should (currently) not be called on linked ID, there
is no reason to treat those differently than local IDs, for the part
that they have in common: needs to be properly sorted.
2021-05-19 17:50:22 +02:00
3620dbbe97 IDManagement: Fix ID sorting in case of linked IDs.
`id_sort_by_name` would simply not deal properly with linked IDs, could
lead to mixing IDs from different libraries, and unsorted IDs within the
same library.
2021-05-19 17:50:22 +02:00
c810672ea7 IDManagement: Add basic test regarding id sorting in their listbase.
Note that while sorting of local IDs is fine, currently sorting of
linked IDs is a total unpredictable failure.

Next commit will fix this, ensuring that linked IDs are grouped by their
library, and properly name-sorted within this library group.
2021-05-19 17:50:22 +02:00
ee5ad46a0e Fix T87621: Win32 Do Not Create Preview Thumbnails for Offline Files
This patch turns off the creation of file thumbnails for files that are
offline and therefore not fully-present on the file system. These types
of files - typically cloud-based or stored on slower backup media -
only have their contents available when actually accessed, at which
point there will be a short delay. If we allow thumbnail creation in
this state then all offline files in a folder will be downloaded just
to view a listing, which can take a long time.

Files in this state will instead get a more generic thumbnail that
still indicates file type (icon in center) and that shows offline state
will a special icon at the bottom-left.

Although this currently only affects Windows users, most of this patch
is platform-agnostic. So other platforms inherit this behavior if they
only add FILE_ATTR_OFFLINE attribute to files in this state.

See D11101 for more information.

Differential Revision: https://developer.blender.org/D11101

Reviewed by Julian Eisel
2021-05-19 08:40:00 -07:00
41b912925c Blender LTS: match download urls with latest changes in buildbot. 2021-05-19 16:41:51 +02:00
0fa4286ade Blender LTS: match download urls with latest changes in buildbot. 2021-05-19 16:39:22 +02:00
38450fc916 Merge branch 'blender-v2.93-release' 2021-05-19 13:58:16 +02:00
Pratik Borhade
9e6c4be731 Fix T88167: Regression: no tooltip for syringe/picker, during picking object
Fix T88167.

Caused by {rB97defd9cd79b6e3ed0e52481a7078107dbe0522b}

`(BLI_rcti_isect_pt` used here to confirm if cursor position is in between active region boundary.
Subtracting min region boundary from the mouse position before the check, fails the condition.

`mval[2]` introduced to hold the region relative mouse position.

Reviewed By: Severin

Maniphest Tasks: T88167

Differential Revision: https://developer.blender.org/D11224
2021-05-19 13:47:40 +02:00
afb66b3c3a Cleanup: use nullptr 2021-05-19 13:04:12 +02:00
c5a898bcac Merge branch 'blender-v2.93-release' 2021-05-19 12:32:06 +02:00
d67223ca29 EEVEE: Subsurface Node: Fix inverted texture blur input
Fixes T86097 EEVEE Subsurface Node give blurry edges
2021-05-19 12:27:38 +02:00
81e3d904db Merge branch 'blender-v2.93-release' 2021-05-19 12:17:23 +02:00
Mikhail Matrosov
a294670bac EEVEE: Geometry info -> tangent: Make consistent with Cycles for hair
Changes output for geometry info node in Eevee to be consistent with Cycles (w/o osl)

Before this patch Eevee outputs Z-tangent even for hair. This patch changes it to output hair tangent (growth direction). Hair tangent is impossible to derive otherwise from normal or view direction.

Reviewed By: fclem, brecht
Differential Revision: https://developer.blender.org/D10841
2021-05-19 12:14:07 +02:00
7adb415070 Geometry Nodes: don't show grease pencil materials for now
Those materials cannot be on geometry handled by geometry nodes yet.
So it makes sense not to show those materials in the material socket.
2021-05-19 11:44:10 +02:00
5e6f3b8564 Geometry Nodes: take materials into account when joining geometries
Materials are now kept intact when using the Join Geometry node
or when realizing instaces.
2021-05-19 11:02:25 +02:00
02b80276b3 Fix issue in previous commit
When `PointerRNA->data` was null, it was interpreted as being
`None` in Python. This caused the materials slots to not show
correctly in the ui.
2021-05-19 10:51:36 +02:00
1a81d268a1 Materials: support changing materials during evaluation
This commit allows that the evaluated geometry of an object has
different materials from the original geometry. This is needed
for geometry nodes.

The main thing that changes for render engines and exporters
is that the number of material slots on an object and its geometry
might not match anymore. For original data, the slot counts are
still equal, but not for evaluated data.

Accessing material slots though rna stays the same. The behavior
adapts automatically depending on whether the object is evaluated.

When accessing materials of an object through `BKE_object_material_*`
one has to use a new api for evaluated objects:
`BKE_object_material_get_eval` and `BKE_object_material_count_eval`.
In the future, the different behavior might be hidden behind a more
general C api, but that would require quite a few more changes.

The ground truth for the number of materials is the number of materials
on the geometry now. This is important in the current design, because
Eevee needs to know the number of materials just based on the mesh in
`mesh_render_mat_len_get` and similar places.

In a few places I had to add a special case for mesh edit mode to get it
to work properly. This is unfortunate, but I don't see a way around that
for now.

Differential Revision: https://developer.blender.org/D11236
2021-05-19 10:23:09 +02:00
5239198f17 Fix T88362: Lineart countour rendering error
This is caused by lacking of consideration of
OB_DUPLICOLLECTION in updateDepsgraph() in the
modifier.

Reviewed by: Antonio Vazquez (antoniov)

https://developer.blender.org/D11292
2021-05-19 15:09:40 +08:00
d373b43f07 VSE: Better handling of effect strip splitting
Splitting of effect strip alone wasn't handled properly. Previously
this resulted in duplicating effect strip, and it was broken at least
from 2.79.

Change in rB8ec6b34b8eb2 was intended to allow splitting strips
individually, so it can be used as RNA API function but also so it
requires as little glue logic as possible.

This is fixed by splitting all dependent strips at once in 2 separate
ListBases for left and right strips. Strips can be finally moved into
original `ListBase`.

With this fix it is still possible to split strips individually with
little glue logic. RNA API function could return list of split strips
as well, currently at least one strip in chain will be provided so
chain can be reconstructed on python side.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D10209
2021-05-18 23:41:47 +02:00
f7a14c116c VSE: Move whole strip chain to meta
Python API function Sequence.move_to_meta() did delete effect chain
when strip with effects was moved.

Use iterator API to query effect strips and move whole chain to meta.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D11206
2021-05-18 23:41:47 +02:00
748b5f025d Merge branch 'blender-v2.93-release' 2021-05-18 22:37:38 +02:00
8dd43ac23e Fluid: Updated Mantaflow source files
Includes update for OpenVDB file IO, i.e. fixes an issue with
compression flag combination that resulted in random segfaults.

Other changes: Cleanup and formatting.
2021-05-18 22:36:51 +02:00
5400be9ffe Fix T88313: GPencil reproject operator projecting on the wrong surface
`ray_start` must start at the position of the gpencil point.
2021-05-18 17:31:10 -03:00
f9567f6c63 Alembic: read/write generated coordinates of meshes
Read and write generated coordinates (also known as "original
coordinates", "reference coordinates", or "orcos") from and to Alembic.
A custom geometry property named "Pref" is used for (hopefully)
interoperability with Maya and Houdini. For now it's only guaranteed for
Blender-to-Blender.

Export: writing generated coordinates is optional (on by default).

Import: generated coordinates are always read whenever the reading of
vertex data is enabled.

Manifest Task: T88081
2021-05-18 19:01:57 +02:00
a881b5272b Fix ID copying outside of Main affecting 'directly linked' status of other IDs.
Copying an ID outside of BMain should not assume that the datablocks it
uses are now directly linked. This would be wrong e.g. in case that new
no-main ID is copied from a linked data-block and is supposed to remain
a linked data.
2021-05-18 18:44:39 +02:00
3826c161ad Cleanup: typo in comment. 2021-05-18 18:44:39 +02:00
019ab774d4 Merge branch 'blender-v2.93-release' 2021-05-18 16:17:29 +02:00
1b909c726b Fix build warning
Value of the size argument in `strncat` is too large, might lead to a
buffer overflow.

Change the argument to be the free space in the destination buffer
minus the terminating null byte.

Introduced in 5368859a66

Reviewed By: Sergey

Differential Revision: https://developer.blender.org/D11289
2021-05-18 16:15:08 +02:00
7aa7cc4ca1 Merge branch 'blender-v2.93-release' 2021-05-18 15:43:06 +02:00
a9da73ab95 Fix T88370: GPencil render crash when use viewlayer and masking
When filtering the render using view layer, could be possible the mask is NULL and just need to be ignored.
2021-05-18 15:42:26 +02:00
2b660e825b Fix T88365: GPUTexture.read returning a buffer with wrong size
The pixel components were not being considered.
2021-05-18 10:01:29 -03:00
b158477551 Fix ocean modifier giving different result on Arm than x86
And re-enable the test on macOS Arm.

Ref T78710
2021-05-18 14:40:22 +02:00
ea7b00c64f Fix T88345: Blender crash on GPUFrameBuffer creation
Misuse of indexes.

`color_attachements` has only color ones while `config` has color and depth.
2021-05-18 09:39:31 -03:00
4402c3006b WM: check missing space-data & constraints in poll functions
Without this, menu search prints many errors in some contexts.
2021-05-18 20:57:40 +10:00
eaf3160f13 Merge branch 'blender-v2.93-release' 2021-05-18 12:52:29 +02:00
53376ec7fc Tests: Disabled physics ocean test on macOS arm64
Test is failing and needs further investigation.
2021-05-18 12:48:35 +02:00
f4eae747f9 Tests: Remove threshold param from physics ocean mesh test
This was not the reason for the test failure on macOS arm64.
2021-05-18 12:44:53 +02:00
1f9eb7e2c6 Merge branch 'blender-v2.93-release' 2021-05-18 19:59:20 +10:00
aa06be9148 Fix error calling select-camera without a 3D view 2021-05-18 19:58:11 +10:00
97f1e4782a Cycles: Avoid unnecessary data updates in viewport
The BlenderSync will do quite a bit of work on every sync_data() call
even if there is nothing changed in the scene. There will be early
outputs done deeper in the call graph, but this is not really enough to
ensure best performance during viewport navigation.

This change makes it so sync_data() is only used when dependency graph
has any update tags: if something changed in the scene the dependency
graph will know it. If nothing changed there will be no IDs tagged for an
update in the dependency graph.

There are two weak parts in the current change:

- With the persistent data there is a special call to ignore the check
  of the dependency graph tags. This is more of a safety, because it is
  not immediately clear what the correct state of recalc flags is.

- Deletion of objects is detected indirectly, via tags of scene and
  collections.

It might not be bad for the first version of the change.

The test file used: {F10117322}

Simply open the file, start viewport render, and navigate the viewport.
On my computer this avoids 0.2sec spend on data_sync() on every
up[date of viewport navigation.

We can do way more granular updates in the future: for example, avoid
heavy objects sync when it is only camera object which changed. This
will need an extended support from the dependency graph API. Doing
nothing if nothing is changed is something we would want to do anyway.

Differential Revision: https://developer.blender.org/D11279
2021-05-18 10:13:09 +02:00
1c1fa15ce1 Merge branch 'blender-v2.93-release' 2021-05-18 17:48:14 +10:00
b236653f8a Fix error in grease pencil flip color operator
- Used try/except instead of a poll function.
- The error case referenced a non-existent error handling module.

Prefer poll functions over exception handling where possible,
also having an operators logic in a try block isn't good practice
as it can hide more serious errors in the code.

Note that duplicate pencil settings access should be moved into a
utility function. This can be part of a separate cleanup.
2021-05-18 17:38:54 +10:00
8a26fec6ea Cleanup: use const casts in mathutils API 2021-05-18 14:16:34 +10:00
4dfbaa34c9 PyAPI: allow any vector sequence for poly_3d_calc(..)
Previously only vectors were accepted for the second point argument:
mathutils.interpolate.poly_3d_calc()
2021-05-18 14:16:34 +10:00
5e67dcf3d2 Cleanup: format 2021-05-18 06:09:46 +02:00
52e53d61f5 Merge branch 'blender-v2.93-release' 2021-05-18 13:39:39 +10:00
dfb963c70d Fix T88026: Repeated switch to rendered viewport crashes
Resolve ownership ambiguity with shared physics pointers.

Previously, LIB_ID_CREATE_NO_MAIN allowed pointer sharing with
the source ID so physics caches can be shared between original and
evaluated data: (Object.soft.shared & Object.rigidbody_object.shared).

This only worked properly for LIB_TAG_COPIED_ON_WRITE ID's,
as LIB_TAG_NO_MAIN can be used in situations where the original ID's
lifetime limited by it's original data.

This commit adds `LIB_ID_COPY_SET_COPIED_ON_WRITE` so ID's only share
memory with original data for ID's evaluated in the depsgraph.

For all other uses, a full copy of physics data is made.

Ref D11228#287094
2021-05-18 13:35:00 +10:00
cdc1ddf20b Cleanup: clang-format 2021-05-18 12:53:27 +10:00
8057b98572 Alembic procedural: fix wrong property used as base for attribute
lookups

We use the schema so that we can access top level attributes as well.
This is already done for polygon meshes and curves, so this only
modifies the behavior for subdivision objects.
2021-05-18 00:59:29 +02:00
38fe8d8b38 Alembic procedural: fix crash accessing invalid ICompoundProperties
Although it is not a pointer, accessing an ICompoundProperty
dereferences a pointer under the hood, so check for validity.
2021-05-18 00:56:45 +02:00
73080ddf38 Add comment regarding Fade Inactive Geometry property in overlays popover
Reviewed By: campbellbarton

Differential Revision: https://developer.blender.org/D11249
2021-05-18 00:51:35 +02:00
f96a0f384d Merge branch 'blender-v2.93-release' 2021-05-18 00:32:48 +02:00
ea11b4e10c Fix T88230: Properly handle Face Set visibility in Expand Face Sets editing
Expand is not expected to update the visibility state of the PBVH, only
the Face Sets IDs. If visibility updates are made accidentally, PBVH
rendering breaks.

In order for this to work properly, the following fixes are needed:

- Expand should always check for active component before attempting to
 modify a Face Set ID

- Expand should always check the visibility state on original_face_sets, as
it is the array that contains the visiblilty state that corresponds with the
current state used for PBVH rendering. This implies that after any modification
done by Expand, the visibility state of ss->face_sets and
expand_cache->original_face_sets should match (like in any other tool that
does not modify visibility).

- Expand should never modify the Face Set ID of a poly that is hidden in
expand_cache->original_face_sets.

- When deleting an ID, hidden Face Sets should be skipped when picking IDs for
content filling. This avoids introducing hidden IDs back into the visible
geometry even after updating its visibility state.

Reviewed By: JulienKaspar, JacquesLucke

Maniphest Tasks: T88230

Differential Revision: https://developer.blender.org/D11243
2021-05-18 00:30:24 +02:00
c17433bb55 Tests: Increased threshold param for physics_ocean test
Adjusting threshold to see what the buildbot will make of this.
2021-05-17 23:21:16 +02:00
c9afea5bbf Tests: Added threshold to physics ocean mesh test
The physics_ocean test is currently failing on macOS arm.
This is one attempt to fix this issue.
2021-05-17 21:13:35 +02:00
43046d82b7 Collada import: use black for Base Color when missing <diffuse>
Treat a missing <diffuse> the same as a black diffuse color.

The easiest way to see this bug is with a Collada shader like

```
          <constant>
            <emission>
              <color sid="emission">1 0 0 1</color>
            </emission>
          </constant>
```

The Collada spec says this should be just

```
color = <emission>
```

ie. red everywhere. The importer slots the red into the Principled Emission socket, but since it leaves the Base Color as the default off-white, this is added to red, and the material looks white-pink in the light and red only in the shadows.

Putting black in the Base Color makes it look red everywhere.

D10939 will also eliminate the much-less-noticeable specular term for this case.

Reviewed By: gaiaclary

Differential Revision: https://developer.blender.org/D10941
2021-05-17 21:10:05 +02:00
542b8da831 Merge branch 'blender-v2.93-release' 2021-05-17 20:18:39 +02:00
1b0ecb2f89 refactor: minor changes to previous commit
- rename boolean is_zero by more descriptive has_specularity
- add some clarifying comments (and TODO)
2021-05-17 20:04:56 +02:00
Scurest
1a4e7b16b2 Collada import: respect zero-specularity
Collada shaders with black <specular> should import with Specular=0.
(A missing <specular> is the same as black.)

The general specular conversion is hard, but this case is common and easy.
Fixes the specular for all <constant>/<lambert> shaders, and <blinn>/<phong>
shaders with black/omitted <specular>. Before this they all looked too "shiny".

Reviewed By: gaiaclary

Differential Revision: https://developer.blender.org/D10939
2021-05-17 19:56:01 +02:00
85be72c1cc Fix T86278: vertex color baking not working with modifiers
As in the old Blender Internal baking code, this still relies on there being a
good mapping to the original vertices.
2021-05-17 19:41:11 +02:00
912f2b1a29 Fix T85892: disable progressive refine when using adaptive sampling
This is giving too bright pixel values, as the sample scaling and random number
sample are wrong. The proper fix for this is complicated. It will be solved in
Cycles X, for now we disable this combination.
2021-05-17 19:41:11 +02:00
e48cdf3d0d Fix T88216: Cycles persistent data fails with animated object transform in instance 2021-05-17 19:41:11 +02:00
88e884bfc6 Fix T88210: Cycles persistent data failing with image sequences 2021-05-17 19:41:11 +02:00
91a5dbbd17 Fix OpenCL group size performance issue on Intel GPUs
Contributed by Intel. On some scenes like classroom with particular integrated
GPUs this speeds up rendering 1.97x. With other benchmarks and GPUs it's
between 0.99-1.14x.
2021-05-17 19:40:57 +02:00
a86e815dd8 Mathutils: add a Matrix.LocRotScale constructor for combining channels.
Combining location, rotation and scale channels into a matrix is
a standard task, so while it is easily accomplished by constructing
and multiplying 3 matrices, having a standard utility allows for
more clear code.

The new constructor builds a 4x4 matrix from separate location,
rotation and scale values. Rotation can be represented as a 3x3
Matrix, Quaternion or Euler value, while the other two inputs
are vectors. Unneeded inputs can be replaced with None.

Differential Revision: https://developer.blender.org/D11264
2021-05-17 19:12:40 +03:00
f09606cc68 File List: micro optimization in the icon preview
No real functional changes.

When `i` is zero, `filelist_cache_previews_push` was called twice with
the same icon.

This caused the preview to be computed twice when only once is needed.
2021-05-17 09:09:12 -03:00
6cfa8531cc Cleanup: reduce indentation in filelist_cache_previews_push 2021-05-17 08:54:53 -03:00
50eceb2a37 Merge branch 'blender-v2.93-release' 2021-05-17 13:06:59 +02:00
bb32ecadb5 GPencil: Fix unreported error exporting to PDF/SVG with animated camera
Before, the camera parameters were calculated only for first frame. 

If the camera is animated, these values need to be recalculated in order to get the new camera view position and export the strokes as expected.

Also fixed the export of PDF when the view is not in camera view. PDF export, needs to be done in camera view.
2021-05-17 13:05:38 +02:00
Wannes Malfait
ba698f0812 Geometry Nodes: update callbacks with 2 new socket types 2021-05-17 11:11:25 +02:00
7e552b356d Merge branch 'blender-v2.93-release' 2021-05-17 18:04:22 +10:00
e3a76feeef Fix PyAPI doc generation error in 43369ca80e
Files without doc-strings were not included.
2021-05-17 18:02:56 +10:00
Henrik Dick
e95f71bf15 UI: use non-linear sliders
* Boolean Modifier > Fast > Overlap Threshold (Logarithmic).
* Remesh Modifier > Voxel > Voxel Size (Logarithmic).
* Sculpt > Dyntopo > Detail Size (Cubic).

Ref D9074
2021-05-17 17:33:59 +10:00
Henrik Dick
0447aedb96 UI: add non-linear slider support
This patch introduces non linear sliders. That means, that the movement
of the mouse doesn't map linearly to the value of the slider.

The following changes have been made.

- Free logarithmic sliders with maximum range of (`0 <= x < inf`)
- Logarithmic sliders with correct value indication bar.
- Free cubic sliders with maximum range of (`-inf < x < inf`)
- Cubic sliders with correct value indication bar.

Cubic mapping has been added as well, because it's used for brush sizes
in other applications (Krita for e.g.).

To make a slider have a different scale type use following line in RNA:
 `RNA_def_property_ui_scale_type(prop, PROP_SCALE_LOGARITHMIC);`
or:
 `RNA_def_property_ui_scale_type(prop, PROP_SCALE_CUBIC);`

Test the precision, step size and soft-min if you change the scale type
of a property as it will feel very different and may need tweaking.

Ref D9074
2021-05-17 17:33:55 +10:00
45863059f7 Cleanup: remove commented code, use function to access header region 2021-05-17 15:55:00 +10:00
67c40cc51c Cleanup: missing declaration warning 2021-05-17 15:22:19 +10:00
27c0d29138 Cleanup: clang-format 2021-05-17 15:12:41 +10:00
6a771175b6 Cleanup: Fix missing-braces warning on Linux / GCC 2021-05-17 15:05:14 +10:00
0e581f6196 Cleanup: Fix missing-braces warning on macOS Clang 2021-05-16 11:19:21 +05:30
49cb30bb0c Cleanup: Fix inconsistent-missing-override warning
macOS Clang
2021-05-16 11:19:04 +05:30
9dabb342ba Cleanup: Improve comments 2021-05-15 18:00:20 -05:00
4da16a0707 Cleanup: Use helper function
Use the `CurveEval` function introduced in the last commit.
2021-05-15 17:50:29 -05:00
038c6e229c Splines: Convenience methods for point offsets of multiple splines
Since spline data is stored separately for each spline, the data often
needs to be flattened into a separate array. It's helpful to have the
necessarily-sequential part of that split off into a separate method.
I've found myself using functions like these in quite a few places.
2021-05-15 17:44:33 -05:00
cb12fb78ca XR Controller Support Step 1: Internal Abstractions for OpenXR Actions
Adds internal API for creating and managing OpenXR actions at the
GHOST and WM layers. Does not bring about any changes for users since
XR action functionality is not yet exposed in the Python API (will be
added in a subsequent patch).

OpenXR actions are a means to communicate with XR input devices and
can be used to retrieve button/pose states or apply haptic feedback.
Actions are bound to device inputs via a semantic path binding
(https://www.khronos.org/registry/OpenXR/specs/1.0/html/xrspec.html#semantic-path-interaction-profiles),
which serves as an XR version of keymaps.

Main features:

- Abstraction of OpenXR action management functions to GHOST-XR,
  WM-XR APIs.
- New "xr_session_start_pre" callback for creating actions at
  appropriate point in the XR session.
- Creation of name-identifiable action sets/actions.
- Binding of actions to controller inputs.
- Acquisition of controller button states.
- Acquisition of controller poses.
- Application of controller haptic feedback.
- Carefully designed error handling and useful error reporting
  (e.g. action set/action name included in error message).

Reviewed By: Julian Eisel

Differential Revision: http://developer.blender.org/D10942
2021-05-16 03:36:31 +09:00
3e6609a0dc Cleanup: silence unused parameter warnings
Introduced in rB48fa029dd11b.
2021-05-15 14:13:22 -03:00
a56cc26b48 Merge branch 'blender-v2.93-release' 2021-05-15 18:34:28 +02:00
610c0ecc3b Fix T88204 EEVEE: Animated world light doesnt work during render
This was a bug uncovered by rB50782df42586.

Previously, the lightcache was always discarded between redraw and forced
to be updated again.

Now we check for update inside the render loop making it compatible with
accumulation motion blur and long exposure.
2021-05-15 18:29:14 +02:00
c2ee7ab2e3 Merge branch 'blender-v2.93-release' 2021-05-15 17:52:26 +02:00
5d97e293c3 EEVEE: Depth of Field: Fix tile artifacts in mixed focus regions
This was caused by the slight focus gather not being wide enough
for small radii. Now the cast to int will properly round the radius to
the nearest integer.

This is related to T86244 Black Artefacts in EEVEE on Transparent BSDF
2021-05-15 17:52:09 +02:00
45d3303ae4 Fix T85845 EEVEE: Depth Of Field: Artifacts with slight out of focus
This was caused by the bokeh LUT being sampled outside the valid range.
But `texelFetch` is only valid if the sample actually exists. This lead to
undefined behavior.

The fix is to increase `DOF_MAX_SLIGHT_FOCUS_RADIUS` (which just offsets the
LUT along the X=Y axis) to avoid any sample outside the defined range.
2021-05-15 17:52:09 +02:00
6e8685e1fd GPencil: Add parameter removed by error in previous commit
This fix the compiler warning too.
2021-05-15 15:53:58 +02:00
65997a058f Merge branch 'blender-v2.93-release' 2021-05-15 02:55:25 +02:00
f432b5ccf5 RigidBodies: Clarify usage of the custom density in the calculate mass
operator

This is only used when the preset is set to 'Custom', make that clear in
the tooltip and disable editing the value if another preset is used.

Issue came up in T88155.

Maniphest Tasks: T88155

Differential Revision: https://developer.blender.org/D11210
2021-05-15 02:53:42 +02:00
ccb22740c1 Merge branch 'blender-v2.93-release' 2021-05-15 02:34:47 +02:00
4707c86179 Fix T88188: Keyframing Vertex Mass in Cloth Simulation doesn't have any
effect

The way it is now, `ClothVertex->mass` is setup once for the clothObject
and then reused if it already exists [so does not change across frames/
steps] which happens in`cloth_from_object` / `do_init_cloth`, where
`SIM_cloth_solver_init` / `SIM_mass_spring_set_vertex_mass` are called
and set up masses in `Implicit_Data`.

Seems possible to update `ClothVertex->mass` every step in
`do_step_cloth`, however it seems more involved to update the masses in
`Implicit_Data` there as well. The masses from `Implicit_Data` are
accessed in many places, so it would be mandatory to have these masses
kept up-to-date (and even then it is unclear if the solver was designed
to work with these animated or if there are assumptions about this being
stable across the sim).

So propose to remove the (broken/not implemented) animation capabilities
on the property instead.

Maniphest Tasks: T88188

Differential Revision: https://developer.blender.org/D11225
2021-05-15 02:32:43 +02:00
3458d09a74 Merge branch 'blender-v2.93-release' 2021-05-15 02:06:02 +02:00
baf7f78f9f Fix T87715: Eevee: Holdout options not available in Outliner Collection context menu
Collection holdouts are now supported by eevee.

Maniphest Tasks: T87715

Differential Revision: https://developer.blender.org/D11233
2021-05-15 01:55:30 +02:00
e7e183aa2b Cleanup: Update / improve comments 2021-05-14 16:37:05 -05:00
3c978a73d1 Docs: PyAPI: Replace use of the bgl module in the RenderEngine example
The bgl module will be discontinued.

This example is seen at:
https://docs.blender.org/api/current/bpy.types.RenderEngine.html uses it.

Differential Revision: https://developer.blender.org/D11262
2021-05-14 18:29:23 -03:00
782ccbde70 Cleanup: Move attribute code to attribute header
This code in the geometry set header was not directly related to
geometry sets, it makes more sense in the attribute access header.
This makes it clearer that code for geometry components uses attribute
code, rather than the other way around. It also allows adding more
functionality to `BKE_attribute_access.hh` that depends on these things
without including `BKE_geometry_set.hh` there.
2021-05-14 15:53:43 -05:00
db59f0b943 Fix T88262: Curve to mesh crash with vector last segment
The code incorrectly used the size of the second to last segment rather
than the last segment's size. That was a problem when the last segment
is a vector segment but the second to last isn't.

I also used the opportunity to slightly refactor the control point
offsets cache, making it one longer so it also contains information
about the size of the last segment, simplifying other code.
2021-05-14 11:26:42 -05:00
7178e54122 Merge branch 'blender-v2.93-release' 2021-05-14 18:18:24 +02:00
d4a36c7ad5 Fix T88234: Crash with Eevee when shader displacement socket connection is muted
In {rB266cd7bb82ce}, support for muting links was added. It might be
debatable if we define a shader as "having" displacement even if the link
is muted, but after said commit, shader_has_displacement() would return
true but still leave the returned node as NULL.

Now also return false if the link is muted (otherwise the caller would
need to additionally check the returned node as well.)

Maniphest Tasks: T88234

Differential Revision: https://developer.blender.org/D11256
2021-05-14 18:16:35 +02:00
7efe92f63b Merge branch 'blender-v2.93-release'
Conflicts:
	source/blender/editors/object/object_add.c
2021-05-14 18:10:12 +02:00
f6cb9433d4 GPencil: Refactor Append operators
Now the operators work like other areas of Blender using the list of selected objects.

Also, the name has been changed to:

```Layers:
- Copy Layer to Selected
- Copy All Layers to Selected

Materials:
- Copy Material to Selected
- Copy All Materials to Selected```

Reviewed By: mendio, pablovazquez, pepeland

Differential Revision: https://developer.blender.org/D11216
2021-05-14 18:02:04 +02:00
8c80267352 CLOG: Fix memleak in own recent improvement to CLOG filter.
Mistake in rBef5782e29744.
2021-05-14 17:50:55 +02:00
Germano Cavalcante
b11499939c Python GPU: Replace part of the code that uses 'bgl' with 'gpu'
This is part of the process described in T80730.

The aim is to deprecate the bgl module.

Reviewed By: fclem, brecht, campbellbarton

Revision: D11147
2021-05-14 12:26:40 -03:00
Germano Cavalcante
48fa029dd1 Python GPU: New 'platform' module
This module exposes the platform utils defined in the GPU module in C.

This will be useful for porting existing code with `bgl` to `gpu`.

Reviewed By: fclem, brecht, campbellbarton

Maniphest Tasks: T80730

Part of D11147
2021-05-14 12:26:40 -03:00
Germano Cavalcante
3f4f109646 Python GPU: New 'capabilities' module
This module exposes the capabilities defined in the GPU module in C.

This will be useful for porting existing code in `bgl` to `gpu`.

Reviewed By: fclem, brecht, campbellbarton

Maniphest Tasks: T80730

Part of D11147
2021-05-14 12:26:40 -03:00
c3e13d5a2b GPencil: fix separate points/strokes freezing with empty selection
Code would still create an object (without setting up materials), code
for removing unused material slots would then freeze.

Now return/cancel early in case of empty selection.

This came up in T88269 [which is still not fully fixed, transforming
curve edit points clear their GP_STROKE_SELECT flag which now results in
the early exit, should be looked at separately]

Maniphest Tasks: T88269

Differential Revision: https://developer.blender.org/D11252
2021-05-14 17:16:05 +02:00
079f415deb Cleanup: use enum types for screen direction variables
The term direction was used in 3 different ways in screen editing code,
making it hard to follow:

- 0-3 for as magic numbers mapped to [west,north,east,south].
- `h`, `v` characters for [horizontal,vertical] axes.
- Cycle direction SPACE_CONTEXT_CYCLE_PREV, SPACE_CONTEXT_CYCLE_NEXT

The following changes have been made:

- Add `eScreenDir` for [west,north,east,south], use variable name `dir`.
- Add `eScreenAxis` for [horizontal,vertical] values, use variable name
  `dir_axis`.
- Add `eScreenCycle` for existing enum `SPACE_CONTEXT_CYCLE_{PREV/NEXT}`.
- Add macros `SCREEN_DIR_IS_VERTICAL(dir)`,
  `SCREEN_DIR_IS_HORIZONTAL(dir)`.
  Replacing `ELEM(dir, 1, 3)`, `ELEM(dir, 0, 2)`.
- Move `ED_screen_draw_join_highlight`, `ED_screen_draw_split_preview`
  to `screen_intern.h`.

Reviewed By: Severin

Ref D11245
2021-05-15 00:49:49 +10:00
d19d79c5a6 LineArt: Custom UI for adding GP object.
This allows extra options (in-front and stroke order) to be shown when adding line art kind of grease pencil object.

Reviewed by: Antonio Vazquez (antoniov)

Diff: https://developer.blender.org/D11130
2021-05-14 22:40:47 +08:00
265d97556a PyAPI: use iterators for ID property methods (keys, values & items)
- Matches changes in Python 3.x dictionary methods.

- Iterating now raises a run-time error if the property-group changes
  size during iteration.

- IDPropertyGroup.iteritems() has been removed.

- IDPropertyGroup View & Iterator types have been added.

- Some set functionality from dict_keys/values/items aren't yet
  supported (isdisjoint method and boolean set style operations).

Proposed as part of T85675.
2021-05-15 00:36:49 +10:00
65f9550813 Nodes: fix connecting wrong sockets when inserting node
Differential Revision: https://developer.blender.org/D11231
2021-05-14 16:01:55 +02:00
a4f0780acf Merge branch 'blender-v2.93-release' 2021-05-14 13:45:48 +02:00
Pratik Borhade
16bcf7dc8e Fix T87417: crash when evaluating geo nodes without window manager
Differential Revision: https://developer.blender.org/D11240
2021-05-14 13:43:30 +02:00
bd5bab961e Merge branch 'blender-v2.93-release' 2021-05-14 19:27:30 +10:00
3c09beb3b1 Fix memory leak in IDPropertyGroup.pop()
When popping ID-property groups/arrays,
ID-property was removed but not freed.

Now the value is converted to a native Python type and freed.
2021-05-14 19:25:29 +10:00
2871fadcad Geometry Nodes: use texture socket in Attribute Sample Texture node
There is a new Texture data-block socket that we can use in Geometry
Nodes now. This commit replaces the texture property of a node and
gives it a texture input socket instead. That increases flexibility.

The texture socket still has some limitations that will be lifted in the
next couple of days (e.g. it's not supported by the switch node and
cannot be exposed the a modifier yet).

Differential Revision: https://developer.blender.org/D11222
2021-05-14 11:23:25 +02:00
a939317862 Cleanup: sort include dirs in CMakeLists.txt
Sort include dirs in just one file, not a big change
(`source/blender/editors/armature/CMakeLists.txt`).

No functional changes.
2021-05-14 11:12:35 +02:00
26b4ef6823 Geometry Nodes: remove some unnecessary updates
This fixes a few "obvious" places where we do unnecessary updates.
Those special cases were added in the early days of geometry nodes
when many updates were missing and we tried to get it to work at all.
There is a fairly high risk that with this change some required updates
will be missing again. Those can be fixed when we find thme.
Some of the update issues might have been fixed by rB58818cba40794905f9323080e60884e090f2d388
and similar changes we added over time.

Differential Revision: https://developer.blender.org/D11238
2021-05-14 11:06:00 +02:00
Christoph Lendenfeld
9797b95f61 Animation: Pose sliding tools improvements
Improve the "In Betweens" tools:
- Push Pose from Rest Pose
- Relax Pose to Rest Pose
- Push Pose from Breakdown
- Relax Pose to Breakdown
- Pose Breakdowner

These all now use the same new sliding tool:
- Actual visual indication of the blending/pushing percentage applied.
- Mouse wrapping to allow for extrapolation without having to worry
  about the initial placement of the mouse. This also means these tools
  are actually usable when chosen from the menu.
- Precision mode by holding {key Shift}.
- Snapping to 10% increments by holding {key Ctrl}.
- Overshoot protection; by default the tool doesn't allow overshoot
  (lower than 0% or higher than 100%), and it can be enabled by pressing
  {key E}.
- Bones are hidden while sliding, so the pose itself can be seen more
  clearly. This can be toggled by pressing {key H} while using the tool.

Reviewed By: #animation_rigging, zeddb, sybren, #user_interface, brecht, Severin, looch

Maniphest Tasks: T81836

Differential Revision: https://developer.blender.org/D9054
2021-05-14 11:00:17 +02:00
a156843112 Cleanup: spelling, rename metas to meta-strips / meta-balls 2021-05-14 17:38:16 +10:00
c1c0b661c0 Cleanup: clang-format 2021-05-14 17:35:08 +10:00
20c3db6604 DrawTest: Make setup method public.
This allows other tests to override it.
2021-05-14 08:34:13 +02:00
55b9b1ff50 Draw: Put DrawTest in its own compile unit.
DrawTest will be used by other tests as well.
2021-05-14 07:55:01 +02:00
bd76184966 Fix sculpt neighbor iterator not taking visibility into account
Sculpting tools are designed to ignore hidden geometry and behave like
hidden geometry does not exist.
When getting the neighbors of a vertex, now this takes into account
hidden geometry to avoid returing neighbors which connected edge is not
visible. This should make corner cases of a lot of tools work properly,
especially when working in low poly meshes when is common to have a
single face loop hidden.

Reviewed By: JacquesLucke

Differential Revision: https://developer.blender.org/D11007
2021-05-14 03:08:01 +02:00
01718ad952 Merge branch 'blender-v2.93-release' 2021-05-14 02:42:19 +02:00
fce795415a Fix T87929: Cycles, missing update when visibility is modified
This issue originates from a missing BVH packing for visibility data
when it is modified.

To fix this, this adds update flags to the managers to carry the
modified visibility information from the Objects' modified flag to the
GeometryManager.

Another set of flags is added to determine which data need to be packed:
geometry, vertices, or visibility. Those flags are then used when
packing the primivites.

Reviewed By: brecht

Maniphest Tasks: T87929

Differential Revision: https://developer.blender.org/D11219
2021-05-14 02:40:43 +02:00
Pablo Dobarro
d70cfb90e0 Fix Sculpt tools showing the cursor of the previous active brush
When using a tool that is not a brush, the previously used brush
preset will still be active in the tool settings, so the cursor will
draw its custom reviews.
This checks if the current active tool is a brush before drawing its
previews. If it is not a brush tools, it draws a default white cursor.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D9418
2021-05-14 00:56:13 +02:00
01a614c699 Fix build after last commit
Part of a rename change in rBc5d38a2be8 was lost when committing.
2021-05-13 17:52:30 -05:00
c5d38a2be8 Functions: Expose set_all method for generic virtual arrays
This is very similar to rB5613c61275fe6 and rB0061150e4c90d, basically
just exposing a `VMutableArray` method to its generic counterpart. This
is quite important for curve point attributes to avoid a lookup for
every point when there are multiple splines.
2021-05-13 17:47:46 -05:00
88aee2986f Cleanup: Clang format 2021-05-13 17:37:51 -05:00
Scurest
2953732fc5 Collada import: connect Emission texture to Emission socket
An emission texture is currently connected to the Base Color socket. It should connect to the Emission socket, like a constant does.

Reviewed By: gaiaclary

Differential Revision: https://developer.blender.org/D10990
2021-05-13 23:43:37 +02:00
d889e9684a Merge branch 'blender-v2.93-release' 2021-05-13 23:07:05 +02:00
37570a7317 Fix T87360 Multi input links aren't placed correctly when created with python
link->multi_input_socket_index, which is used to calculate the links
position on the multi-input socket, was not set.
Now it is set to the sockets current link count.

Review: Jacques Lucke (JacquesLucke)
Differential Revision: https://developer.blender.org/D11082
2021-05-13 23:05:38 +02:00
725 changed files with 24480 additions and 14495 deletions

View File

@@ -29,7 +29,7 @@ set(BLOSC_EXTRA_ARGS
-DCMAKE_POSITION_INDEPENDENT_CODE=ON
)
# Prevent blosc from including it's own local copy of zlib in the object file
# Prevent blosc from including its own local copy of zlib in the object file
# and cause linker errors with everybody else.
set(BLOSC_EXTRA_ARGS ${BLOSC_EXTRA_ARGS}
-DPREFER_EXTERNAL_ZLIB=ON

View File

@@ -43,7 +43,7 @@ set(JPEG_FILE libjpeg-turbo-${JPEG_VERSION}.tar.gz)
set(BOOST_VERSION 1.73.0)
set(BOOST_VERSION_NODOTS 1_73_0)
set(BOOST_VERSION_NODOTS_SHORT 1_73)
set(BOOST_URI https://dl.bintray.com/boostorg/release/${BOOST_VERSION}/source/boost_${BOOST_VERSION_NODOTS}.tar.gz)
set(BOOST_URI https://boostorg.jfrog.io/artifactory/main/release/${BOOST_VERSION}/source/boost_${BOOST_VERSION_NODOTS}.tar.gz)
set(BOOST_HASH 4036cd27ef7548b8d29c30ea10956196)
set(BOOST_HASH_TYPE MD5)
set(BOOST_FILE boost_${BOOST_VERSION_NODOTS}.tar.gz)
@@ -297,10 +297,10 @@ set(OPENJPEG_HASH 63f5a4713ecafc86de51bfad89cc07bb788e9bba24ebbf0c4ca637621aadb6
set(OPENJPEG_HASH_TYPE SHA256)
set(OPENJPEG_FILE openjpeg-v${OPENJPEG_VERSION}.tar.gz)
set(FFMPEG_VERSION 4.2.3)
set(FFMPEG_VERSION 4.4)
set(FFMPEG_URI http://ffmpeg.org/releases/ffmpeg-${FFMPEG_VERSION}.tar.bz2)
set(FFMPEG_HASH 695fad11f3baf27784e24cb0e977b65a)
set(FFMPEG_HASH_TYPE MD5)
set(FFMPEG_HASH 42093549751b582cf0f338a21a3664f52e0a9fbe0d238d3c992005e493607d0e)
set(FFMPEG_HASH_TYPE SHA256)
set(FFMPEG_FILE ffmpeg-${FFMPEG_VERSION}.tar.bz2)
set(FFTW_VERSION 3.3.8)

View File

@@ -37,7 +37,7 @@ if [ $USE_DEBUG_TRAP -ne 0 ]; then
trap 'err_report $LINENO' ERR
fi
# Noisy, show every line that runs with it's line number.
# Noisy, show every line that runs with its line number.
if [ $USE_DEBUG_LOG -ne 0 ]; then
PS4='\e[0;33m$(printf %4d ${LINENO}):\e\033[0m '
set -x
@@ -563,9 +563,9 @@ OIDN_SKIP=false
ISPC_VERSION="1.14.1"
FFMPEG_VERSION="4.2.3"
FFMPEG_VERSION_SHORT="4.2"
FFMPEG_VERSION_MIN="3.0"
FFMPEG_VERSION="4.4"
FFMPEG_VERSION_SHORT="4.4"
FFMPEG_VERSION_MIN="4.4"
FFMPEG_VERSION_MAX="5.0"
FFMPEG_FORCE_BUILD=false
FFMPEG_FORCE_REBUILD=false

View File

@@ -20,7 +20,7 @@
# ILMBASE_LIBRARIES - list of libraries to link against when using IlmBase.
# ILMBASE_FOUND - True if IlmBase was found.
# Other standarnd issue macros
# Other standard issue macros
include(FindPackageHandleStandardArgs)
include(FindPackageMessage)
include(SelectLibraryConfigurations)

View File

@@ -22,7 +22,7 @@
# These are defined by the FindIlmBase module.
# OPENEXR_FOUND - True if OpenEXR was found.
# Other standarnd issue macros
# Other standard issue macros
include(SelectLibraryConfigurations)
include(FindPackageHandleStandardArgs)
include(FindPackageMessage)

View File

@@ -1,70 +1,4 @@
Blender Buildbot
================
Buildbot Configuration
=====================
Code signing
------------
Code signing is done as part of INSTALL target, which makes it possible to sign
files which are aimed into a bundle and coming from a non-signed source (such as
libraries SVN).
This is achieved by specifying `worker_codesign.cmake` as a post-install script
run by CMake. This CMake script simply involves an utility script written in
Python which takes care of an actual signing.
### Configuration
Client configuration doesn't need anything special, other than variable
`SHARED_STORAGE_DIR` pointing to a location which is watched by a server.
This is done in `config_builder.py` file and is stored in Git (which makes it
possible to have almost zero-configuration buildbot machines).
Server configuration requires copying `config_server_template.py` under the
name of `config_server.py` and tweaking values, which are platform-specific.
#### Windows configuration
There are two things which are needed on Windows in order to have code signing
to work:
- `TIMESTAMP_AUTHORITY_URL` which is most likely set http://timestamp.digicert.com
- `CERTIFICATE_FILEPATH` which is a full file path to a PKCS #12 key (.pfx).
## Tips
### Self-signed certificate on Windows
It is easiest to test configuration using self-signed certificate.
The certificate manipulation utilities are coming with Windows SDK.
Unfortunately, they are not added to PATH. Here is an example of how to make
sure they are easily available:
```
set PATH=C:\Program Files (x86)\Windows Kits\10\App Certification Kit;%PATH%
set PATH=C:\Program Files (x86)\Windows Kits\10\bin\10.0.18362.0\x64;%PATH%
```
Generate CA:
```
makecert -r -pe -n "CN=Blender Test CA" -ss CA -sr CurrentUser -a sha256 ^
-cy authority -sky signature -sv BlenderTestCA.pvk BlenderTestCA.cer
```
Import the generated CA:
```
certutil -user -addstore Root BlenderTestCA.cer
```
Create self-signed certificate and pack it into PKCS #12:
```
makecert -pe -n "CN=Blender Test SPC" -a sha256 -cy end ^
-sky signature ^
-ic BlenderTestCA.cer -iv BlenderTestCA.pvk ^
-sv BlenderTestSPC.pvk BlenderTestSPC.cer
pvk2pfx -pvk BlenderTestSPC.pvk -spc BlenderTestSPC.cer -pfx BlenderTestSPC.pfx
```
Files used by Buildbot's `compile-code` step.

View File

@@ -1,127 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import argparse
import os
import re
import subprocess
import sys
def is_tool(name):
"""Check whether `name` is on PATH and marked as executable."""
# from whichcraft import which
from shutil import which
return which(name) is not None
class Builder:
def __init__(self, name, branch, codesign):
self.name = name
self.branch = branch
self.is_release_branch = re.match("^blender-v(.*)-release$", branch) is not None
self.codesign = codesign
# Buildbot runs from build/ directory
self.blender_dir = os.path.abspath(os.path.join('..', 'blender.git'))
self.build_dir = os.path.abspath(os.path.join('..', 'build'))
self.install_dir = os.path.abspath(os.path.join('..', 'install'))
self.upload_dir = os.path.abspath(os.path.join('..', 'install'))
# Detect platform
if name.startswith('mac'):
self.platform = 'mac'
self.command_prefix = []
elif name.startswith('linux'):
self.platform = 'linux'
if is_tool('scl'):
self.command_prefix = ['scl', 'enable', 'devtoolset-9', '--']
else:
self.command_prefix = []
elif name.startswith('win'):
self.platform = 'win'
self.command_prefix = []
else:
raise ValueError('Unkonw platform for builder ' + self.platform)
# Always 64 bit now
self.bits = 64
def create_builder_from_arguments():
parser = argparse.ArgumentParser()
parser.add_argument('builder_name')
parser.add_argument('branch', default='master', nargs='?')
parser.add_argument("--codesign", action="store_true")
args = parser.parse_args()
return Builder(args.builder_name, args.branch, args.codesign)
class VersionInfo:
def __init__(self, builder):
# Get version information
buildinfo_h = os.path.join(builder.build_dir, "source", "creator", "buildinfo.h")
blender_h = os.path.join(builder.blender_dir, "source", "blender", "blenkernel", "BKE_blender_version.h")
version_number = int(self._parse_header_file(blender_h, 'BLENDER_VERSION'))
version_number_patch = int(self._parse_header_file(blender_h, 'BLENDER_VERSION_PATCH'))
version_numbers = (version_number // 100, version_number % 100, version_number_patch)
self.short_version = "%d.%d" % (version_numbers[0], version_numbers[1])
self.version = "%d.%d.%d" % version_numbers
self.version_cycle = self._parse_header_file(blender_h, 'BLENDER_VERSION_CYCLE')
self.hash = self._parse_header_file(buildinfo_h, 'BUILD_HASH')[1:-1]
if self.version_cycle == "release":
# Final release
self.full_version = self.version
self.is_development_build = False
elif self.version_cycle == "rc":
# Release candidate
self.full_version = self.version + self.version_cycle
self.is_development_build = False
else:
# Development build
self.full_version = self.version + '-' + self.hash
self.is_development_build = True
def _parse_header_file(self, filename, define):
import re
regex = re.compile(r"^#\s*define\s+%s\s+(.*)" % define)
with open(filename, "r") as file:
for l in file:
match = regex.match(l)
if match:
return match.group(1)
return None
def call(cmd, env=None, exit_on_error=True):
print(' '.join(cmd))
# Flush to ensure correct order output on Windows.
sys.stdout.flush()
sys.stderr.flush()
retcode = subprocess.call(cmd, env=env)
if exit_on_error and retcode != 0:
sys.exit(retcode)
return retcode

View File

@@ -1,81 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
from dataclasses import dataclass
from pathlib import Path
from typing import List
@dataclass
class AbsoluteAndRelativeFileName:
"""
Helper class which keeps track of absolute file path for a direct access and
corresponding relative path against given base.
The relative part is used to construct a file name within an archive which
contains files which are to be signed or which has been signed already
(depending on whether the archive is addressed to signing server or back
to the buildbot worker).
"""
# Base directory which is where relative_filepath is relative to.
base_dir: Path
# Full absolute path of the corresponding file.
absolute_filepath: Path
# Derived from full file path, contains part of the path which is relative
# to a desired base path.
relative_filepath: Path
def __init__(self, base_dir: Path, filepath: Path):
self.base_dir = base_dir
self.absolute_filepath = filepath.resolve()
self.relative_filepath = self.absolute_filepath.relative_to(
self.base_dir)
@classmethod
def from_path(cls, path: Path) -> 'AbsoluteAndRelativeFileName':
assert path.is_absolute()
assert path.is_file()
base_dir = path.parent
return AbsoluteAndRelativeFileName(base_dir, path)
@classmethod
def recursively_from_directory(cls, base_dir: Path) \
-> List['AbsoluteAndRelativeFileName']:
"""
Create list of AbsoluteAndRelativeFileName for all the files in the
given directory.
NOTE: Result will be pointing to a resolved paths.
"""
assert base_dir.is_absolute()
assert base_dir.is_dir()
base_dir = base_dir.resolve()
result = []
for filename in base_dir.glob('**/*'):
if not filename.is_file():
continue
result.append(AbsoluteAndRelativeFileName(base_dir, filename))
return result

View File

@@ -1,245 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import dataclasses
import json
import os
from pathlib import Path
from typing import Optional
import codesign.util as util
class ArchiveStateError(Exception):
message: str
def __init__(self, message):
self.message = message
super().__init__(self.message)
@dataclasses.dataclass
class ArchiveState:
"""
Additional information (state) of the archive
Includes information like expected file size of the archive file in the case
the archive file is expected to be successfully created.
If the archive can not be created, this state will contain error message
indicating details of error.
"""
# Size in bytes of the corresponding archive.
file_size: Optional[int] = None
# Non-empty value indicates that error has happenned.
error_message: str = ''
def has_error(self) -> bool:
"""
Check whether the archive is at error state
"""
return self.error_message
def serialize_to_string(self) -> str:
payload = dataclasses.asdict(self)
return json.dumps(payload, sort_keys=True, indent=4)
def serialize_to_file(self, filepath: Path) -> None:
string = self.serialize_to_string()
filepath.write_text(string)
@classmethod
def deserialize_from_string(cls, string: str) -> 'ArchiveState':
try:
object_as_dict = json.loads(string)
except json.decoder.JSONDecodeError:
raise ArchiveStateError('Error parsing JSON')
return cls(**object_as_dict)
@classmethod
def deserialize_from_file(cls, filepath: Path):
string = filepath.read_text()
return cls.deserialize_from_string(string)
class ArchiveWithIndicator:
"""
The idea of this class is to wrap around logic which takes care of keeping
track of a name of an archive and synchronization routines between buildbot
worker and signing server.
The synchronization is done based on creating a special file after the
archive file is knowingly ready for access.
"""
# Base directory where the archive is stored (basically, a basename() of
# the absolute archive file name).
#
# For example, 'X:\\TEMP\\'.
base_dir: Path
# Absolute file name of the archive.
#
# For example, 'X:\\TEMP\\FOO.ZIP'.
archive_filepath: Path
# Absolute name of a file which acts as an indication of the fact that the
# archive is ready and is available for access.
#
# This is how synchronization between buildbot worker and signing server is
# done:
# - First, the archive is created under archive_filepath name.
# - Second, the indication file is created under ready_indicator_filepath
# name.
# - Third, the colleague of whoever created the indicator name watches for
# the indication file to appear, and once it's there it access the
# archive.
ready_indicator_filepath: Path
def __init__(
self, base_dir: Path, archive_name: str, ready_indicator_name: str):
"""
Construct the object from given base directory and name of the archive
file:
ArchiveWithIndicator(Path('X:\\TEMP'), 'FOO.ZIP', 'INPUT_READY')
"""
self.base_dir = base_dir
self.archive_filepath = self.base_dir / archive_name
self.ready_indicator_filepath = self.base_dir / ready_indicator_name
def is_ready_unsafe(self) -> bool:
"""
Check whether the archive is ready for access.
No guarding about possible network failres is done here.
"""
if not self.ready_indicator_filepath.exists():
return False
try:
archive_state = ArchiveState.deserialize_from_file(
self.ready_indicator_filepath)
except ArchiveStateError as error:
print(f'Error deserializing archive state: {error.message}')
return False
if archive_state.has_error():
# If the error did happen during codesign procedure there will be no
# corresponding archive file.
# The caller code will deal with the error check further.
return True
# Sometimes on macOS indicator file appears prior to the actual archive
# despite the order of creation and os.sync() used in tag_ready().
# So consider archive not ready if there is an indicator without an
# actual archive.
if not self.archive_filepath.exists():
print('Found indicator without actual archive, waiting for archive '
f'({self.archive_filepath}) to appear.')
return False
# Wait for until archive is fully stored.
actual_archive_size = self.archive_filepath.stat().st_size
if actual_archive_size != archive_state.file_size:
print('Partial/invalid archive size (expected '
f'{archive_state.file_size} got {actual_archive_size})')
return False
return True
def is_ready(self) -> bool:
"""
Check whether the archive is ready for access.
Will tolerate possible network failures: if there is a network failure
or if there is still no proper permission on a file False is returned.
"""
# There are some intermitten problem happening at a random which is
# translates to "OSError : [WinError 59] An unexpected network error occurred".
# Some reports suggests it might be due to lack of permissions to the file,
# which might be applicable in our case since it's possible that file is
# initially created with non-accessible permissions and gets chmod-ed
# after initial creation.
try:
return self.is_ready_unsafe()
except OSError as e:
print(f'Exception checking archive: {e}')
return False
def tag_ready(self, error_message='') -> None:
"""
Tag the archive as ready by creating the corresponding indication file.
NOTE: It is expected that the archive was never tagged as ready before
and that there are no subsequent tags of the same archive.
If it is violated, an assert will fail.
"""
assert not self.is_ready()
# Try the best to make sure everything is synced to the file system,
# to avoid any possibility of stamp appearing on a network share prior to
# an actual file.
if util.get_current_platform() != util.Platform.WINDOWS:
os.sync()
archive_size = -1
if self.archive_filepath.exists():
archive_size = self.archive_filepath.stat().st_size
archive_info = ArchiveState(
file_size=archive_size, error_message=error_message)
self.ready_indicator_filepath.write_text(
archive_info.serialize_to_string())
def get_state(self) -> ArchiveState:
"""
Get state object for this archive
The state is read from the corresponding state file.
"""
try:
return ArchiveState.deserialize_from_file(self.ready_indicator_filepath)
except ArchiveStateError as error:
return ArchiveState(error_message=f'Error in information format: {error}')
def clean(self) -> None:
"""
Remove both archive and the ready indication file.
"""
util.ensure_file_does_not_exist_or_die(self.ready_indicator_filepath)
util.ensure_file_does_not_exist_or_die(self.archive_filepath)
def is_fully_absent(self) -> bool:
"""
Check whether both archive and its ready indicator are absent.
Is used for a sanity check during code signing process by both
buildbot worker and signing server.
"""
return (not self.archive_filepath.exists() and
not self.ready_indicator_filepath.exists())

View File

@@ -1,501 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# Signing process overview.
#
# From buildbot worker side:
# - Files which needs to be signed are collected from either a directory to
# sign all signable files in there, or by filename of a single file to sign.
# - Those files gets packed into an archive and stored in a location location
# which is watched by the signing server.
# - A marker READY file is created which indicates the archive is ready for
# access.
# - Wait for the server to provide an archive with signed files.
# This is done by watching for the READY file which corresponds to an archive
# coming from the signing server.
# - Unpack the signed signed files from the archives and replace original ones.
#
# From code sign server:
# - Watch special location for a READY file which indicates the there is an
# archive with files which are to be signed.
# - Unpack the archive to a temporary location.
# - Run codesign tool and make sure all the files are signed.
# - Pack the signed files and store them in a location which is watched by
# the buildbot worker.
# - Create a READY file which indicates that the archive with signed files is
# ready.
import abc
import logging
import shutil
import subprocess
import time
import tarfile
import uuid
from pathlib import Path
from tempfile import TemporaryDirectory
from typing import Iterable, List
import codesign.util as util
from codesign.absolute_and_relative_filename import AbsoluteAndRelativeFileName
from codesign.archive_with_indicator import ArchiveWithIndicator
from codesign.exception import CodeSignException
logger = logging.getLogger(__name__)
logger_builder = logger.getChild('builder')
logger_server = logger.getChild('server')
def pack_files(files: Iterable[AbsoluteAndRelativeFileName],
archive_filepath: Path) -> None:
"""
Create tar archive from given files for the signing pipeline.
Is used by buildbot worker to create an archive of files which are to be
signed, and by signing server to send signed files back to the worker.
"""
with tarfile.TarFile.open(archive_filepath, 'w') as tar_file_handle:
for file_info in files:
tar_file_handle.add(file_info.absolute_filepath,
arcname=file_info.relative_filepath)
def extract_files(archive_filepath: Path,
extraction_dir: Path) -> None:
"""
Extract all files form the given archive into the given direcotry.
"""
# TODO(sergey): Verify files in the archive have relative path.
with tarfile.TarFile.open(archive_filepath, mode='r') as tar_file_handle:
tar_file_handle.extractall(path=extraction_dir)
class BaseCodeSigner(metaclass=abc.ABCMeta):
"""
Base class for a platform-specific signer of binaries.
Contains all the logic shared across platform-specific implementations, such
as synchronization and notification logic.
Platform specific bits (such as actual command for signing the binary) are
to be implemented as a subclass.
Provides utilities code signing as a whole, including functionality needed
by a signing server and a buildbot worker.
The signer and builder may run on separate machines, the only requirement is
that they have access to a directory which is shared between them. For the
security concerns this is to be done as a separate machine (or as a Shared
Folder configuration in VirtualBox configuration). This directory might be
mounted under different base paths, but its underlying storage is to be
the same.
The code signer is short-lived on a buildbot worker side, and is living
forever on a code signing server side.
"""
# TODO(sergey): Find a neat way to have config annotated.
# config: Config
# Storage directory where builder puts files which are requested to be
# signed.
# Consider this an input of the code signing server.
unsigned_storage_dir: Path
# Storage where signed files are stored.
# Consider this an output of the code signer server.
signed_storage_dir: Path
# Platform the code is currently executing on.
platform: util.Platform
def __init__(self, config):
self.config = config
absolute_shared_storage_dir = config.SHARED_STORAGE_DIR.resolve()
# Unsigned (signing server input) configuration.
self.unsigned_storage_dir = absolute_shared_storage_dir / 'unsigned'
# Signed (signing server output) configuration.
self.signed_storage_dir = absolute_shared_storage_dir / 'signed'
self.platform = util.get_current_platform()
def cleanup_environment_for_builder(self) -> None:
# TODO(sergey): Revisit need of cleaning up the existing files.
# In practice it wasn't so helpful, and with multiple clients
# talking to the same server it becomes even more tricky.
pass
def cleanup_environment_for_signing_server(self) -> None:
# TODO(sergey): Revisit need of cleaning up the existing files.
# In practice it wasn't so helpful, and with multiple clients
# talking to the same server it becomes even more tricky.
pass
def generate_request_id(self) -> str:
"""
Generate an unique identifier for code signing request.
"""
return str(uuid.uuid4())
def archive_info_for_request_id(
self, path: Path, request_id: str) -> ArchiveWithIndicator:
return ArchiveWithIndicator(
path, f'{request_id}.tar', f'{request_id}.ready')
def signed_archive_info_for_request_id(
self, request_id: str) -> ArchiveWithIndicator:
return self.archive_info_for_request_id(
self.signed_storage_dir, request_id)
def unsigned_archive_info_for_request_id(
self, request_id: str) -> ArchiveWithIndicator:
return self.archive_info_for_request_id(
self.unsigned_storage_dir, request_id)
############################################################################
# Buildbot worker side helpers.
@abc.abstractmethod
def check_file_is_to_be_signed(
self, file: AbsoluteAndRelativeFileName) -> bool:
"""
Check whether file is to be signed.
Is used by both single file signing pipeline and recursive directory
signing pipeline.
This is where code signer is to check whether file is to be signed or
not. This check might be based on a simple extension test or on actual
test whether file have a digital signature already or not.
"""
def collect_files_to_sign(self, path: Path) \
-> List[AbsoluteAndRelativeFileName]:
"""
Get all files which need to be signed from the given path.
NOTE: The path might either be a file or directory.
This function is run from the buildbot worker side.
"""
# If there is a single file provided trust the buildbot worker that it
# is eligible for signing.
if path.is_file():
file = AbsoluteAndRelativeFileName.from_path(path)
if not self.check_file_is_to_be_signed(file):
return []
return [file]
all_files = AbsoluteAndRelativeFileName.recursively_from_directory(
path)
files_to_be_signed = [file for file in all_files
if self.check_file_is_to_be_signed(file)]
return files_to_be_signed
def wait_for_signed_archive_or_die(self, request_id) -> None:
"""
Wait until archive with signed files is available.
Will only return if the archive with signed files is available. If there
was an error during code sign procedure the SystemExit exception is
raised, with the message set to the error reported by the codesign
server.
Will only wait for the configured time. If that time exceeds and there
is still no responce from the signing server the application will exit
with a non-zero exit code.
"""
signed_archive_info = self.signed_archive_info_for_request_id(
request_id)
unsigned_archive_info = self.unsigned_archive_info_for_request_id(
request_id)
timeout_in_seconds = self.config.TIMEOUT_IN_SECONDS
time_start = time.monotonic()
while not signed_archive_info.is_ready():
time.sleep(1)
time_slept_in_seconds = time.monotonic() - time_start
if time_slept_in_seconds > timeout_in_seconds:
signed_archive_info.clean()
unsigned_archive_info.clean()
raise SystemExit("Signing server didn't finish signing in "
f'{timeout_in_seconds} seconds, dying :(')
archive_state = signed_archive_info.get_state()
if archive_state.has_error():
signed_archive_info.clean()
unsigned_archive_info.clean()
raise SystemExit(
f'Error happenned during codesign procedure: {archive_state.error_message}')
def copy_signed_files_to_directory(
self, signed_dir: Path, destination_dir: Path) -> None:
"""
Copy all files from signed_dir to destination_dir.
This function will overwrite any existing file. Permissions are copied
from the source files, but other metadata, such as timestamps, are not.
"""
for signed_filepath in signed_dir.glob('**/*'):
if not signed_filepath.is_file():
continue
relative_filepath = signed_filepath.relative_to(signed_dir)
destination_filepath = destination_dir / relative_filepath
destination_filepath.parent.mkdir(parents=True, exist_ok=True)
shutil.copy(signed_filepath, destination_filepath)
def run_buildbot_path_sign_pipeline(self, path: Path) -> None:
"""
Run all steps needed to make given path signed.
Path points to an unsigned file or a directory which contains unsigned
files.
If the path points to a single file then this file will be signed.
This is used to sign a final bundle such as .msi on Windows or .dmg on
macOS.
NOTE: The code signed implementation might actually reject signing the
file, in which case the file will be left unsigned. This isn't anything
to be considered a failure situation, just might happen when buildbot
worker can not detect whether signing is really required in a specific
case or not.
If the path points to a directory then code signer will sign all
signable files from it (finding them recursively).
"""
self.cleanup_environment_for_builder()
# Make sure storage directory exists.
self.unsigned_storage_dir.mkdir(parents=True, exist_ok=True)
# Collect all files which needs to be signed and pack them into a single
# archive which will be sent to the signing server.
logger_builder.info('Collecting files which are to be signed...')
files = self.collect_files_to_sign(path)
if not files:
logger_builder.info('No files to be signed, ignoring.')
return
logger_builder.info('Found %d files to sign.', len(files))
request_id = self.generate_request_id()
signed_archive_info = self.signed_archive_info_for_request_id(
request_id)
unsigned_archive_info = self.unsigned_archive_info_for_request_id(
request_id)
pack_files(files=files,
archive_filepath=unsigned_archive_info.archive_filepath)
unsigned_archive_info.tag_ready()
# Wait for the signing server to finish signing.
logger_builder.info('Waiting signing server to sign the files...')
self.wait_for_signed_archive_or_die(request_id)
# Extract signed files from archive and move files to final location.
with TemporaryDirectory(prefix='blender-buildbot-') as temp_dir_str:
unpacked_signed_files_dir = Path(temp_dir_str)
logger_builder.info('Extracting signed files from archive...')
extract_files(
archive_filepath=signed_archive_info.archive_filepath,
extraction_dir=unpacked_signed_files_dir)
destination_dir = path
if destination_dir.is_file():
destination_dir = destination_dir.parent
self.copy_signed_files_to_directory(
unpacked_signed_files_dir, destination_dir)
logger_builder.info('Removing archive with signed files...')
signed_archive_info.clean()
############################################################################
# Signing server side helpers.
def wait_for_sign_request(self) -> str:
"""
Wait for the buildbot to request signing of an archive.
Returns an identifier of signing request.
"""
# TOOD(sergey): Support graceful shutdown on Ctrl-C.
logger_server.info(
f'Waiting for a request directory {self.unsigned_storage_dir} to appear.')
while not self.unsigned_storage_dir.exists():
time.sleep(1)
logger_server.info(
'Waiting for a READY indicator of any signing request.')
request_id = None
while request_id is None:
for file in self.unsigned_storage_dir.iterdir():
if file.suffix != '.ready':
continue
request_id = file.stem
logger_server.info(f'Found READY for request ID {request_id}.')
if request_id is None:
time.sleep(1)
unsigned_archive_info = self.unsigned_archive_info_for_request_id(
request_id)
while not unsigned_archive_info.is_ready():
time.sleep(1)
return request_id
@abc.abstractmethod
def sign_all_files(self, files: List[AbsoluteAndRelativeFileName]) -> None:
"""
Sign all files in the given directory.
NOTE: Signing should happen in-place.
"""
def run_signing_pipeline(self, request_id: str):
"""
Run the full signing pipeline starting from the point when buildbot
worker have requested signing.
"""
# Make sure storage directory exists.
self.signed_storage_dir.mkdir(parents=True, exist_ok=True)
with TemporaryDirectory(prefix='blender-codesign-') as temp_dir_str:
temp_dir = Path(temp_dir_str)
signed_archive_info = self.signed_archive_info_for_request_id(
request_id)
unsigned_archive_info = self.unsigned_archive_info_for_request_id(
request_id)
logger_server.info('Extracting unsigned files from archive...')
extract_files(
archive_filepath=unsigned_archive_info.archive_filepath,
extraction_dir=temp_dir)
logger_server.info('Collecting all files which needs signing...')
files = AbsoluteAndRelativeFileName.recursively_from_directory(
temp_dir)
logger_server.info('Signing all requested files...')
try:
self.sign_all_files(files)
except CodeSignException as error:
signed_archive_info.tag_ready(error_message=error.message)
unsigned_archive_info.clean()
logger_server.info('Signing is complete with errors.')
return
logger_server.info('Packing signed files...')
pack_files(files=files,
archive_filepath=signed_archive_info.archive_filepath)
signed_archive_info.tag_ready()
logger_server.info('Removing signing request...')
unsigned_archive_info.clean()
logger_server.info('Signing is complete.')
def run_signing_server(self):
logger_server.info('Starting new code signing server...')
self.cleanup_environment_for_signing_server()
logger_server.info('Code signing server is ready')
while True:
logger_server.info('Waiting for the signing request in %s...',
self.unsigned_storage_dir)
request_id = self.wait_for_sign_request()
logger_server.info(
f'Beging signign procedure for request ID {request_id}.')
self.run_signing_pipeline(request_id)
############################################################################
# Command executing.
#
# Abstracted to a degree that allows to run commands from a foreign
# platform.
# The goal with this is to allow performing dry-run tests of code signer
# server from other platforms (for example, to test that macOS code signer
# does what it is supposed to after doing a refactor on Linux).
# TODO(sergey): What is the type annotation for the command?
def run_command_or_mock(self, command, platform: util.Platform) -> None:
"""
Run given command if current platform matches given one
If the platform is different then it will only be printed allowing
to verify logic of the code signing process.
"""
if platform != self.platform:
logger_server.info(
f'Will run command for {platform}: {command}')
return
logger_server.info(f'Running command: {command}')
subprocess.run(command)
# TODO(sergey): What is the type annotation for the command?
def check_output_or_mock(self, command,
platform: util.Platform,
allow_nonzero_exit_code=False) -> str:
"""
Run given command if current platform matches given one
If the platform is different then it will only be printed allowing
to verify logic of the code signing process.
If allow_nonzero_exit_code is truth then the output will be returned
even if application quit with non-zero exit code.
Otherwise an subprocess.CalledProcessError exception will be raised
in such case.
"""
if platform != self.platform:
logger_server.info(
f'Will run command for {platform}: {command}')
return
if allow_nonzero_exit_code:
process = subprocess.Popen(command,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT)
output = process.communicate()[0]
return output.decode()
logger_server.info(f'Running command: {command}')
return subprocess.check_output(
command, stderr=subprocess.STDOUT).decode()

View File

@@ -1,62 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# Configuration of a code signer which is specific to the code running from
# buildbot's worker.
import sys
from pathlib import Path
import codesign.util as util
from codesign.config_common import *
platform = util.get_current_platform()
if platform == util.Platform.LINUX:
SHARED_STORAGE_DIR = Path('/data/codesign')
elif platform == util.Platform.WINDOWS:
SHARED_STORAGE_DIR = Path('Z:\\codesign')
elif platform == util.Platform.MACOS:
SHARED_STORAGE_DIR = Path('/Volumes/codesign_macos/codesign')
# https://docs.python.org/3/library/logging.config.html#configuration-dictionary-schema
LOGGING = {
'version': 1,
'formatters': {
'default': {'format': '%(asctime)-15s %(levelname)8s %(name)s %(message)s'}
},
'handlers': {
'console': {
'class': 'logging.StreamHandler',
'formatter': 'default',
'stream': 'ext://sys.stderr',
}
},
'loggers': {
'codesign': {'level': 'INFO'},
},
'root': {
'level': 'WARNING',
'handlers': [
'console',
],
}
}

View File

@@ -1,36 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
from pathlib import Path
# Timeout in seconds for the signing process.
#
# This is how long buildbot packing step will wait signing server to
# perform signing.
#
# NOTE: Notarization could take a long time, hence the rather high value
# here. Might consider using different timeout for different platforms.
TIMEOUT_IN_SECONDS = 45 * 60 * 60
# Directory which is shared across buildbot worker and signing server.
#
# This is where worker puts files requested for signing as well as where
# server puts signed files.
SHARED_STORAGE_DIR: Path

View File

@@ -1,101 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# Configuration of a code signer which is specific to the code signing server.
#
# NOTE: DO NOT put any sensitive information here, put it in an actual
# configuration on the signing machine.
from pathlib import Path
from codesign.config_common import *
CODESIGN_DIRECTORY = Path(__file__).absolute().parent
BLENDER_GIT_ROOT_DIRECTORY = CODESIGN_DIRECTORY.parent.parent.parent
################################################################################
# Common configuration.
# Directory where folders for codesign requests and signed result are stored.
# For example, /data/codesign
SHARED_STORAGE_DIR: Path
################################################################################
# macOS-specific configuration.
MACOS_ENTITLEMENTS_FILE = \
BLENDER_GIT_ROOT_DIRECTORY / 'release' / 'darwin' / 'entitlements.plist'
# Identity of the Developer ID Application certificate which is to be used for
# codesign tool.
# Use `security find-identity -v -p codesigning` to find the identity.
#
# NOTE: This identity is just an example from release/darwin/README.txt.
MACOS_CODESIGN_IDENTITY = 'AE825E26F12D08B692F360133210AF46F4CF7B97'
# User name (Apple ID) which will be used to request notarization.
MACOS_XCRUN_USERNAME = 'me@example.com'
# One-time application password which will be used to request notarization.
MACOS_XCRUN_PASSWORD = '@keychain:altool-password'
# Timeout in seconds within which the notarial office is supposed to reply.
MACOS_NOTARIZE_TIMEOUT_IN_SECONDS = 60 * 60
################################################################################
# Windows-specific configuration.
# URL to the timestamping authority.
WIN_TIMESTAMP_AUTHORITY_URL = 'http://timestamp.digicert.com'
# Full path to the certificate used for signing.
#
# The path and expected file format might vary depending on a platform.
#
# On Windows it is usually is a PKCS #12 key (.pfx), so the path will look
# like Path('C:\\Secret\\Blender.pfx').
WIN_CERTIFICATE_FILEPATH: Path
################################################################################
# Logging configuration, common for all platforms.
# https://docs.python.org/3/library/logging.config.html#configuration-dictionary-schema
LOGGING = {
'version': 1,
'formatters': {
'default': {'format': '%(asctime)-15s %(levelname)8s %(name)s %(message)s'}
},
'handlers': {
'console': {
'class': 'logging.StreamHandler',
'formatter': 'default',
'stream': 'ext://sys.stderr',
}
},
'loggers': {
'codesign': {'level': 'INFO'},
},
'root': {
'level': 'WARNING',
'handlers': [
'console',
],
}
}

View File

@@ -1,26 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
class CodeSignException(Exception):
message: str
def __init__(self, message):
self.message = message
super().__init__(self.message)

View File

@@ -1,72 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# NOTE: This is a no-op signer (since there isn't really a procedure to sign
# Linux binaries yet). Used to debug and verify the code signing routines on
# a Linux environment.
import logging
from pathlib import Path
from typing import List
from codesign.absolute_and_relative_filename import AbsoluteAndRelativeFileName
from codesign.base_code_signer import BaseCodeSigner
logger = logging.getLogger(__name__)
logger_server = logger.getChild('server')
class LinuxCodeSigner(BaseCodeSigner):
def is_active(self) -> bool:
"""
Check whether this signer is active.
if it is inactive, no files will be signed.
Is used to be able to debug code signing pipeline on Linux, where there
is no code signing happening in the actual buildbot and release
environment.
"""
return False
def check_file_is_to_be_signed(
self, file: AbsoluteAndRelativeFileName) -> bool:
if file.relative_filepath == Path('blender'):
return True
if (file.relative_filepath.parts[-3:-1] == ('python', 'bin') and
file.relative_filepath.name.startwith('python')):
return True
if file.relative_filepath.suffix == '.so':
return True
return False
def collect_files_to_sign(self, path: Path) \
-> List[AbsoluteAndRelativeFileName]:
if not self.is_active():
return []
return super().collect_files_to_sign(path)
def sign_all_files(self, files: List[AbsoluteAndRelativeFileName]) -> None:
num_files = len(files)
for file_index, file in enumerate(files):
logger.info('Server: Signed file [%d/%d] %s',
file_index + 1, num_files, file.relative_filepath)

View File

@@ -1,456 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import logging
import re
import stat
import subprocess
import time
from pathlib import Path
from typing import List
import codesign.util as util
from buildbot_utils import Builder
from codesign.absolute_and_relative_filename import AbsoluteAndRelativeFileName
from codesign.base_code_signer import BaseCodeSigner
from codesign.exception import CodeSignException
logger = logging.getLogger(__name__)
logger_server = logger.getChild('server')
# NOTE: Check is done as filename.endswith(), so keep the dot
EXTENSIONS_TO_BE_SIGNED = {'.dylib', '.so', '.dmg'}
# Prefixes of a file (not directory) name which are to be signed.
# Used to sign extra executable files in Contents/Resources.
NAME_PREFIXES_TO_BE_SIGNED = {'python'}
class NotarizationException(CodeSignException):
pass
def is_file_from_bundle(file: AbsoluteAndRelativeFileName) -> bool:
"""
Check whether file is coming from an .app bundle
"""
parts = file.relative_filepath.parts
if not parts:
return False
if not parts[0].endswith('.app'):
return False
return True
def get_bundle_from_file(
file: AbsoluteAndRelativeFileName) -> AbsoluteAndRelativeFileName:
"""
Get AbsoluteAndRelativeFileName descriptor of bundle
"""
assert(is_file_from_bundle(file))
parts = file.relative_filepath.parts
bundle_name = parts[0]
base_dir = file.base_dir
bundle_filepath = file.base_dir / bundle_name
return AbsoluteAndRelativeFileName(base_dir, bundle_filepath)
def is_bundle_executable_file(file: AbsoluteAndRelativeFileName) -> bool:
"""
Check whether given file is an executable within an app bundle
"""
if not is_file_from_bundle(file):
return False
parts = file.relative_filepath.parts
num_parts = len(parts)
if num_parts < 3:
return False
if parts[1:3] != ('Contents', 'MacOS'):
return False
return True
def xcrun_field_value_from_output(field: str, output: str) -> str:
"""
Get value of a given field from xcrun output.
If field is not found empty string is returned.
"""
field_prefix = field + ': '
for line in output.splitlines():
line = line.strip()
if line.startswith(field_prefix):
return line[len(field_prefix):]
return ''
class MacOSCodeSigner(BaseCodeSigner):
def check_file_is_to_be_signed(
self, file: AbsoluteAndRelativeFileName) -> bool:
if file.relative_filepath.name.startswith('.'):
return False
if is_bundle_executable_file(file):
return True
base_name = file.relative_filepath.name
if any(base_name.startswith(prefix)
for prefix in NAME_PREFIXES_TO_BE_SIGNED):
return True
mode = file.absolute_filepath.lstat().st_mode
if mode & stat.S_IXUSR != 0:
file_output = subprocess.check_output(
("file", file.absolute_filepath)).decode()
if "64-bit executable" in file_output:
return True
return file.relative_filepath.suffix in EXTENSIONS_TO_BE_SIGNED
def collect_files_to_sign(self, path: Path) \
-> List[AbsoluteAndRelativeFileName]:
# Include all files when signing app or dmg bundle: all the files are
# needed to do valid signature of bundle.
if path.name.endswith('.app'):
return AbsoluteAndRelativeFileName.recursively_from_directory(path)
if path.is_dir():
files = []
for child in path.iterdir():
if child.name.endswith('.app'):
current_files = AbsoluteAndRelativeFileName.recursively_from_directory(
child)
else:
current_files = super().collect_files_to_sign(child)
for current_file in current_files:
files.append(AbsoluteAndRelativeFileName(
path, current_file.absolute_filepath))
return files
return super().collect_files_to_sign(path)
############################################################################
# Codesign.
def codesign_remove_signature(
self, file: AbsoluteAndRelativeFileName) -> None:
"""
Make sure given file does not have codesign signature
This is needed because codesigning is not possible for file which has
signature already.
"""
logger_server.info(
'Removing codesign signature from %s...', file.relative_filepath)
command = ['codesign', '--remove-signature', file.absolute_filepath]
self.run_command_or_mock(command, util.Platform.MACOS)
def codesign_file(
self, file: AbsoluteAndRelativeFileName) -> None:
"""
Sign given file
NOTE: File must not have any signatures.
"""
logger_server.info(
'Codesigning %s...', file.relative_filepath)
entitlements_file = self.config.MACOS_ENTITLEMENTS_FILE
command = ['codesign',
'--timestamp',
'--options', 'runtime',
f'--entitlements={entitlements_file}',
'--sign', self.config.MACOS_CODESIGN_IDENTITY,
file.absolute_filepath]
self.run_command_or_mock(command, util.Platform.MACOS)
def codesign_all_files(self, files: List[AbsoluteAndRelativeFileName]) -> None:
"""
Run codesign tool on all eligible files in the given list.
Will ignore all files which are not to be signed. For the rest will
remove possible existing signature and add a new signature.
"""
num_files = len(files)
have_ignored_files = False
signed_files = []
for file_index, file in enumerate(files):
# Ignore file if it is not to be signed.
# Allows to manually construct ZIP of a bundle and get it signed.
if not self.check_file_is_to_be_signed(file):
logger_server.info(
'Ignoring file [%d/%d] %s',
file_index + 1, num_files, file.relative_filepath)
have_ignored_files = True
continue
logger_server.info(
'Running codesigning routines for file [%d/%d] %s...',
file_index + 1, num_files, file.relative_filepath)
self.codesign_remove_signature(file)
self.codesign_file(file)
signed_files.append(file)
if have_ignored_files:
logger_server.info('Signed %d files:', len(signed_files))
num_signed_files = len(signed_files)
for file_index, signed_file in enumerate(signed_files):
logger_server.info(
'- [%d/%d] %s',
file_index + 1, num_signed_files,
signed_file.relative_filepath)
def codesign_bundles(
self, files: List[AbsoluteAndRelativeFileName]) -> None:
"""
Codesign all .app bundles in the given list of files.
Bundle is deducted from paths of the files, and every bundle is only
signed once.
"""
signed_bundles = set()
extra_files = []
for file in files:
if not is_file_from_bundle(file):
continue
bundle = get_bundle_from_file(file)
bundle_name = bundle.relative_filepath
if bundle_name in signed_bundles:
continue
logger_server.info('Running codesign routines on bundle %s',
bundle_name)
# It is not possible to remove signature from DMG.
if bundle.relative_filepath.name.endswith('.app'):
self.codesign_remove_signature(bundle)
self.codesign_file(bundle)
signed_bundles.add(bundle_name)
# Codesign on a bundle adds an extra folder with information.
# It needs to be compied to the source.
code_signature_directory = \
bundle.absolute_filepath / 'Contents' / '_CodeSignature'
code_signature_files = \
AbsoluteAndRelativeFileName.recursively_from_directory(
code_signature_directory)
for code_signature_file in code_signature_files:
bundle_relative_file = AbsoluteAndRelativeFileName(
bundle.base_dir,
code_signature_directory /
code_signature_file.relative_filepath)
extra_files.append(bundle_relative_file)
files.extend(extra_files)
############################################################################
# Notarization.
def notarize_get_bundle_id(self, file: AbsoluteAndRelativeFileName) -> str:
"""
Get bundle ID which will be used to notarize DMG
"""
name = file.relative_filepath.name
app_name = name.split('-', 2)[0].lower()
app_name_words = app_name.split()
if len(app_name_words) > 1:
app_name_id = ''.join(word.capitalize() for word in app_name_words)
else:
app_name_id = app_name_words[0]
# TODO(sergey): Consider using "alpha" for buildbot builds.
return f'org.blenderfoundation.{app_name_id}.release'
def notarize_request(self, file) -> str:
"""
Request notarization of the given file.
Returns UUID of the notarization request. If error occurred None is
returned instead of UUID.
"""
bundle_id = self.notarize_get_bundle_id(file)
logger_server.info('Bundle ID: %s', bundle_id)
logger_server.info('Submitting file to the notarial office.')
command = [
'xcrun', 'altool', '--notarize-app', '--verbose',
'-f', file.absolute_filepath,
'--primary-bundle-id', bundle_id,
'--username', self.config.MACOS_XCRUN_USERNAME,
'--password', self.config.MACOS_XCRUN_PASSWORD]
output = self.check_output_or_mock(
command, util.Platform.MACOS, allow_nonzero_exit_code=True)
for line in output.splitlines():
line = line.strip()
if line.startswith('RequestUUID = '):
request_uuid = line[14:]
return request_uuid
# Check whether the package has been already submitted.
if 'The software asset has already been uploaded.' in line:
request_uuid = re.sub(
'.*The upload ID is ([A-Fa-f0-9\-]+).*', '\\1', line)
logger_server.warning(
f'The package has been already submitted under UUID {request_uuid}')
return request_uuid
logger_server.error(output)
logger_server.error('xcrun command did not report RequestUUID')
return None
def notarize_review_status(self, xcrun_output: str) -> bool:
"""
Review status returned by xcrun's notarization info
Returns truth if the notarization process has finished.
If there are errors during notarization, a NotarizationException()
exception is thrown with status message from the notarial office.
"""
# Parse status and message
status = xcrun_field_value_from_output('Status', xcrun_output)
status_message = xcrun_field_value_from_output(
'Status Message', xcrun_output)
if status == 'success':
logger_server.info(
'Package successfully notarized: %s', status_message)
return True
if status == 'invalid':
logger_server.error(xcrun_output)
logger_server.error(
'Package notarization has failed: %s', status_message)
raise NotarizationException(status_message)
if status == 'in progress':
return False
logger_server.info(
'Unknown notarization status %s (%s)', status, status_message)
return False
def notarize_wait_result(self, request_uuid: str) -> None:
"""
Wait for until notarial office have a reply
"""
logger_server.info(
'Waiting for a result from the notarization office.')
command = ['xcrun', 'altool',
'--notarization-info', request_uuid,
'--username', self.config.MACOS_XCRUN_USERNAME,
'--password', self.config.MACOS_XCRUN_PASSWORD]
time_start = time.monotonic()
timeout_in_seconds = self.config.MACOS_NOTARIZE_TIMEOUT_IN_SECONDS
while True:
xcrun_output = self.check_output_or_mock(
command, util.Platform.MACOS, allow_nonzero_exit_code=True)
if self.notarize_review_status(xcrun_output):
break
logger_server.info('Keep waiting for notarization office.')
time.sleep(30)
time_slept_in_seconds = time.monotonic() - time_start
if time_slept_in_seconds > timeout_in_seconds:
logger_server.error(
"Notarial office didn't reply in %f seconds.",
timeout_in_seconds)
def notarize_staple(self, file: AbsoluteAndRelativeFileName) -> bool:
"""
Staple notarial label on the file
"""
logger_server.info('Stapling notarial stamp.')
command = ['xcrun', 'stapler', 'staple', '-v', file.absolute_filepath]
self.check_output_or_mock(command, util.Platform.MACOS)
def notarize_dmg(self, file: AbsoluteAndRelativeFileName) -> bool:
"""
Run entire pipeline to get DMG notarized.
"""
logger_server.info('Begin notarization routines on %s',
file.relative_filepath)
# Submit file for notarization.
request_uuid = self.notarize_request(file)
if not request_uuid:
return False
logger_server.info('Received Request UUID: %s', request_uuid)
# Wait for the status from the notarization office.
if not self.notarize_wait_result(request_uuid):
return False
# Staple.
self.notarize_staple(file)
def notarize_all_dmg(
self, files: List[AbsoluteAndRelativeFileName]) -> bool:
"""
Notarize all DMG images from the input.
Images are supposed to be codesigned already.
"""
for file in files:
if not file.relative_filepath.name.endswith('.dmg'):
continue
if not self.check_file_is_to_be_signed(file):
continue
self.notarize_dmg(file)
############################################################################
# Entry point.
def sign_all_files(self, files: List[AbsoluteAndRelativeFileName]) -> None:
# TODO(sergey): Handle errors somehow.
self.codesign_all_files(files)
self.codesign_bundles(files)
self.notarize_all_dmg(files)

View File

@@ -1,52 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import logging.config
import sys
from pathlib import Path
from typing import Optional
import codesign.config_builder
import codesign.util as util
from codesign.base_code_signer import BaseCodeSigner
class SimpleCodeSigner:
code_signer: Optional[BaseCodeSigner]
def __init__(self):
platform = util.get_current_platform()
if platform == util.Platform.LINUX:
from codesign.linux_code_signer import LinuxCodeSigner
self.code_signer = LinuxCodeSigner(codesign.config_builder)
elif platform == util.Platform.MACOS:
from codesign.macos_code_signer import MacOSCodeSigner
self.code_signer = MacOSCodeSigner(codesign.config_builder)
elif platform == util.Platform.WINDOWS:
from codesign.windows_code_signer import WindowsCodeSigner
self.code_signer = WindowsCodeSigner(codesign.config_builder)
else:
self.code_signer = None
def sign_file_or_directory(self, path: Path) -> None:
logging.config.dictConfig(codesign.config_builder.LOGGING)
self.code_signer.run_buildbot_path_sign_pipeline(path)

View File

@@ -1,54 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import sys
from enum import Enum
from pathlib import Path
class Platform(Enum):
LINUX = 1
MACOS = 2
WINDOWS = 3
def get_current_platform() -> Platform:
if sys.platform == 'linux':
return Platform.LINUX
elif sys.platform == 'darwin':
return Platform.MACOS
elif sys.platform == 'win32':
return Platform.WINDOWS
raise Exception(f'Unknown platform {sys.platform}')
def ensure_file_does_not_exist_or_die(filepath: Path) -> None:
"""
If the file exists, unlink it.
If the file path exists and is not a file an assert will trigger.
If the file path does not exists nothing happens.
"""
if not filepath.exists():
return
if not filepath.is_file():
# TODO(sergey): Provide information about what the filepath actually is.
raise SystemExit(f'{filepath} is expected to be a file, but is not')
filepath.unlink()

View File

@@ -1,117 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import logging
from pathlib import Path
from typing import List
import codesign.util as util
from buildbot_utils import Builder
from codesign.absolute_and_relative_filename import AbsoluteAndRelativeFileName
from codesign.base_code_signer import BaseCodeSigner
from codesign.exception import CodeSignException
logger = logging.getLogger(__name__)
logger_server = logger.getChild('server')
# NOTE: Check is done as filename.endswith(), so keep the dot
EXTENSIONS_TO_BE_SIGNED = {'.exe', '.dll', '.pyd', '.msi'}
BLACKLIST_FILE_PREFIXES = (
'api-ms-', 'concrt', 'msvcp', 'ucrtbase', 'vcomp', 'vcruntime')
class SigntoolException(CodeSignException):
pass
class WindowsCodeSigner(BaseCodeSigner):
def check_file_is_to_be_signed(
self, file: AbsoluteAndRelativeFileName) -> bool:
base_name = file.relative_filepath.name
if any(base_name.startswith(prefix)
for prefix in BLACKLIST_FILE_PREFIXES):
return False
return file.relative_filepath.suffix in EXTENSIONS_TO_BE_SIGNED
def get_sign_command_prefix(self) -> List[str]:
return [
'signtool', 'sign', '/v',
'/f', self.config.WIN_CERTIFICATE_FILEPATH,
'/tr', self.config.WIN_TIMESTAMP_AUTHORITY_URL]
def run_codesign_tool(self, filepath: Path) -> None:
command = self.get_sign_command_prefix() + [filepath]
try:
codesign_output = self.check_output_or_mock(command, util.Platform.WINDOWS)
except subprocess.CalledProcessError as e:
raise SigntoolException(f'Error running signtool {e}')
logger_server.info(f'signtool output:\n{codesign_output}')
got_number_of_success = False
for line in codesign_output.split('\n'):
line_clean = line.strip()
line_clean_lower = line_clean.lower()
if line_clean_lower.startswith('number of warnings') or \
line_clean_lower.startswith('number of errors'):
number = int(line_clean_lower.split(':')[1])
if number != 0:
raise SigntoolException('Non-clean success of signtool')
if line_clean_lower.startswith('number of files successfully signed'):
got_number_of_success = True
number = int(line_clean_lower.split(':')[1])
if number != 1:
raise SigntoolException('Signtool did not consider codesign a success')
if not got_number_of_success:
raise SigntoolException('Signtool did not report number of files signed')
def sign_all_files(self, files: List[AbsoluteAndRelativeFileName]) -> None:
# NOTE: Sign files one by one to avoid possible command line length
# overflow (which could happen if we ever decide to sign every binary
# in the install folder, for example).
#
# TODO(sergey): Consider doing batched signing of handful of files in
# one go (but only if this actually known to be much faster).
num_files = len(files)
for file_index, file in enumerate(files):
# Ignore file if it is not to be signed.
# Allows to manually construct ZIP of package and get it signed.
if not self.check_file_is_to_be_signed(file):
logger_server.info(
'Ignoring file [%d/%d] %s',
file_index + 1, num_files, file.relative_filepath)
continue
logger_server.info(
'Running signtool command for file [%d/%d] %s...',
file_index + 1, num_files, file.relative_filepath)
self.run_codesign_tool(file.absolute_filepath)

View File

@@ -1,37 +0,0 @@
#!/usr/bin/env python3
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# NOTE: This is a no-op signer (since there isn't really a procedure to sign
# Linux binaries yet). Used to debug and verify the code signing routines on
# a Linux environment.
import logging.config
from pathlib import Path
from typing import List
from codesign.linux_code_signer import LinuxCodeSigner
import codesign.config_server
if __name__ == "__main__":
logging.config.dictConfig(codesign.config_server.LOGGING)
code_signer = LinuxCodeSigner(codesign.config_server)
code_signer.run_signing_server()

View File

@@ -1,41 +0,0 @@
#!/usr/bin/env python3
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import logging.config
from pathlib import Path
from typing import List
from codesign.macos_code_signer import MacOSCodeSigner
import codesign.config_server
if __name__ == "__main__":
entitlements_file = codesign.config_server.MACOS_ENTITLEMENTS_FILE
if not entitlements_file.exists():
raise SystemExit(
'Entitlements file {entitlements_file} does not exist.')
if not entitlements_file.is_file():
raise SystemExit(
'Entitlements file {entitlements_file} is not a file.')
logging.config.dictConfig(codesign.config_server.LOGGING)
code_signer = MacOSCodeSigner(codesign.config_server)
code_signer.run_signing_server()

View File

@@ -1,11 +0,0 @@
@echo off
rem This is an entry point of the codesign server for Windows.
rem It makes sure that signtool.exe is within the current PATH and can be
rem used by the Python script.
SETLOCAL
set PATH=C:\Program Files (x86)\Windows Kits\10\App Certification Kit;%PATH%
codesign_server_windows.py

View File

@@ -1,54 +0,0 @@
#!/usr/bin/env python3
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# Implementation of codesign server for Windows.
#
# NOTE: If signtool.exe is not in the PATH use codesign_server_windows.bat
import logging.config
import shutil
from pathlib import Path
from typing import List
import codesign.util as util
from codesign.windows_code_signer import WindowsCodeSigner
import codesign.config_server
if __name__ == "__main__":
logging.config.dictConfig(codesign.config_server.LOGGING)
logger = logging.getLogger(__name__)
logger_server = logger.getChild('server')
# TODO(sergey): Consider moving such sanity checks into
# CodeSigner.check_environment_or_die().
if not shutil.which('signtool.exe'):
if util.get_current_platform() == util.Platform.WINDOWS:
raise SystemExit("signtool.exe is not found in %PATH%")
logger_server.info(
'signtool.exe not found, '
'but will not be used on this foreign platform')
code_signer = WindowsCodeSigner(codesign.config_server)
code_signer.run_signing_server()

View File

@@ -1,551 +0,0 @@
#!/usr/bin/env python3
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
import argparse
import re
import shutil
import subprocess
import sys
import time
from pathlib import Path
from tempfile import TemporaryDirectory, NamedTemporaryFile
from typing import List
BUILDBOT_DIRECTORY = Path(__file__).absolute().parent
CODESIGN_SCRIPT = BUILDBOT_DIRECTORY / 'worker_codesign.py'
BLENDER_GIT_ROOT_DIRECTORY = BUILDBOT_DIRECTORY.parent.parent
DARWIN_DIRECTORY = BLENDER_GIT_ROOT_DIRECTORY / 'release' / 'darwin'
# Extra size which is added on top of actual files size when estimating size
# of destination DNG.
EXTRA_DMG_SIZE_IN_BYTES = 800 * 1024 * 1024
################################################################################
# Common utilities
def get_directory_size(root_directory: Path) -> int:
"""
Get size of directory on disk
"""
total_size = 0
for file in root_directory.glob('**/*'):
total_size += file.lstat().st_size
return total_size
################################################################################
# DMG bundling specific logic
def create_argument_parser():
parser = argparse.ArgumentParser()
parser.add_argument(
'source_dir',
type=Path,
help='Source directory which points to either existing .app bundle'
'or to a directory with .app bundles.')
parser.add_argument(
'--background-image',
type=Path,
help="Optional background picture which will be set on the DMG."
"If not provided default Blender's one is used.")
parser.add_argument(
'--volume-name',
type=str,
help='Optional name of a volume which will be used for DMG.')
parser.add_argument(
'--dmg',
type=Path,
help='Optional argument which points to a final DMG file name.')
parser.add_argument(
'--applescript',
type=Path,
help="Optional path to applescript to set up folder looks of DMG."
"If not provided default Blender's one is used.")
parser.add_argument(
'--codesign',
action="store_true",
help="Code sign and notarize DMG contents.")
return parser
def collect_app_bundles(source_dir: Path) -> List[Path]:
"""
Collect all app bundles which are to be put into DMG
If the source directory points to FOO.app it will be the only app bundle
packed.
Otherwise all .app bundles from given directory are placed to a single
DMG.
"""
if source_dir.name.endswith('.app'):
return [source_dir]
app_bundles = []
for filename in source_dir.glob('*'):
if not filename.is_dir():
continue
if not filename.name.endswith('.app'):
continue
app_bundles.append(filename)
return app_bundles
def collect_and_log_app_bundles(source_dir: Path) -> List[Path]:
app_bundles = collect_app_bundles(source_dir)
if not app_bundles:
print('No app bundles found for packing')
return
print(f'Found {len(app_bundles)} to pack:')
for app_bundle in app_bundles:
print(f'- {app_bundle}')
return app_bundles
def estimate_dmg_size(app_bundles: List[Path]) -> int:
"""
Estimate size of DMG to hold requested app bundles
The size is based on actual size of all files in all bundles plus some
space to compensate for different size-on-disk plus some space to hold
codesign signatures.
Is better to be on a high side since the empty space is compressed, but
lack of space might cause silent failures later on.
"""
app_bundles_size = 0
for app_bundle in app_bundles:
app_bundles_size += get_directory_size(app_bundle)
return app_bundles_size + EXTRA_DMG_SIZE_IN_BYTES
def copy_app_bundles_to_directory(app_bundles: List[Path],
directory: Path) -> None:
"""
Copy all bundles to a given directory
This directory is what the DMG will be created from.
"""
for app_bundle in app_bundles:
print(f'Copying {app_bundle.name}...')
shutil.copytree(app_bundle, directory / app_bundle.name)
def get_main_app_bundle(app_bundles: List[Path]) -> Path:
"""
Get application bundle main for the installation
"""
return app_bundles[0]
def create_dmg_image(app_bundles: List[Path],
dmg_filepath: Path,
volume_name: str) -> None:
"""
Create DMG disk image and put app bundles in it
No DMG configuration or codesigning is happening here.
"""
if dmg_filepath.exists():
print(f'Removing existing writable DMG {dmg_filepath}...')
dmg_filepath.unlink()
print('Preparing directory with app bundles for the DMG...')
with TemporaryDirectory(prefix='blender-dmg-content-') as content_dir_str:
# Copy all bundles to a clean directory.
content_dir = Path(content_dir_str)
copy_app_bundles_to_directory(app_bundles, content_dir)
# Estimate size of the DMG.
dmg_size = estimate_dmg_size(app_bundles)
print(f'Estimated DMG size: {dmg_size:,} bytes.')
# Create the DMG.
print(f'Creating writable DMG {dmg_filepath}')
command = ('hdiutil',
'create',
'-size', str(dmg_size),
'-fs', 'HFS+',
'-srcfolder', content_dir,
'-volname', volume_name,
'-format', 'UDRW',
dmg_filepath)
subprocess.run(command)
def get_writable_dmg_filepath(dmg_filepath: Path):
"""
Get file path for writable DMG image
"""
parent = dmg_filepath.parent
return parent / (dmg_filepath.stem + '-temp.dmg')
def mount_readwrite_dmg(dmg_filepath: Path) -> None:
"""
Mount writable DMG
Mounting point would be /Volumes/<volume name>
"""
print(f'Mounting read-write DMG ${dmg_filepath}')
command = ('hdiutil',
'attach', '-readwrite',
'-noverify',
'-noautoopen',
dmg_filepath)
subprocess.run(command)
def get_mount_directory_for_volume_name(volume_name: str) -> Path:
"""
Get directory under which the volume will be mounted
"""
return Path('/Volumes') / volume_name
def eject_volume(volume_name: str) -> None:
"""
Eject given volume, if mounted
"""
mount_directory = get_mount_directory_for_volume_name(volume_name)
if not mount_directory.exists():
return
mount_directory_str = str(mount_directory)
print(f'Ejecting volume {volume_name}')
# Figure out which device to eject.
mount_output = subprocess.check_output(['mount']).decode()
device = ''
for line in mount_output.splitlines():
if f'on {mount_directory_str} (' not in line:
continue
tokens = line.split(' ', 3)
if len(tokens) < 3:
continue
if tokens[1] != 'on':
continue
if device:
raise Exception(
f'Multiple devices found for mounting point {mount_directory}')
device = tokens[0]
if not device:
raise Exception(
f'No device found for mounting point {mount_directory}')
print(f'{mount_directory} is mounted as device {device}, ejecting...')
subprocess.run(['diskutil', 'eject', device])
def copy_background_if_needed(background_image_filepath: Path,
mount_directory: Path) -> None:
"""
Copy background to the DMG
If the background image is not specified it will not be copied.
"""
if not background_image_filepath:
print('No background image provided.')
return
print(f'Copying background image {background_image_filepath}')
destination_dir = mount_directory / '.background'
destination_dir.mkdir(exist_ok=True)
destination_filepath = destination_dir / background_image_filepath.name
shutil.copy(background_image_filepath, destination_filepath)
def create_applications_link(mount_directory: Path) -> None:
"""
Create link to /Applications in the given location
"""
print('Creating link to /Applications')
command = ('ln', '-s', '/Applications', mount_directory / ' ')
subprocess.run(command)
def run_applescript(applescript: Path,
volume_name: str,
app_bundles: List[Path],
background_image_filepath: Path) -> None:
"""
Run given applescript to adjust look and feel of the DMG
"""
main_app_bundle = get_main_app_bundle(app_bundles)
with NamedTemporaryFile(
mode='w', suffix='.applescript') as temp_applescript:
print('Adjusting applescript for volume name...')
# Adjust script to the specific volume name.
with open(applescript, mode='r') as input:
for line in input.readlines():
stripped_line = line.strip()
if stripped_line.startswith('tell disk'):
line = re.sub('tell disk ".*"',
f'tell disk "{volume_name}"',
line)
elif stripped_line.startswith('set background picture'):
if not background_image_filepath:
continue
else:
background_image_short = \
'.background:' + background_image_filepath.name
line = re.sub('to file ".*"',
f'to file "{background_image_short}"',
line)
line = line.replace('blender.app', main_app_bundle.name)
temp_applescript.write(line)
temp_applescript.flush()
print('Running applescript...')
command = ('osascript', temp_applescript.name)
subprocess.run(command)
print('Waiting for applescript...')
# NOTE: This is copied from bundle.sh. The exact reason for sleep is
# still remained a mystery.
time.sleep(5)
def codesign(subject: Path):
"""
Codesign file or directory
NOTE: For DMG it will also notarize.
"""
command = (CODESIGN_SCRIPT, subject)
subprocess.run(command)
def codesign_app_bundles_in_dmg(mount_directory: str) -> None:
"""
Code sign all binaries and bundles in the mounted directory
"""
print(f'Codesigning all app bundles in {mount_directory}')
codesign(mount_directory)
def codesign_and_notarize_dmg(dmg_filepath: Path) -> None:
"""
Run codesign and notarization pipeline on the DMG
"""
print(f'Codesigning and notarizing DMG {dmg_filepath}')
codesign(dmg_filepath)
def compress_dmg(writable_dmg_filepath: Path,
final_dmg_filepath: Path) -> None:
"""
Compress temporary read-write DMG
"""
command = ('hdiutil', 'convert',
writable_dmg_filepath,
'-format', 'UDZO',
'-o', final_dmg_filepath)
if final_dmg_filepath.exists():
print(f'Removing old compressed DMG {final_dmg_filepath}')
final_dmg_filepath.unlink()
print('Compressing disk image...')
subprocess.run(command)
def create_final_dmg(app_bundles: List[Path],
dmg_filepath: Path,
background_image_filepath: Path,
volume_name: str,
applescript: Path,
codesign: bool) -> None:
"""
Create DMG with all app bundles
Will take care configuring background, signing all binaries and app bundles
and notarizing the DMG.
"""
print('Running all routines to create final DMG')
writable_dmg_filepath = get_writable_dmg_filepath(dmg_filepath)
mount_directory = get_mount_directory_for_volume_name(volume_name)
# Make sure volume is not mounted.
# If it is mounted it will prevent removing old DMG files and could make
# it so app bundles are copied to the wrong place.
eject_volume(volume_name)
create_dmg_image(app_bundles, writable_dmg_filepath, volume_name)
mount_readwrite_dmg(writable_dmg_filepath)
# Run codesign first, prior to copying amything else.
#
# This allows to recurs into the content of bundles without worrying about
# possible interfereice of Application symlink.
if codesign:
codesign_app_bundles_in_dmg(mount_directory)
copy_background_if_needed(background_image_filepath, mount_directory)
create_applications_link(mount_directory)
run_applescript(applescript, volume_name, app_bundles,
background_image_filepath)
print('Ejecting read-write DMG image...')
eject_volume(volume_name)
compress_dmg(writable_dmg_filepath, dmg_filepath)
writable_dmg_filepath.unlink()
if codesign:
codesign_and_notarize_dmg(dmg_filepath)
def ensure_dmg_extension(filepath: Path) -> Path:
"""
Make sure given file have .dmg extension
"""
if filepath.suffix != '.dmg':
return filepath.with_suffix(f'{filepath.suffix}.dmg')
return filepath
def get_dmg_filepath(requested_name: Path, app_bundles: List[Path]) -> Path:
"""
Get full file path for the final DMG image
Will use the provided one when possible, otherwise will deduct it from
app bundles.
If the name is deducted, the DMG is stored in the current directory.
"""
if requested_name:
return ensure_dmg_extension(requested_name.absolute())
# TODO(sergey): This is not necessarily the main one.
main_bundle = app_bundles[0]
# Strip .app from the name
return Path(main_bundle.name[:-4] + '.dmg').absolute()
def get_background_image(requested_background_image: Path) -> Path:
"""
Get effective filepath for the background image
"""
if requested_background_image:
return requested_background_image.absolute()
return DARWIN_DIRECTORY / 'background.tif'
def get_applescript(requested_applescript: Path) -> Path:
"""
Get effective filepath for the applescript
"""
if requested_applescript:
return requested_applescript.absolute()
return DARWIN_DIRECTORY / 'blender.applescript'
def get_volume_name_from_dmg_filepath(dmg_filepath: Path) -> str:
"""
Deduct volume name from the DMG path
Will use first part of the DMG file name prior to dash.
"""
tokens = dmg_filepath.stem.split('-')
words = tokens[0].split()
return ' '.join(word.capitalize() for word in words)
def get_volume_name(requested_volume_name: str,
dmg_filepath: Path) -> str:
"""
Get effective name for DMG volume
"""
if requested_volume_name:
return requested_volume_name
return get_volume_name_from_dmg_filepath(dmg_filepath)
def main():
parser = create_argument_parser()
args = parser.parse_args()
# Get normalized input parameters.
source_dir = args.source_dir.absolute()
background_image_filepath = get_background_image(args.background_image)
applescript = get_applescript(args.applescript)
codesign = args.codesign
app_bundles = collect_and_log_app_bundles(source_dir)
if not app_bundles:
return
dmg_filepath = get_dmg_filepath(args.dmg, app_bundles)
volume_name = get_volume_name(args.volume_name, dmg_filepath)
print(f'Will produce DMG "{dmg_filepath.name}" (without quotes)')
create_final_dmg(app_bundles,
dmg_filepath,
background_image_filepath,
volume_name,
applescript,
codesign)
if __name__ == "__main__":
main()

View File

@@ -1,44 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# This is a script which is used as POST-INSTALL one for regular CMake's
# INSTALL target.
# It is used by buildbot workers to sign every binary which is going into
# the final buundle.
# On Windows Python 3 there only is python.exe, no python3.exe.
#
# On other platforms it is possible to have python2 and python3, and a
# symbolic link to python to either of them. So on those platforms use
# an explicit Python version.
if(WIN32)
set(PYTHON_EXECUTABLE python)
else()
set(PYTHON_EXECUTABLE python3)
endif()
execute_process(
COMMAND ${PYTHON_EXECUTABLE} "${CMAKE_CURRENT_LIST_DIR}/worker_codesign.py"
"${CMAKE_INSTALL_PREFIX}"
WORKING_DIRECTORY ${CMAKE_CURRENT_LIST_DIR}
RESULT_VARIABLE exit_code
)
if(NOT exit_code EQUAL "0")
message(FATAL_ERROR "Non-zero exit code of codesign tool")
endif()

View File

@@ -1,74 +0,0 @@
#!/usr/bin/env python3
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Helper script which takes care of signing provided location.
#
# The location can either be a directory (in which case all eligible binaries
# will be signed) or a single file (in which case a single file will be signed).
#
# This script takes care of all the complexity of communicating between process
# which requests file to be signed and the code signing server.
#
# NOTE: Signing happens in-place.
import argparse
import sys
from pathlib import Path
from codesign.simple_code_signer import SimpleCodeSigner
def create_argument_parser():
parser = argparse.ArgumentParser()
parser.add_argument('path_to_sign', type=Path)
return parser
def main():
parser = create_argument_parser()
args = parser.parse_args()
path_to_sign = args.path_to_sign.absolute()
if sys.platform == 'win32':
# When WIX packed is used to generate .msi on Windows the CPack will
# install two different projects and install them to different
# installation prefix:
#
# - C:\b\build\_CPack_Packages\WIX\Blender
# - C:\b\build\_CPack_Packages\WIX\Unspecified
#
# Annoying part is: CMake's post-install script will only be run
# once, with the install prefix which corresponds to a project which
# was installed last. But we want to sign binaries from all projects.
# So in order to do so we detect that we are running for a CPack's
# project used for WIX and force parent directory (which includes both
# projects) to be signed.
#
# Here we force both projects to be signed.
if path_to_sign.name == 'Unspecified' and 'WIX' in str(path_to_sign):
path_to_sign = path_to_sign.parent
code_signer = SimpleCodeSigner()
code_signer.sign_file_or_directory(path_to_sign)
if __name__ == "__main__":
main()

View File

@@ -1,135 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import os
import shutil
import buildbot_utils
def get_cmake_options(builder):
codesign_script = os.path.join(
builder.blender_dir, 'build_files', 'buildbot', 'worker_codesign.cmake')
config_file = "build_files/cmake/config/blender_release.cmake"
options = ['-DCMAKE_BUILD_TYPE:STRING=Release',
'-DWITH_GTESTS=ON']
if builder.platform == 'mac':
options.append('-DCMAKE_OSX_ARCHITECTURES:STRING=x86_64')
options.append('-DCMAKE_OSX_DEPLOYMENT_TARGET=10.9')
elif builder.platform == 'win':
options.extend(['-G', 'Visual Studio 16 2019', '-A', 'x64'])
if builder.codesign:
options.extend(['-DPOSTINSTALL_SCRIPT:PATH=' + codesign_script])
elif builder.platform == 'linux':
config_file = "build_files/buildbot/config/blender_linux.cmake"
optix_sdk_dir = os.path.join(builder.blender_dir, '..', '..', 'NVIDIA-Optix-SDK-7.1')
options.append('-DOPTIX_ROOT_DIR:PATH=' + optix_sdk_dir)
# Workaround to build sm_30 kernels with CUDA 10, since CUDA 11 no longer supports that architecture
if builder.platform == 'win':
options.append('-DCUDA10_TOOLKIT_ROOT_DIR:PATH=C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.1')
options.append('-DCUDA10_NVCC_EXECUTABLE:FILEPATH=C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.1/bin/nvcc.exe')
options.append('-DCUDA11_TOOLKIT_ROOT_DIR:PATH=C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.1')
options.append('-DCUDA11_NVCC_EXECUTABLE:FILEPATH=C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.1/bin/nvcc.exe')
elif builder.platform == 'linux':
options.append('-DCUDA10_TOOLKIT_ROOT_DIR:PATH=/usr/local/cuda-10.1')
options.append('-DCUDA10_NVCC_EXECUTABLE:FILEPATH=/usr/local/cuda-10.1/bin/nvcc')
options.append('-DCUDA11_TOOLKIT_ROOT_DIR:PATH=/usr/local/cuda-11.1')
options.append('-DCUDA11_NVCC_EXECUTABLE:FILEPATH=/usr/local/cuda-11.1/bin/nvcc')
options.append("-C" + os.path.join(builder.blender_dir, config_file))
options.append("-DCMAKE_INSTALL_PREFIX=%s" % (builder.install_dir))
return options
def update_git(builder):
# Do extra git fetch because not all platform/git/buildbot combinations
# update the origin remote, causing buildinfo to detect local changes.
os.chdir(builder.blender_dir)
print("Fetching remotes")
command = ['git', 'fetch', '--all']
buildbot_utils.call(builder.command_prefix + command)
def clean_directories(builder):
# Make sure no garbage remained from the previous run
if os.path.isdir(builder.install_dir):
shutil.rmtree(builder.install_dir)
# Make sure build directory exists and enter it
os.makedirs(builder.build_dir, exist_ok=True)
# Remove buildinfo files to force buildbot to re-generate them.
for buildinfo in ('buildinfo.h', 'buildinfo.h.txt', ):
full_path = os.path.join(builder.build_dir, 'source', 'creator', buildinfo)
if os.path.exists(full_path):
print("Removing {}" . format(buildinfo))
os.remove(full_path)
def cmake_configure(builder):
# CMake configuration
os.chdir(builder.build_dir)
cmake_cache = os.path.join(builder.build_dir, 'CMakeCache.txt')
if os.path.exists(cmake_cache):
print("Removing CMake cache")
os.remove(cmake_cache)
print("CMake configure:")
cmake_options = get_cmake_options(builder)
command = ['cmake', builder.blender_dir] + cmake_options
buildbot_utils.call(builder.command_prefix + command)
def cmake_build(builder):
# CMake build
os.chdir(builder.build_dir)
# NOTE: CPack will build an INSTALL target, which would mean that code
# signing will happen twice when using `make install` and CPack.
# The tricky bit here is that it is not possible to know whether INSTALL
# target is used by CPack or by a buildbot itaself. Extra level on top of
# this is that on Windows it is required to build INSTALL target in order
# to have unit test binaries to run.
# So on the one hand we do an extra unneeded code sign on Windows, but on
# a positive side we don't add complexity and don't make build process more
# fragile trying to avoid this. The signing process is way faster than just
# a clean build of buildbot, especially with regression tests enabled.
if builder.platform == 'win':
command = ['cmake', '--build', '.', '--target', 'install', '--config', 'Release']
else:
command = ['make', '-s', '-j16', 'install']
print("CMake build:")
buildbot_utils.call(builder.command_prefix + command)
if __name__ == "__main__":
builder = buildbot_utils.create_builder_from_arguments()
update_git(builder)
clean_directories(builder)
cmake_configure(builder)
cmake_build(builder)

View File

@@ -1,208 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
# Runs on buildbot worker, creating a release package using the build
# system and zipping it into buildbot_upload.zip. This is then uploaded
# to the master in the next buildbot step.
import os
import sys
from pathlib import Path
import buildbot_utils
def get_package_name(builder, platform=None):
info = buildbot_utils.VersionInfo(builder)
package_name = 'blender-' + info.full_version
if platform:
package_name += '-' + platform
if not (builder.branch == 'master' or builder.is_release_branch):
if info.is_development_build:
package_name = builder.branch + "-" + package_name
return package_name
def sign_file_or_directory(path):
from codesign.simple_code_signer import SimpleCodeSigner
code_signer = SimpleCodeSigner()
code_signer.sign_file_or_directory(Path(path))
def create_buildbot_upload_zip(builder, package_files):
import zipfile
buildbot_upload_zip = os.path.join(builder.upload_dir, "buildbot_upload.zip")
if os.path.exists(buildbot_upload_zip):
os.remove(buildbot_upload_zip)
try:
z = zipfile.ZipFile(buildbot_upload_zip, "w", compression=zipfile.ZIP_STORED)
for filepath, filename in package_files:
print("Packaged", filename)
z.write(filepath, arcname=filename)
z.close()
except Exception as ex:
sys.stderr.write('Create buildbot_upload.zip failed: ' + str(ex) + '\n')
sys.exit(1)
def create_tar_xz(src, dest, package_name):
# One extra to remove leading os.sep when cleaning root for package_root
ln = len(src) + 1
flist = list()
# Create list of tuples containing file and archive name
for root, dirs, files in os.walk(src):
package_root = os.path.join(package_name, root[ln:])
flist.extend([(os.path.join(root, file), os.path.join(package_root, file)) for file in files])
import tarfile
# Set UID/GID of archived files to 0, otherwise they'd be owned by whatever
# user compiled the package. If root then unpacks it to /usr/local/ you get
# a security issue.
def _fakeroot(tarinfo):
tarinfo.gid = 0
tarinfo.gname = "root"
tarinfo.uid = 0
tarinfo.uname = "root"
return tarinfo
package = tarfile.open(dest, 'w:xz', preset=9)
for entry in flist:
package.add(entry[0], entry[1], recursive=False, filter=_fakeroot)
package.close()
def cleanup_files(dirpath, extension):
for f in os.listdir(dirpath):
filepath = os.path.join(dirpath, f)
if os.path.isfile(filepath) and f.endswith(extension):
os.remove(filepath)
def pack_mac(builder):
info = buildbot_utils.VersionInfo(builder)
os.chdir(builder.build_dir)
cleanup_files(builder.build_dir, '.dmg')
package_name = get_package_name(builder, 'macOS')
package_filename = package_name + '.dmg'
package_filepath = os.path.join(builder.build_dir, package_filename)
release_dir = os.path.join(builder.blender_dir, 'release', 'darwin')
buildbot_dir = os.path.join(builder.blender_dir, 'build_files', 'buildbot')
bundle_script = os.path.join(buildbot_dir, 'worker_bundle_dmg.py')
command = [bundle_script]
command += ['--dmg', package_filepath]
if info.is_development_build:
background_image = os.path.join(release_dir, 'buildbot', 'background.tif')
command += ['--background-image', background_image]
if builder.codesign:
command += ['--codesign']
command += [builder.install_dir]
buildbot_utils.call(command)
create_buildbot_upload_zip(builder, [(package_filepath, package_filename)])
def pack_win(builder):
info = buildbot_utils.VersionInfo(builder)
os.chdir(builder.build_dir)
cleanup_files(builder.build_dir, '.zip')
# CPack will add the platform name
cpack_name = get_package_name(builder, None)
package_name = get_package_name(builder, 'windows' + str(builder.bits))
command = ['cmake', '-DCPACK_OVERRIDE_PACKAGENAME:STRING=' + cpack_name, '.']
buildbot_utils.call(builder.command_prefix + command)
command = ['cpack', '-G', 'ZIP']
buildbot_utils.call(builder.command_prefix + command)
package_filename = package_name + '.zip'
package_filepath = os.path.join(builder.build_dir, package_filename)
package_files = [(package_filepath, package_filename)]
if info.version_cycle == 'release':
# Installer only for final release builds, otherwise will get
# 'this product is already installed' messages.
command = ['cpack', '-G', 'WIX']
buildbot_utils.call(builder.command_prefix + command)
package_filename = package_name + '.msi'
package_filepath = os.path.join(builder.build_dir, package_filename)
if builder.codesign:
sign_file_or_directory(package_filepath)
package_files += [(package_filepath, package_filename)]
create_buildbot_upload_zip(builder, package_files)
def pack_linux(builder):
blender_executable = os.path.join(builder.install_dir, 'blender')
info = buildbot_utils.VersionInfo(builder)
# Strip all unused symbols from the binaries
print("Stripping binaries...")
buildbot_utils.call(builder.command_prefix + ['strip', '--strip-all', blender_executable])
print("Stripping python...")
py_target = os.path.join(builder.install_dir, info.short_version)
buildbot_utils.call(
builder.command_prefix + [
'find', py_target, '-iname', '*.so', '-exec', 'strip', '-s', '{}', ';',
],
)
# Construct package name
platform_name = 'linux64'
package_name = get_package_name(builder, platform_name)
package_filename = package_name + ".tar.xz"
print("Creating .tar.xz archive")
package_filepath = builder.install_dir + '.tar.xz'
create_tar_xz(builder.install_dir, package_filepath, package_name)
# Create buildbot_upload.zip
create_buildbot_upload_zip(builder, [(package_filepath, package_filename)])
if __name__ == "__main__":
builder = buildbot_utils.create_builder_from_arguments()
# Make sure install directory always exists
os.makedirs(builder.install_dir, exist_ok=True)
if builder.platform == 'mac':
pack_mac(builder)
elif builder.platform == 'win':
pack_win(builder)
elif builder.platform == 'linux':
pack_linux(builder)

View File

@@ -1,42 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import buildbot_utils
import os
import sys
def get_ctest_arguments(builder):
args = ['--output-on-failure']
if builder.platform == 'win':
args += ['-C', 'Release']
return args
def test(builder):
os.chdir(builder.build_dir)
command = builder.command_prefix + ['ctest'] + get_ctest_arguments(builder)
buildbot_utils.call(command)
if __name__ == "__main__":
builder = buildbot_utils.create_builder_from_arguments()
test(builder)

View File

@@ -1,31 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# <pep8 compliant>
import buildbot_utils
import os
import sys
if __name__ == "__main__":
builder = buildbot_utils.create_builder_from_arguments()
os.chdir(builder.blender_dir)
# Run make update which handles all libraries and submodules.
make_update = os.path.join(builder.blender_dir, "build_files", "utils", "make_update.py")
buildbot_utils.call([sys.executable, make_update, '--no-blender', "--use-tests", "--use-centos-libraries"])

View File

@@ -79,7 +79,7 @@ if(EXISTS ${SOURCE_DIR}/.git)
ERROR_QUIET)
if(NOT _git_below_check STREQUAL "")
# If there're commits between HEAD and upstream this means
# that we're reset-ed to older revision. Use it's hash then.
# that we're reset-ed to older revision. Use its hash then.
execute_process(COMMAND git rev-parse --short=12 HEAD
WORKING_DIRECTORY ${SOURCE_DIR}
OUTPUT_VARIABLE MY_WC_HASH

View File

@@ -56,10 +56,6 @@ set(WITH_TBB ON CACHE BOOL "" FORCE)
set(WITH_USD ON CACHE BOOL "" FORCE)
set(WITH_MEM_JEMALLOC ON CACHE BOOL "" FORCE)
set(WITH_CYCLES_CUDA_BINARIES ON CACHE BOOL "" FORCE)
set(WITH_CYCLES_CUBIN_COMPILER OFF CACHE BOOL "" FORCE)
set(CYCLES_CUDA_BINARIES_ARCH sm_30;sm_35;sm_37;sm_50;sm_52;sm_60;sm_61;sm_70;sm_75;sm_86;compute_75 CACHE STRING "" FORCE)
set(WITH_CYCLES_DEVICE_OPTIX ON CACHE BOOL "" FORCE)
# platform dependent options
if(APPLE)
@@ -80,4 +76,8 @@ if(UNIX AND NOT APPLE)
endif()
if(NOT APPLE)
set(WITH_XR_OPENXR ON CACHE BOOL "" FORCE)
set(WITH_CYCLES_DEVICE_OPTIX ON CACHE BOOL "" FORCE)
set(WITH_CYCLES_CUDA_BINARIES ON CACHE BOOL "" FORCE)
set(WITH_CYCLES_CUBIN_COMPILER OFF CACHE BOOL "" FORCE)
endif()

View File

@@ -104,8 +104,8 @@ if(WIN32)
set(CPACK_WIX_LIGHT_EXTRA_FLAGS -dcl:medium)
endif()
set(CPACK_PACKAGE_EXECUTABLES "blender" "blender")
set(CPACK_CREATE_DESKTOP_LINKS "blender" "blender")
set(CPACK_PACKAGE_EXECUTABLES "blender-launcher" "blender")
set(CPACK_CREATE_DESKTOP_LINKS "blender-launcher" "blender")
include(CPack)

View File

@@ -20,12 +20,6 @@
# Libraries configuration for Apple.
if("${CMAKE_OSX_ARCHITECTURES}" STREQUAL "arm64")
set(MACOSX_DEPLOYMENT_TARGET 11.00)
else()
set(MACOSX_DEPLOYMENT_TARGET 10.13)
endif()
macro(find_package_wrapper)
# do nothing, just satisfy the macro
endmacro()

View File

@@ -168,21 +168,15 @@ endif()
unset(OSX_SDKROOT)
# 10.13 is our min. target, if you use higher sdk, weak linking happens
if("${CMAKE_OSX_ARCHITECTURES}" STREQUAL "arm64")
# M1 chips run Big Sur onwards.
set(OSX_MIN_DEPLOYMENT_TARGET 11.00)
else()
# 10.13 is our min. target, if you use higher sdk, weak linking happens
set(OSX_MIN_DEPLOYMENT_TARGET 10.13)
endif()
if(CMAKE_OSX_DEPLOYMENT_TARGET)
if(${CMAKE_OSX_DEPLOYMENT_TARGET} VERSION_LESS ${OSX_MIN_DEPLOYMENT_TARGET})
message(STATUS "Setting deployment target to ${OSX_MIN_DEPLOYMENT_TARGET}, lower versions are not supported")
set(CMAKE_OSX_DEPLOYMENT_TARGET "${OSX_MIN_DEPLOYMENT_TARGET}" CACHE STRING "" FORCE)
endif()
else()
set(CMAKE_OSX_DEPLOYMENT_TARGET "${OSX_MIN_DEPLOYMENT_TARGET}" CACHE STRING "" FORCE)
endif()
set(CMAKE_OSX_DEPLOYMENT_TARGET "${OSX_MIN_DEPLOYMENT_TARGET}" CACHE STRING "" FORCE)
if(NOT ${CMAKE_GENERATOR} MATCHES "Xcode")
# Force CMAKE_OSX_DEPLOYMENT_TARGET for makefiles, will not work else (CMake bug?)

View File

@@ -119,7 +119,7 @@ string(APPEND CMAKE_MODULE_LINKER_FLAGS " /SAFESEH:NO /ignore:4099")
list(APPEND PLATFORM_LINKLIBS
ws2_32 vfw32 winmm kernel32 user32 gdi32 comdlg32 Comctl32 version
advapi32 shfolder shell32 ole32 oleaut32 uuid psapi Dbghelp Shlwapi
pathcch
pathcch Shcore
)
if(WITH_INPUT_IME)
@@ -144,8 +144,8 @@ add_definitions(-D_ALLOW_KEYWORD_MACROS)
# that both /GR and /GR- are specified.
remove_cc_flag("/GR")
# We want to support Windows 7 level ABI
add_definitions(-D_WIN32_WINNT=0x601)
# Make the Windows 8.1 API available for use.
add_definitions(-D_WIN32_WINNT=0x603)
include(build_files/cmake/platform/platform_win32_bundle_crt.cmake)
remove_cc_flag("/MDd" "/MD" "/Zi")

View File

@@ -0,0 +1,8 @@
Pipeline Config
===============
This configuration file is used by buildbot new pipeline for the `update-code` step.
It will soon be used by the ../utils/make_update.py script.
Both buildbot and developers will eventually use the same configuration file.

View File

@@ -0,0 +1,87 @@
{
"update-code":
{
"git" :
{
"submodules":
[
{ "path": "release/scripts/addons", "branch": "master", "commit_id": "HEAD" },
{ "path": "release/scripts/addons_contrib", "branch": "master", "commit_id": "HEAD" },
{ "path": "release/datafiles/locale", "branch": "master", "commit_id": "HEAD" },
{ "path": "source/tools", "branch": "master", "commit_id": "HEAD" }
]
},
"svn":
{
"tests": { "path": "lib/tests", "branch": "trunk", "commit_id": "HEAD" },
"libraries":
{
"darwin-x86_64": { "path": "lib/darwin", "branch": "trunk", "commit_id": "HEAD" },
"darwin-arm64": { "path": "lib/darwin_arm64", "branch": "trunk", "commit_id": "HEAD" },
"linux-x86_64": { "path": "lib/linux_centos7_x86_64", "branch": "trunk", "commit_id": "HEAD" },
"windows-amd64": { "path": "lib/win64_vc15", "branch": "trunk", "commit_id": "HEAD" }
}
}
},
"buildbot":
{
"gcc":
{
"version": "9.0"
},
"sdks":
{
"optix":
{
"version": "7.1.0"
},
"cuda10":
{
"version": "10.1"
},
"cuda11":
{
"version": "11.3"
}
},
"cmake":
{
"default":
{
"version": "any",
"overrides":
{
}
},
"darwin-x86_64":
{
"overrides":
{
}
},
"darwin-arm64":
{
"overrides":
{
}
},
"linux-x86_64":
{
"overrides":
{
}
},
"windows-amd64":
{
"overrides":
{
}
}
}
}
}

View File

@@ -0,0 +1,5 @@
Make Utility Scripts
====================
Scripts used only by developers for now

View File

@@ -85,7 +85,7 @@ def openBlendFile(filename):
'''
handle = open(filename, 'rb')
magic = ReadString(handle, 7)
if magic in ("BLENDER", "BULLETf"):
if magic in {"BLENDER", "BULLETf"}:
log.debug("normal blendfile detected")
handle.seek(0, os.SEEK_SET)
return handle
@@ -137,7 +137,7 @@ class BlendFile:
fileblock = BlendFileBlock(handle, self)
found_dna_block = False
while not found_dna_block:
if fileblock.Header.Code in ("DNA1", "SDNA"):
if fileblock.Header.Code in {"DNA1", "SDNA"}:
self.Catalog = DNACatalog(self.Header, handle)
found_dna_block = True
else:

View File

@@ -4,7 +4,9 @@ Simple Render Engine
"""
import bpy
import bgl
import array
import gpu
from gpu_extras.presets import draw_texture_2d
class CustomRenderEngine(bpy.types.RenderEngine):
@@ -100,8 +102,7 @@ class CustomRenderEngine(bpy.types.RenderEngine):
dimensions = region.width, region.height
# Bind shader that converts from scene linear to display space,
bgl.glEnable(bgl.GL_BLEND)
bgl.glBlendFunc(bgl.GL_ONE, bgl.GL_ONE_MINUS_SRC_ALPHA)
gpu.state.blend_set('ALPHA_PREMULT')
self.bind_display_space_shader(scene)
if not self.draw_data or self.draw_data.dimensions != dimensions:
@@ -110,7 +111,7 @@ class CustomRenderEngine(bpy.types.RenderEngine):
self.draw_data.draw()
self.unbind_display_space_shader()
bgl.glDisable(bgl.GL_BLEND)
gpu.state.blend_set('NONE')
class CustomDrawData:
@@ -119,68 +120,21 @@ class CustomDrawData:
self.dimensions = dimensions
width, height = dimensions
pixels = [0.1, 0.2, 0.1, 1.0] * width * height
pixels = bgl.Buffer(bgl.GL_FLOAT, width * height * 4, pixels)
pixels = width * height * array.array('f', [0.1, 0.2, 0.1, 1.0])
pixels = gpu.types.Buffer('FLOAT', width * height * 4, pixels)
# Generate texture
self.texture = bgl.Buffer(bgl.GL_INT, 1)
bgl.glGenTextures(1, self.texture)
bgl.glActiveTexture(bgl.GL_TEXTURE0)
bgl.glBindTexture(bgl.GL_TEXTURE_2D, self.texture[0])
bgl.glTexImage2D(bgl.GL_TEXTURE_2D, 0, bgl.GL_RGBA16F, width, height, 0, bgl.GL_RGBA, bgl.GL_FLOAT, pixels)
bgl.glTexParameteri(bgl.GL_TEXTURE_2D, bgl.GL_TEXTURE_MIN_FILTER, bgl.GL_LINEAR)
bgl.glTexParameteri(bgl.GL_TEXTURE_2D, bgl.GL_TEXTURE_MAG_FILTER, bgl.GL_LINEAR)
bgl.glBindTexture(bgl.GL_TEXTURE_2D, 0)
self.texture = gpu.types.GPUTexture((width, height), format='RGBA16F', data=pixels)
# Bind shader that converts from scene linear to display space,
# use the scene's color management settings.
shader_program = bgl.Buffer(bgl.GL_INT, 1)
bgl.glGetIntegerv(bgl.GL_CURRENT_PROGRAM, shader_program)
# Generate vertex array
self.vertex_array = bgl.Buffer(bgl.GL_INT, 1)
bgl.glGenVertexArrays(1, self.vertex_array)
bgl.glBindVertexArray(self.vertex_array[0])
texturecoord_location = bgl.glGetAttribLocation(shader_program[0], "texCoord")
position_location = bgl.glGetAttribLocation(shader_program[0], "pos")
bgl.glEnableVertexAttribArray(texturecoord_location)
bgl.glEnableVertexAttribArray(position_location)
# Generate geometry buffers for drawing textured quad
position = [0.0, 0.0, width, 0.0, width, height, 0.0, height]
position = bgl.Buffer(bgl.GL_FLOAT, len(position), position)
texcoord = [0.0, 0.0, 1.0, 0.0, 1.0, 1.0, 0.0, 1.0]
texcoord = bgl.Buffer(bgl.GL_FLOAT, len(texcoord), texcoord)
self.vertex_buffer = bgl.Buffer(bgl.GL_INT, 2)
bgl.glGenBuffers(2, self.vertex_buffer)
bgl.glBindBuffer(bgl.GL_ARRAY_BUFFER, self.vertex_buffer[0])
bgl.glBufferData(bgl.GL_ARRAY_BUFFER, 32, position, bgl.GL_STATIC_DRAW)
bgl.glVertexAttribPointer(position_location, 2, bgl.GL_FLOAT, bgl.GL_FALSE, 0, None)
bgl.glBindBuffer(bgl.GL_ARRAY_BUFFER, self.vertex_buffer[1])
bgl.glBufferData(bgl.GL_ARRAY_BUFFER, 32, texcoord, bgl.GL_STATIC_DRAW)
bgl.glVertexAttribPointer(texturecoord_location, 2, bgl.GL_FLOAT, bgl.GL_FALSE, 0, None)
bgl.glBindBuffer(bgl.GL_ARRAY_BUFFER, 0)
bgl.glBindVertexArray(0)
# Note: This is just a didactic example.
# In this case it would be more convenient to fill the texture with:
# self.texture.clear('FLOAT', value=[0.1, 0.2, 0.1, 1.0])
def __del__(self):
bgl.glDeleteBuffers(2, self.vertex_buffer)
bgl.glDeleteVertexArrays(1, self.vertex_array)
bgl.glBindTexture(bgl.GL_TEXTURE_2D, 0)
bgl.glDeleteTextures(1, self.texture)
del self.texture
def draw(self):
bgl.glActiveTexture(bgl.GL_TEXTURE0)
bgl.glBindTexture(bgl.GL_TEXTURE_2D, self.texture[0])
bgl.glBindVertexArray(self.vertex_array[0])
bgl.glDrawArrays(bgl.GL_TRIANGLE_FAN, 0, 4)
bgl.glBindVertexArray(0)
bgl.glBindTexture(bgl.GL_TEXTURE_2D, 0)
draw_texture_2d(self.texture, (0, 0), self.texture.width, self.texture.height)
# RenderEngines also need to tell UI Panels that they are compatible with.

View File

@@ -0,0 +1,5 @@
# Compute local object transformation matrix:
if obj.rotation_mode == 'QUATERNION':
matrix = mathutils.Matrix.LocRotScale(obj.location, obj.rotation_quaternion, obj.scale)
else:
matrix = mathutils.Matrix.LocRotScale(obj.location, obj.rotation_euler, obj.scale)

View File

@@ -14,10 +14,14 @@ mat_rot = mathutils.Matrix.Rotation(math.radians(45.0), 4, 'X')
mat_out = mat_loc @ mat_rot @ mat_sca
print(mat_out)
# extract components back out of the matrix
# extract components back out of the matrix as two vectors and a quaternion
loc, rot, sca = mat_out.decompose()
print(loc, rot, sca)
# recombine extracted components
mat_out2 = mathutils.Matrix.LocRotScale(loc, rot, sca)
print(mat_out2)
# it can also be useful to access components of a matrix directly
mat = mathutils.Matrix()
mat[0][0], mat[1][0], mat[2][0] = 0.0, 1.0, 2.0

View File

@@ -545,6 +545,13 @@ def range_str(val):
def example_extract_docstring(filepath):
'''
Return (text, line_no, line_no_has_content) where:
- ``text`` is the doc-string text.
- ``line_no`` is the line the doc-string text ends.
- ``line_no_has_content`` when False, this file only contains a doc-string.
There is no need to include the remainder.
'''
file = open(filepath, "r", encoding="utf-8")
line = file.readline()
line_no = 0
@@ -553,7 +560,7 @@ def example_extract_docstring(filepath):
line_no += 1
else:
file.close()
return "", 0, False
return "", 0, True
for line in file:
line_no += 1
@@ -947,7 +954,7 @@ def pymodule2sphinx(basepath, module_name, module, title, module_all_extra):
# constant, not much fun we can do here except to list it.
# TODO, figure out some way to document these!
fw(".. data:: %s\n\n" % attribute)
write_indented_lines(" ", fw, "constant value %s" % repr(value), False)
write_indented_lines(" ", fw, "Constant value %s" % repr(value), False)
fw("\n")
else:
BPY_LOGGER.debug("\tnot documenting %s.%s of %r type" % (module_name, attribute, value_type.__name__))
@@ -1239,7 +1246,7 @@ def pyrna_enum2sphinx(prop, use_empty_descriptions=False):
"%s.\n" % (
identifier,
# Account for multi-line enum descriptions, allowing this to be a block of text.
indent(", ".join(escape_rst(val) for val in (name, description) if val) or "Undocumented", " "),
indent(" -- ".join(escape_rst(val) for val in (name, description) if val) or "Undocumented", " "),
)
for identifier, name, description in prop.enum_items
])

View File

@@ -75,6 +75,7 @@ void FFMPEGWriter::encode()
m_frame->nb_samples = m_input_samples;
m_frame->format = m_codecCtx->sample_fmt;
m_frame->channel_layout = m_codecCtx->channel_layout;
m_frame->channels = m_specs.channels;
if(avcodec_fill_audio_frame(m_frame, m_specs.channels, m_codecCtx->sample_fmt, reinterpret_cast<data_t*>(data), m_input_buffer.getSize(), 0) < 0)
AUD_THROW(FileException, "File couldn't be written, filling the audio frame failed with ffmpeg.");

View File

@@ -29,10 +29,10 @@
#if OPENVDB == 1
# include "openvdb/openvdb.h"
# include <openvdb/points/PointConversion.h>
# include <openvdb/points/PointCount.h>
# include <openvdb/tools/Clip.h>
# include <openvdb/tools/Dense.h>
# include "openvdb/points/PointConversion.h"
# include "openvdb/points/PointCount.h"
# include "openvdb/tools/Clip.h"
# include "openvdb/tools/Dense.h"
#endif
#define POSITION_NAME "P"
@@ -519,7 +519,7 @@ int writeObjectsVDB(const string &filename,
}
}
// Write only if the is at least one grid, optionally write with compression.
// Write only if there is at least one grid, optionally write with compression.
if (gridsVDB.size()) {
int vdb_flags = openvdb::io::COMPRESS_ACTIVE_MASK;
switch (compression) {
@@ -534,7 +534,8 @@ int writeObjectsVDB(const string &filename,
}
case COMPRESSION_BLOSC: {
# if OPENVDB_BLOSC == 1
vdb_flags |= openvdb::io::COMPRESS_BLOSC;
// Cannot use |= here, causes segfault with blosc 1.5.0 (== recommended version)
vdb_flags = openvdb::io::COMPRESS_BLOSC;
# else
debMsg("OpenVDB was built without Blosc support, using Zip compression instead", 1);
vdb_flags |= openvdb::io::COMPRESS_ZIP;

View File

@@ -384,6 +384,7 @@ class FluidSolver : public PbClass {
GridStorage<Real> mGrids4dReal;
GridStorage<Vec3> mGrids4dVec;
GridStorage<Vec4> mGrids4dVec4;
public:
PbArgs _args;
}

View File

@@ -42,7 +42,7 @@ inline void updateQtGui(bool full, int frame, float time, const std::string &cur
# ifdef _DEBUG
# define DEBUG 1
# endif // _DEBUG
#endif // DEBUG
#endif // DEBUG
// Standard exception
class Error : public std::exception {

View File

@@ -1,3 +1,3 @@
#define MANTA_GIT_VERSION "commit 39b7a415721ecbf6643612a24e8eadd221aeb934"
#define MANTA_GIT_VERSION "commit 9c505cd22e289b98c9aa717efba8ef3201c7e458"

View File

@@ -389,6 +389,7 @@ class GridBase : public PbClass {
Real mDx;
bool m3D; // precomputed Z shift: to ensure 2D compatibility, always use this instead of sx*sy !
IndexInt mStrideZ;
public:
PbArgs _args;
}

View File

@@ -326,6 +326,7 @@ class Grid4dBase : public PbClass {
// precomputed Z,T shift: to ensure 2D compatibility, always use this instead of sx*sy !
IndexInt mStrideZ;
IndexInt mStrideT;
public:
PbArgs _args;
}
@@ -950,6 +951,7 @@ template<class T> class Grid4d : public Grid4dBase {
protected:
T *mData;
public:
PbArgs _args;
}

View File

@@ -266,6 +266,7 @@ class LevelsetGrid : public Grid<Real> {
}
static Real invalidTimeValue();
public:
PbArgs _args;
}

View File

@@ -796,6 +796,7 @@ class Mesh : public PbClass {
std::vector<MeshDataImpl<int> *>
mMdataInt; //! indicate that mdata of this mesh is copied, and needs to be freed
bool mFreeMdata;
public:
PbArgs _args;
}
@@ -881,6 +882,7 @@ class MeshDataBase : public PbClass {
protected:
Mesh *mMesh;
public:
PbArgs _args;
}
@@ -1645,6 +1647,7 @@ template<class T> class MeshDataImpl : public MeshDataBase {
//! optionally , we might have an associated grid from which to grab new data
Grid<T> *mpGridSource; //! unfortunately , we need to distinguish mac vs regular vec3
bool mGridSourceMAC;
public:
PbArgs _args;
}

View File

@@ -154,6 +154,7 @@ class MovingObstacle : public PbClass {
int mEmptyType;
int mID;
static int sIDcnt;
public:
PbArgs _args;
}

View File

@@ -236,6 +236,7 @@ class WaveletNoiseField : public PbClass {
static int randomSeed;
// global reference count for noise tile
static std::atomic<int> mNoiseReferenceCount;
public:
PbArgs _args;
}

View File

@@ -205,6 +205,7 @@ class ParticleBase : public PbClass {
//! custom seed for particle systems, used by plugins
int mSeed; //! fix global random seed storage, used mainly by functions in this class
static int globalSeed;
public:
PbArgs _args;
}
@@ -628,6 +629,7 @@ template<class S> class ParticleSystem : public ParticleBase {
std::vector<S> mData;
//! reduce storage , called by doCompress
virtual void compress();
public:
PbArgs _args;
}
@@ -918,6 +920,7 @@ class ParticleIndexSystem : public ParticleSystem<ParticleIndexData> {
return -1;
}
};
public:
PbArgs _args;
}
@@ -982,6 +985,7 @@ template<class DATA, class CON> class ConnectedParticleSystem : public ParticleS
protected:
std::vector<CON> mSegments;
virtual void compress();
public:
PbArgs _args;
}
@@ -1071,6 +1075,7 @@ class ParticleDataBase : public PbClass {
protected:
ParticleBase *mpParticleSys;
public:
PbArgs _args;
}
@@ -1843,6 +1848,7 @@ template<class T> class ParticleDataImpl : public ParticleDataBase {
//! optionally , we might have an associated grid from which to grab new data
Grid<T> *mpGridSource; //! unfortunately , we need to distinguish mac vs regular vec3
bool mGridSourceMAC;
public:
PbArgs _args;
}

View File

@@ -234,10 +234,10 @@ void subdivideMesh(
normalize(ne2);
// Real thisArea = sqrMag(cross(-e2,e0));
// small angle approximation says sin(x) = arcsin(x) = x,
// arccos(x) = pi/2 - arcsin(x),
// cos(x) = dot(A,B),
// so angle is approximately 1 - dot(A,B).
// small angle approximation says sin(x) = arcsin(x) = x,
// arccos(x) = pi/2 - arcsin(x),
// cos(x) = dot(A,B),
// so angle is approximately 1 - dot(A,B).
Real angle[3];
angle[0] = 1.0 - dot(ne0, -ne2);
angle[1] = 1.0 - dot(ne1, -ne0);

View File

@@ -2287,9 +2287,10 @@ struct knFlipComputePotentialTrappedAir : public KernelBase {
const Vec3 &vj = scaleFromManta * v.getCentered(x, y, z);
const Vec3 xij = xi - xj;
const Vec3 vij = vi - vj;
Real h = !pot.is3D() ? 1.414 * radius :
1.732 * radius; // estimate sqrt(2)*radius resp. sqrt(3)*radius
// for h, due to squared resp. cubic neighbor area
Real h = !pot.is3D() ?
1.414 * radius :
1.732 * radius; // estimate sqrt(2)*radius resp. sqrt(3)*radius for h, due
// to squared resp. cubic neighbor area
vdiff += norm(vij) * (1 - dot(getNormalized(vij), getNormalized(xij))) *
(1 - norm(xij) / h);
}

View File

@@ -269,6 +269,7 @@ class Shape : public PbClass {
protected:
GridType mType;
public:
PbArgs _args;
}
@@ -319,6 +320,7 @@ class NullShape : public Shape {
{
gridSetConst<Real>(phi, 1000.0f);
}
public:
PbArgs _args;
}
@@ -394,6 +396,7 @@ class Box : public Shape {
protected:
Vec3 mP0, mP1;
public:
PbArgs _args;
}
@@ -455,6 +458,7 @@ class Sphere : public Shape {
protected:
Vec3 mCenter, mScale;
Real mRadius;
public:
PbArgs _args;
}
@@ -579,6 +583,7 @@ class Cylinder : public Shape {
protected:
Vec3 mCenter, mZDir;
Real mRadius, mZ;
public:
PbArgs _args;
}
@@ -655,6 +660,7 @@ class Slope : public Shape {
Real mAnglexy, mAngleyz;
Real mOrigin;
Vec3 mGs;
public:
PbArgs _args;
}

View File

@@ -199,6 +199,7 @@ class TurbulenceParticleSystem : public ParticleSystem<TurbulenceParticleData> {
private:
WaveletNoiseField &noise;
public:
PbArgs _args;
}

View File

@@ -127,6 +127,7 @@ class VortexParticleSystem : public ParticleSystem<VortexParticleData> {
}
virtual ParticleBase *clone();
public:
PbArgs _args;
}

View File

@@ -240,6 +240,7 @@ class VortexSheetMesh : public Mesh {
VorticityChannel mVorticity;
TexCoord3Channel mTex1, mTex2;
TurbulenceChannel mTurb;
public:
PbArgs _args;
}

View File

@@ -322,7 +322,9 @@ static bool clg_ctx_filter_check(CLogContext *ctx, const char *identifier)
if (flt->match[0] == '*' && flt->match[len - 1] == '*') {
char *match = MEM_callocN(sizeof(char) * len - 1, __func__);
memcpy(match, flt->match + 1, len - 2);
if (strstr(identifier, match) != NULL) {
const bool success = (strstr(identifier, match) != NULL);
MEM_freeN(match);
if (success) {
return (bool)i;
}
}

View File

@@ -19,16 +19,16 @@ from __future__ import annotations
def _is_using_buggy_driver():
import bgl
import gpu
# We need to be conservative here because in multi-GPU systems display card
# might be quite old, but others one might be just good.
#
# So We shouldn't disable possible good dedicated cards just because display
# card seems weak. And instead we only blacklist configurations which are
# proven to cause problems.
if bgl.glGetString(bgl.GL_VENDOR) == "ATI Technologies Inc.":
if gpu.platform.vendor_get() == "ATI Technologies Inc.":
import re
version = bgl.glGetString(bgl.GL_VERSION)
version = gpu.platform.version_get()
if version.endswith("Compatibility Profile Context"):
# Old HD 4xxx and 5xxx series drivers did not have driver version
# in the version string, but those cards do not quite work and

View File

@@ -725,7 +725,7 @@ class CYCLES_RENDER_PT_performance_tiles(CyclesButtonsPanel, Panel):
col.prop(cscene, "tile_order", text="Order")
sub = col.column()
sub.active = not rd.use_save_buffers
sub.active = not rd.use_save_buffers and not cscene.use_adaptive_sampling
sub.prop(cscene, "use_progressive_refine")

View File

@@ -83,6 +83,8 @@ struct BlenderCamera {
BoundBox2D pano_viewplane;
BoundBox2D viewport_camera_border;
float passepartout_alpha;
Transform matrix;
float offscreen_dicing_scale;
@@ -125,6 +127,7 @@ static void blender_camera_init(BlenderCamera *bcam, BL::RenderSettings &b_rende
bcam->pano_viewplane.top = 1.0f;
bcam->viewport_camera_border.right = 1.0f;
bcam->viewport_camera_border.top = 1.0f;
bcam->passepartout_alpha = 0.5f;
bcam->offscreen_dicing_scale = 1.0f;
bcam->matrix = transform_identity();
@@ -212,6 +215,8 @@ static void blender_camera_from_object(BlenderCamera *bcam,
bcam->lens = b_camera.lens();
bcam->passepartout_alpha = b_camera.show_passepartout() ? b_camera.passepartout_alpha() : 0.0f;
if (b_camera.dof().use_dof()) {
/* allow f/stop number to change aperture_size but still
* give manual control over aperture radius */
@@ -834,15 +839,19 @@ static void blender_camera_border(BlenderCamera *bcam,
full_border,
&bcam->viewport_camera_border);
if (!b_render.use_border()) {
if (b_render.use_border()) {
bcam->border.left = b_render.border_min_x();
bcam->border.right = b_render.border_max_x();
bcam->border.bottom = b_render.border_min_y();
bcam->border.top = b_render.border_max_y();
}
else if (bcam->passepartout_alpha == 1.0f) {
bcam->border = full_border;
}
else {
return;
}
bcam->border.left = b_render.border_min_x();
bcam->border.right = b_render.border_max_x();
bcam->border.bottom = b_render.border_min_y();
bcam->border.top = b_render.border_max_y();
/* Determine viewport subset matching camera border. */
blender_camera_border_subset(b_engine,
b_render,
@@ -885,8 +894,7 @@ void BlenderSync::sync_view(BL::SpaceView3D &b_v3d,
}
}
BufferParams BlenderSync::get_buffer_params(BL::RenderSettings &b_render,
BL::SpaceView3D &b_v3d,
BufferParams BlenderSync::get_buffer_params(BL::SpaceView3D &b_v3d,
BL::RegionView3D &b_rv3d,
Camera *cam,
int width,
@@ -902,7 +910,8 @@ BufferParams BlenderSync::get_buffer_params(BL::RenderSettings &b_render,
if (b_v3d && b_rv3d && b_rv3d.view_perspective() != BL::RegionView3D::view_perspective_CAMERA)
use_border = b_v3d.use_render_border();
else
use_border = b_render.use_border();
/* the camera can always have a passepartout */
use_border = true;
if (use_border) {
/* border render */

View File

@@ -261,10 +261,8 @@ Object *BlenderSync::sync_object(BL::Depsgraph &b_depsgraph,
}
/* test if we need to sync */
bool object_updated = false;
if (object_map.add_or_update(&object, b_ob, b_parent, key))
object_updated = true;
bool object_updated = object_map.add_or_update(&object, b_ob, b_parent, key) ||
(tfm != object->get_tfm());
/* mesh sync */
/* b_ob is owned by the iterator and will go out of scope at the end of the block.
@@ -313,8 +311,7 @@ Object *BlenderSync::sync_object(BL::Depsgraph &b_depsgraph,
* transform comparison should not be needed, but duplis don't work perfect
* in the depsgraph and may not signal changes, so this is a workaround */
if (object->is_modified() || object_updated ||
(object->get_geometry() && object->get_geometry()->is_modified()) ||
tfm != object->get_tfm()) {
(object->get_geometry() && object->get_geometry()->is_modified())) {
object->name = b_ob.name().c_str();
object->set_pass_id(b_ob.pass_index());
object->set_color(get_float3(b_ob.color()));

View File

@@ -155,7 +155,7 @@ void BlenderSession::create_session()
/* set buffer parameters */
BufferParams buffer_params = BlenderSync::get_buffer_params(
b_render, b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
session->reset(buffer_params, session_params.samples);
b_engine.use_highlight_tiles(session_params.progressive_refine == false);
@@ -242,8 +242,7 @@ void BlenderSession::reset_session(BL::BlendData &b_data, BL::Depsgraph &b_depsg
BL::SpaceView3D b_null_space_view3d(PointerRNA_NULL);
BL::RegionView3D b_null_region_view3d(PointerRNA_NULL);
BufferParams buffer_params = BlenderSync::get_buffer_params(b_render,
b_null_space_view3d,
BufferParams buffer_params = BlenderSync::get_buffer_params(b_null_space_view3d,
b_null_region_view3d,
scene->camera,
width,
@@ -486,7 +485,7 @@ void BlenderSession::render(BL::Depsgraph &b_depsgraph_)
SessionParams session_params = BlenderSync::get_session_params(
b_engine, b_userpref, b_scene, background, b_view_layer);
BufferParams buffer_params = BlenderSync::get_buffer_params(
b_render, b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
/* temporary render result to find needed passes and views */
BL::RenderResult b_rr = begin_render_result(
@@ -810,7 +809,7 @@ void BlenderSession::synchronize(BL::Depsgraph &b_depsgraph_)
/* get buffer parameters */
BufferParams buffer_params = BlenderSync::get_buffer_params(
b_render, b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
if (!buffer_params.denoising_data_pass) {
session_params.denoising.use = false;
@@ -889,7 +888,7 @@ bool BlenderSession::draw(int w, int h)
SessionParams session_params = BlenderSync::get_session_params(
b_engine, b_userpref, b_scene, background);
BufferParams buffer_params = BlenderSync::get_buffer_params(
b_render, b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
b_v3d, b_rv3d, scene->camera, width, height, session_params.denoising.use);
bool session_pause = BlenderSync::get_session_pause(b_scene, background);
if (session_pause == false) {
@@ -907,7 +906,7 @@ bool BlenderSession::draw(int w, int h)
/* draw */
BufferParams buffer_params = BlenderSync::get_buffer_params(
b_render, b_v3d, b_rv3d, scene->camera, width, height, session->params.denoising.use);
b_v3d, b_rv3d, scene->camera, width, height, session->params.denoising.use);
DeviceDrawParams draw_params;
if (session->params.display_buffer_linear) {

View File

@@ -1373,7 +1373,7 @@ void BlenderSync::sync_world(BL::Depsgraph &b_depsgraph, BL::SpaceView3D &b_v3d,
BlenderViewportParameters new_viewport_parameters(b_v3d);
if (world_recalc || update_all || b_world.ptr.data != world_map ||
viewport_parameters.modified(new_viewport_parameters)) {
viewport_parameters.shader_modified(new_viewport_parameters)) {
Shader *shader = scene->default_background;
ShaderGraph *graph = new ShaderGraph();
@@ -1501,8 +1501,8 @@ void BlenderSync::sync_world(BL::Depsgraph &b_depsgraph, BL::SpaceView3D &b_v3d,
background->set_transparent_roughness_threshold(0.0f);
}
background->set_use_shader(view_layer.use_background_shader |
viewport_parameters.custom_viewport_parameters());
background->set_use_shader(view_layer.use_background_shader ||
viewport_parameters.use_custom_shader());
background->set_use_ao(background->get_use_ao() && view_layer.use_background_ao);
background->tag_update(scene);
@@ -1553,13 +1553,9 @@ void BlenderSync::sync_lights(BL::Depsgraph &b_depsgraph, bool update_all)
void BlenderSync::sync_shaders(BL::Depsgraph &b_depsgraph, BL::SpaceView3D &b_v3d)
{
/* for auto refresh images */
bool auto_refresh_update = false;
if (preview) {
ImageManager *image_manager = scene->image_manager;
int frame = b_scene.frame_current();
auto_refresh_update = image_manager->set_animation_frame_update(frame);
}
ImageManager *image_manager = scene->image_manager;
const int frame = b_scene.frame_current();
const bool auto_refresh_update = image_manager->set_animation_frame_update(frame);
shader_map.pre_sync();

View File

@@ -69,7 +69,8 @@ BlenderSync::BlenderSync(BL::RenderEngine &b_engine,
experimental(false),
dicing_rate(1.0f),
max_subdivisions(12),
progress(progress)
progress(progress),
has_updates_(true)
{
PointerRNA cscene = RNA_pointer_get(&b_scene.ptr, "cycles");
dicing_rate = preview ? RNA_float_get(&cscene, "preview_dicing_rate") :
@@ -84,7 +85,9 @@ BlenderSync::~BlenderSync()
void BlenderSync::reset(BL::BlendData &b_data, BL::Scene &b_scene)
{
/* Update data and scene pointers in case they change in session reset,
* for example after undo. */
* for example after undo.
* Note that we do not modify the `has_updates_` flag here because the sync
* reset is also used during viewport navigation. */
this->b_data = b_data;
this->b_scene = b_scene;
}
@@ -117,6 +120,8 @@ void BlenderSync::sync_recalc(BL::Depsgraph &b_depsgraph, BL::SpaceView3D &b_v3d
}
if (dicing_prop_changed) {
has_updates_ = true;
for (const pair<const GeometryKey, Geometry *> &iter : geometry_map.key_to_scene_data()) {
Geometry *geom = iter.second;
if (geom->is_mesh()) {
@@ -133,6 +138,12 @@ void BlenderSync::sync_recalc(BL::Depsgraph &b_depsgraph, BL::SpaceView3D &b_v3d
/* Iterate over all IDs in this depsgraph. */
for (BL::DepsgraphUpdate &b_update : b_depsgraph.updates) {
/* TODO(sergey): Can do more selective filter here. For example, ignore changes made to
* screen datablock. Note that sync_data() needs to be called after object deletion, and
* currently this is ensured by the scene ID tagged for update, which sets the `has_updates_`
* flag. */
has_updates_ = true;
BL::ID b_id(b_update.id());
/* Material */
@@ -213,9 +224,13 @@ void BlenderSync::sync_recalc(BL::Depsgraph &b_depsgraph, BL::SpaceView3D &b_v3d
if (b_v3d) {
BlenderViewportParameters new_viewport_parameters(b_v3d);
if (viewport_parameters.modified(new_viewport_parameters)) {
if (viewport_parameters.shader_modified(new_viewport_parameters)) {
world_recalc = true;
has_updates_ = true;
}
has_updates_ |= viewport_parameters.modified(new_viewport_parameters);
}
}
@@ -227,11 +242,15 @@ void BlenderSync::sync_data(BL::RenderSettings &b_render,
int height,
void **python_thread_state)
{
if (!has_updates_) {
return;
}
scoped_timer timer;
BL::ViewLayer b_view_layer = b_depsgraph.view_layer_eval();
sync_view_layer(b_v3d, b_view_layer);
sync_view_layer(b_view_layer);
sync_integrator();
sync_film(b_v3d);
sync_shaders(b_depsgraph, b_v3d);
@@ -254,6 +273,8 @@ void BlenderSync::sync_data(BL::RenderSettings &b_render,
free_data_after_sync(b_depsgraph);
VLOG(1) << "Total time spent synchronizing data: " << timer.get_time();
has_updates_ = false;
}
/* Integrator */
@@ -424,7 +445,7 @@ void BlenderSync::sync_film(BL::SpaceView3D &b_v3d)
/* Render Layer */
void BlenderSync::sync_view_layer(BL::SpaceView3D & /*b_v3d*/, BL::ViewLayer &b_view_layer)
void BlenderSync::sync_view_layer(BL::ViewLayer &b_view_layer)
{
view_layer.name = b_view_layer.name();
@@ -875,6 +896,9 @@ SessionParams BlenderSync::get_session_params(BL::RenderEngine &b_engine,
/* Clamp samples. */
params.samples = min(params.samples, Integrator::MAX_SAMPLES);
/* Adaptive sampling. */
params.adaptive_sampling = RNA_boolean_get(&cscene, "use_adaptive_sampling");
/* tiles */
const bool is_cpu = (params.device.type == DEVICE_CPU);
if (!is_cpu && !background) {
@@ -927,7 +951,7 @@ SessionParams BlenderSync::get_session_params(BL::RenderEngine &b_engine,
BL::RenderSettings b_r = b_scene.render();
params.progressive_refine = b_engine.is_preview() ||
get_boolean(cscene, "use_progressive_refine");
if (b_r.use_save_buffers())
if (b_r.use_save_buffers() || params.adaptive_sampling)
params.progressive_refine = false;
if (background) {
@@ -963,8 +987,6 @@ SessionParams BlenderSync::get_session_params(BL::RenderEngine &b_engine,
params.use_profiling = params.device.has_profiling && !b_engine.is_preview() && background &&
BlenderSession::print_render_stats;
params.adaptive_sampling = RNA_boolean_get(&cscene, "use_adaptive_sampling");
return params;
}

View File

@@ -73,7 +73,7 @@ class BlenderSync {
int width,
int height,
void **python_thread_state);
void sync_view_layer(BL::SpaceView3D &b_v3d, BL::ViewLayer &b_view_layer);
void sync_view_layer(BL::ViewLayer &b_view_layer);
vector<Pass> sync_render_passes(BL::Scene &b_scene,
BL::RenderLayer &b_render_layer,
BL::ViewLayer &b_view_layer,
@@ -104,8 +104,7 @@ class BlenderSync {
bool background,
BL::ViewLayer b_view_layer = BL::ViewLayer(PointerRNA_NULL));
static bool get_session_pause(BL::Scene &b_scene, bool background);
static BufferParams get_buffer_params(BL::RenderSettings &b_render,
BL::SpaceView3D &b_v3d,
static BufferParams get_buffer_params(BL::SpaceView3D &b_v3d,
BL::RegionView3D &b_rv3d,
Camera *cam,
int width,
@@ -264,6 +263,12 @@ class BlenderSync {
} view_layer;
Progress &progress;
protected:
/* Indicates that `sync_recalc()` detected changes in the scene.
* If this flag is false then the data is considered to be up-to-date and will not be
* synchronized at all. */
bool has_updates_ = true;
};
CCL_NAMESPACE_END

View File

@@ -13,6 +13,7 @@
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "blender_viewport.h"
#include "blender_util.h"
@@ -25,29 +26,39 @@ BlenderViewportParameters::BlenderViewportParameters()
studiolight_rotate_z(0.0f),
studiolight_intensity(1.0f),
studiolight_background_alpha(1.0f),
studiolight_path(ustring())
display_pass(PASS_COMBINED)
{
}
BlenderViewportParameters::BlenderViewportParameters(BL::SpaceView3D &b_v3d)
: BlenderViewportParameters()
{
/* We only copy the parameters if we are in look dev mode. otherwise
if (!b_v3d) {
return;
}
BL::View3DShading shading = b_v3d.shading();
PointerRNA cshading = RNA_pointer_get(&shading.ptr, "cycles");
/* We only copy the shading parameters if we are in look dev mode. otherwise
* defaults are being used. These defaults mimic normal render settings */
if (b_v3d && b_v3d.shading().type() == BL::View3DShading::type_RENDERED) {
use_scene_world = b_v3d.shading().use_scene_world_render();
use_scene_lights = b_v3d.shading().use_scene_lights_render();
if (shading.type() == BL::View3DShading::type_RENDERED) {
use_scene_world = shading.use_scene_world_render();
use_scene_lights = shading.use_scene_lights_render();
if (!use_scene_world) {
studiolight_rotate_z = b_v3d.shading().studiolight_rotate_z();
studiolight_intensity = b_v3d.shading().studiolight_intensity();
studiolight_background_alpha = b_v3d.shading().studiolight_background_alpha();
studiolight_path = b_v3d.shading().selected_studio_light().path();
studiolight_rotate_z = shading.studiolight_rotate_z();
studiolight_intensity = shading.studiolight_intensity();
studiolight_background_alpha = shading.studiolight_background_alpha();
studiolight_path = shading.selected_studio_light().path();
}
}
/* Film. */
display_pass = (PassType)get_enum(cshading, "render_pass", -1, -1);
}
/* Check if two instances are different. */
const bool BlenderViewportParameters::modified(const BlenderViewportParameters &other) const
bool BlenderViewportParameters::shader_modified(const BlenderViewportParameters &other) const
{
return use_scene_world != other.use_scene_world || use_scene_lights != other.use_scene_lights ||
studiolight_rotate_z != other.studiolight_rotate_z ||
@@ -56,26 +67,26 @@ const bool BlenderViewportParameters::modified(const BlenderViewportParameters &
studiolight_path != other.studiolight_path;
}
const bool BlenderViewportParameters::custom_viewport_parameters() const
bool BlenderViewportParameters::film_modified(const BlenderViewportParameters &other) const
{
return !(use_scene_world && use_scene_lights);
return display_pass != other.display_pass;
}
PassType BlenderViewportParameters::get_viewport_display_render_pass(BL::SpaceView3D &b_v3d)
bool BlenderViewportParameters::modified(const BlenderViewportParameters &other) const
{
PassType display_pass = PASS_NONE;
if (b_v3d) {
BL::View3DShading b_view3dshading = b_v3d.shading();
PointerRNA cshading = RNA_pointer_get(&b_view3dshading.ptr, "cycles");
display_pass = (PassType)get_enum(cshading, "render_pass", -1, -1);
}
return display_pass;
return shader_modified(other) || film_modified(other);
}
bool BlenderViewportParameters::use_custom_shader() const
{
return !(use_scene_world && use_scene_lights);
}
PassType update_viewport_display_passes(BL::SpaceView3D &b_v3d, vector<Pass> &passes)
{
if (b_v3d) {
PassType display_pass = BlenderViewportParameters::get_viewport_display_render_pass(b_v3d);
const BlenderViewportParameters viewport_parameters(b_v3d);
const PassType display_pass = viewport_parameters.display_pass;
passes.clear();
Pass::add(display_pass, passes);

View File

@@ -18,17 +18,18 @@
#define __BLENDER_VIEWPORT_H__
#include "MEM_guardedalloc.h"
#include "RNA_access.h"
#include "RNA_blender_cpp.h"
#include "RNA_types.h"
#include "render/film.h"
#include "util/util_param.h"
CCL_NAMESPACE_BEGIN
class BlenderViewportParameters {
private:
public:
/* Shader. */
bool use_scene_world;
bool use_scene_lights;
float studiolight_rotate_z;
@@ -36,17 +37,24 @@ class BlenderViewportParameters {
float studiolight_background_alpha;
ustring studiolight_path;
/* Film. */
PassType display_pass;
BlenderViewportParameters();
BlenderViewportParameters(BL::SpaceView3D &b_v3d);
explicit BlenderViewportParameters(BL::SpaceView3D &b_v3d);
const bool modified(const BlenderViewportParameters &other) const;
const bool custom_viewport_parameters() const;
friend class BlenderSync;
/* Check whether any of shading related settings are different from the given parameters. */
bool shader_modified(const BlenderViewportParameters &other) const;
public:
/* Retrieve the render pass that needs to be displayed on the given `SpaceView3D`
* When the `b_v3d` parameter is not given `PASS_NONE` will be returned. */
static PassType get_viewport_display_render_pass(BL::SpaceView3D &b_v3d);
/* Check whether any of film related settings are different from the given parameters. */
bool film_modified(const BlenderViewportParameters &other) const;
/* Check whether any of settings are different from the given parameters. */
bool modified(const BlenderViewportParameters &other) const;
/* Returns truth when a custom shader defined by the viewport is to be used instead of the
* regular background shader or scene light. */
bool use_custom_shader() const;
};
PassType update_viewport_display_passes(BL::SpaceView3D &b_v3d, vector<Pass> &passes);

View File

@@ -17,6 +17,8 @@
#ifdef WITH_OPTIX
# include "device/device.h"
# include "bvh/bvh_optix.h"
CCL_NAMESPACE_BEGIN
@@ -26,6 +28,7 @@ BVHOptiX::BVHOptiX(const BVHParams &params_,
const vector<Object *> &objects_,
Device *device)
: BVH(params_, geometry_, objects_),
device(device),
traversable_handle(0),
as_data(device, params_.top_level ? "optix tlas" : "optix blas", false),
motion_transform_data(device, "optix motion transform", false)
@@ -34,7 +37,9 @@ BVHOptiX::BVHOptiX(const BVHParams &params_,
BVHOptiX::~BVHOptiX()
{
// Acceleration structure memory is freed via the 'as_data' destructor
// Acceleration structure memory is delayed freed on device, since deleting the
// BVH may happen while still being used for rendering.
device->release_optix_bvh(this);
}
CCL_NAMESPACE_END

View File

@@ -28,6 +28,7 @@ CCL_NAMESPACE_BEGIN
class BVHOptiX : public BVH {
public:
Device *device;
uint64_t traversable_handle;
device_only_memory<char> as_data;
device_only_memory<char> motion_transform_data;

View File

@@ -61,7 +61,6 @@ enum DeviceTypeMask {
};
enum DeviceKernelStatus {
DEVICE_KERNEL_WAITING_FOR_FEATURE_KERNEL = 0,
DEVICE_KERNEL_FEATURE_KERNEL_AVAILABLE,
DEVICE_KERNEL_USING_FEATURE_KERNEL,
DEVICE_KERNEL_FEATURE_KERNEL_INVALID,
@@ -427,6 +426,9 @@ class Device {
/* acceleration structure building */
virtual void build_bvh(BVH *bvh, Progress &progress, bool refit);
/* OptiX specific destructor. */
virtual void release_optix_bvh(BVH * /*bvh*/){};
#ifdef WITH_NETWORK
/* networking */
void server_run();

View File

@@ -35,10 +35,54 @@ device_memory::device_memory(Device *device, const char *name, MemoryType type)
device_pointer(0),
host_pointer(0),
shared_pointer(0),
shared_counter(0)
shared_counter(0),
original_device_ptr(0),
original_device_size(0),
original_device(0),
need_realloc_(false),
modified(false)
{
}
device_memory::device_memory(device_memory &&other) noexcept
: data_type(other.data_type),
data_elements(other.data_elements),
data_size(other.data_size),
device_size(other.device_size),
data_width(other.data_width),
data_height(other.data_height),
data_depth(other.data_depth),
type(other.type),
name(other.name),
device(other.device),
device_pointer(other.device_pointer),
host_pointer(other.host_pointer),
shared_pointer(other.shared_pointer),
shared_counter(other.shared_counter),
original_device_ptr(other.original_device_ptr),
original_device_size(other.original_device_size),
original_device(other.original_device),
need_realloc_(other.need_realloc_),
modified(other.modified)
{
other.data_elements = 0;
other.data_size = 0;
other.device_size = 0;
other.data_width = 0;
other.data_height = 0;
other.data_depth = 0;
other.device = 0;
other.device_pointer = 0;
other.host_pointer = 0;
other.shared_pointer = 0;
other.shared_counter = 0;
other.original_device_ptr = 0;
other.original_device_size = 0;
other.original_device = 0;
other.need_realloc_ = false;
other.modified = false;
}
device_memory::~device_memory()
{
assert(shared_pointer == 0);

View File

@@ -238,6 +238,7 @@ class device_memory {
/* Only create through subclasses. */
device_memory(Device *device, const char *name, MemoryType type);
device_memory(device_memory &&other) noexcept;
/* No copying allowed. */
device_memory(const device_memory &) = delete;
@@ -277,6 +278,10 @@ template<typename T> class device_only_memory : public device_memory {
data_elements = max(device_type_traits<T>::num_elements, 1);
}
device_only_memory(device_only_memory &&other) noexcept : device_memory(std::move(other))
{
}
virtual ~device_only_memory()
{
free();

View File

@@ -232,10 +232,6 @@ class MultiDevice : public Device {
foreach (SubDevice &sub, devices) {
DeviceKernelStatus subresult = sub.device->get_active_kernel_switch_state();
switch (subresult) {
case DEVICE_KERNEL_WAITING_FOR_FEATURE_KERNEL:
result = subresult;
break;
case DEVICE_KERNEL_FEATURE_KERNEL_INVALID:
case DEVICE_KERNEL_FEATURE_KERNEL_AVAILABLE:
return subresult;

View File

@@ -193,6 +193,9 @@ class OptiXDevice : public CUDADevice {
device_only_memory<unsigned char> denoiser_state;
int denoiser_input_passes = 0;
vector<device_only_memory<char>> delayed_free_bvh_memory;
thread_mutex delayed_free_bvh_mutex;
public:
OptiXDevice(DeviceInfo &info_, Stats &stats_, Profiler &profiler_, bool background_)
: CUDADevice(info_, stats_, profiler_, background_),
@@ -258,6 +261,8 @@ class OptiXDevice : public CUDADevice {
// Make CUDA context current
const CUDAContextScope scope(cuContext);
free_bvh_memory_delayed();
sbt_data.free();
texture_info.free();
launch_params.free();
@@ -721,7 +726,11 @@ class OptiXDevice : public CUDADevice {
}
}
else if (task.type == DeviceTask::SHADER) {
launch_shader_eval(task, thread_index);
// CUDA kernels are used when doing baking
if (optix_module == NULL)
CUDADevice::shader(task);
else
launch_shader_eval(task, thread_index);
}
else if (task.type == DeviceTask::DENOISE_BUFFER) {
// Set up a single tile that covers the whole task and denoise it
@@ -1297,6 +1306,8 @@ class OptiXDevice : public CUDADevice {
return;
}
free_bvh_memory_delayed();
BVHOptiX *const bvh_optix = static_cast<BVHOptiX *>(bvh);
progress.set_substatus("Building OptiX acceleration structure");
@@ -1767,6 +1778,24 @@ class OptiXDevice : public CUDADevice {
}
}
void release_optix_bvh(BVH *bvh) override
{
thread_scoped_lock lock(delayed_free_bvh_mutex);
/* Do delayed free of BVH memory, since geometry holding BVH might be deleted
* while GPU is still rendering. */
BVHOptiX *const bvh_optix = static_cast<BVHOptiX *>(bvh);
delayed_free_bvh_memory.emplace_back(std::move(bvh_optix->as_data));
delayed_free_bvh_memory.emplace_back(std::move(bvh_optix->motion_transform_data));
bvh_optix->traversable_handle = 0;
}
void free_bvh_memory_delayed()
{
thread_scoped_lock lock(delayed_free_bvh_mutex);
delayed_free_bvh_memory.free_memory();
}
void const_copy_to(const char *name, void *host, size_t size) override
{
// Set constant memory for CUDA module

View File

@@ -269,7 +269,6 @@ class OpenCLDevice : public Device {
cl_device_id cdDevice;
cl_int ciErr;
int device_num;
bool use_preview_kernels;
class OpenCLProgram {
public:
@@ -369,8 +368,7 @@ class OpenCLDevice : public Device {
/* Load the kernels and put the created kernels in the given
* `programs` parameter. */
void load_kernels(vector<OpenCLProgram *> &programs,
const DeviceRequestedFeatures &requested_features,
bool is_preview = false);
const DeviceRequestedFeatures &requested_features);
};
DeviceSplitKernel *split_kernel;
@@ -382,7 +380,6 @@ class OpenCLDevice : public Device {
OpenCLProgram denoising_program;
OpenCLSplitPrograms kernel_programs;
OpenCLSplitPrograms preview_programs;
typedef map<string, device_vector<uchar> *> ConstMemMap;
typedef map<string, device_ptr> MemMap;
@@ -412,7 +409,6 @@ class OpenCLDevice : public Device {
string device_md5_hash(string kernel_custom_build_options = "");
bool load_kernels(const DeviceRequestedFeatures &requested_features);
void load_required_kernels(const DeviceRequestedFeatures &requested_features);
void load_preview_kernels();
bool wait_for_availability(const DeviceRequestedFeatures &requested_features);
DeviceKernelStatus get_active_kernel_switch_state();
@@ -422,8 +418,7 @@ class OpenCLDevice : public Device {
/* Get the program file name to compile (*.cl) for the given kernel */
const string get_opencl_program_filename(const string &kernel_name);
string get_build_options(const DeviceRequestedFeatures &requested_features,
const string &opencl_program_name,
bool preview_kernel = false);
const string &opencl_program_name);
/* Enable the default features to reduce recompilation events */
void enable_default_features(DeviceRequestedFeatures &features);

View File

@@ -107,8 +107,7 @@ void OpenCLDevice::enable_default_features(DeviceRequestedFeatures &features)
}
string OpenCLDevice::get_build_options(const DeviceRequestedFeatures &requested_features,
const string &opencl_program_name,
bool preview_kernel)
const string &opencl_program_name)
{
/* first check for non-split kernel programs */
if (opencl_program_name == "base" || opencl_program_name == "denoising") {
@@ -185,13 +184,7 @@ string OpenCLDevice::get_build_options(const DeviceRequestedFeatures &requested_
enable_default_features(nofeatures);
/* Add program specific optimized compile directives */
if (preview_kernel) {
DeviceRequestedFeatures preview_features;
preview_features.use_hair = true;
build_options += "-D__KERNEL_AO_PREVIEW__ ";
build_options += preview_features.get_build_options();
}
else if (opencl_program_name == "split_do_volume" && !requested_features.use_volume) {
if (opencl_program_name == "split_do_volume" && !requested_features.use_volume) {
build_options += nofeatures.get_build_options();
}
else {
@@ -238,9 +231,7 @@ OpenCLDevice::OpenCLSplitPrograms::~OpenCLSplitPrograms()
}
void OpenCLDevice::OpenCLSplitPrograms::load_kernels(
vector<OpenCLProgram *> &programs,
const DeviceRequestedFeatures &requested_features,
bool is_preview)
vector<OpenCLProgram *> &programs, const DeviceRequestedFeatures &requested_features)
{
if (!requested_features.use_baking) {
# define ADD_SPLIT_KERNEL_BUNDLE_PROGRAM(kernel_name) \
@@ -251,7 +242,7 @@ void OpenCLDevice::OpenCLSplitPrograms::load_kernels(
device, \
program_name_##kernel_name, \
"kernel_" #kernel_name ".cl", \
device->get_build_options(requested_features, program_name_##kernel_name, is_preview)); \
device->get_build_options(requested_features, program_name_##kernel_name)); \
program_##kernel_name.add_kernel(ustring("path_trace_" #kernel_name)); \
programs.push_back(&program_##kernel_name);
@@ -259,7 +250,7 @@ void OpenCLDevice::OpenCLSplitPrograms::load_kernels(
ADD_SPLIT_KERNEL_PROGRAM(subsurface_scatter);
ADD_SPLIT_KERNEL_PROGRAM(direct_lighting);
ADD_SPLIT_KERNEL_PROGRAM(indirect_background);
if (requested_features.use_volume || is_preview) {
if (requested_features.use_volume) {
ADD_SPLIT_KERNEL_PROGRAM(do_volume);
}
ADD_SPLIT_KERNEL_PROGRAM(shader_eval);
@@ -274,7 +265,7 @@ void OpenCLDevice::OpenCLSplitPrograms::load_kernels(
device,
"split_bundle",
"kernel_split_bundle.cl",
device->get_build_options(requested_features, "split_bundle", is_preview));
device->get_build_options(requested_features, "split_bundle"));
ADD_SPLIT_KERNEL_BUNDLE_PROGRAM(data_init);
ADD_SPLIT_KERNEL_BUNDLE_PROGRAM(state_buffer_size);
@@ -403,7 +394,7 @@ class OpenCLSplitKernel : public DeviceSplitKernel {
device,
program_name,
device->get_opencl_program_filename(kernel_name),
device->get_build_options(requested_features, program_name, device->use_preview_kernels));
device->get_build_options(requested_features, program_name));
kernel->program.add_kernel(ustring("path_trace_" + kernel_name));
kernel->program.load();
@@ -569,6 +560,11 @@ class OpenCLSplitKernel : public DeviceSplitKernel {
size_t num_elements = max_elements_for_max_buffer_size(kg, data, max_buffer_size);
int2 global_size = make_int2(max(round_down((int)sqrt(num_elements), 64), 64),
(int)sqrt(num_elements));
if (device->info.description.find("Intel") != string::npos) {
global_size = make_int2(min(512, global_size.x), min(512, global_size.y));
}
VLOG(1) << "Global size: " << global_size << ".";
return global_size;
}
@@ -612,7 +608,6 @@ OpenCLDevice::OpenCLDevice(DeviceInfo &info, Stats &stats, Profiler &profiler, b
: Device(info, stats, profiler, background),
load_kernel_num_compiling(0),
kernel_programs(this),
preview_programs(this),
memory_manager(this),
texture_info(this, "__texture_info", MEM_GLOBAL)
{
@@ -622,7 +617,6 @@ OpenCLDevice::OpenCLDevice(DeviceInfo &info, Stats &stats, Profiler &profiler, b
cqCommandQueue = NULL;
device_initialized = false;
textures_need_update = true;
use_preview_kernels = !background;
vector<OpenCLPlatformDevice> usable_devices;
OpenCLInfo::get_usable_devices(&usable_devices);
@@ -678,9 +672,6 @@ OpenCLDevice::OpenCLDevice(DeviceInfo &info, Stats &stats, Profiler &profiler, b
device_initialized = true;
split_kernel = new OpenCLSplitKernel(this);
if (use_preview_kernels) {
load_preview_kernels();
}
}
OpenCLDevice::~OpenCLDevice()
@@ -771,7 +762,7 @@ bool OpenCLDevice::load_kernels(const DeviceRequestedFeatures &requested_feature
load_required_kernels(requested_features);
vector<OpenCLProgram *> programs;
kernel_programs.load_kernels(programs, requested_features, false);
kernel_programs.load_kernels(programs, requested_features);
if (!requested_features.use_baking && requested_features.use_denoising) {
denoising_program = OpenCLProgram(
@@ -849,19 +840,6 @@ void OpenCLDevice::load_required_kernels(const DeviceRequestedFeatures &requeste
}
}
void OpenCLDevice::load_preview_kernels()
{
DeviceRequestedFeatures no_features;
vector<OpenCLProgram *> programs;
preview_programs.load_kernels(programs, no_features, true);
foreach (OpenCLProgram *program, programs) {
if (!program->load()) {
load_required_kernel_task_pool.push(function_bind(&OpenCLProgram::compile, program));
}
}
}
bool OpenCLDevice::wait_for_availability(const DeviceRequestedFeatures &requested_features)
{
if (requested_features.use_baking) {
@@ -869,59 +847,18 @@ bool OpenCLDevice::wait_for_availability(const DeviceRequestedFeatures &requeste
return true;
}
if (background) {
load_kernel_task_pool.wait_work();
use_preview_kernels = false;
}
else {
/* We use a device setting to determine to load preview kernels or not
* Better to check on device level than per kernel as mixing preview and
* non-preview kernels does not work due to different data types */
if (use_preview_kernels) {
use_preview_kernels = load_kernel_num_compiling.load() > 0;
}
}
load_kernel_task_pool.wait_work();
return split_kernel->load_kernels(requested_features);
}
OpenCLDevice::OpenCLSplitPrograms *OpenCLDevice::get_split_programs()
{
return use_preview_kernels ? &preview_programs : &kernel_programs;
return &kernel_programs;
}
DeviceKernelStatus OpenCLDevice::get_active_kernel_switch_state()
{
/* Do not switch kernels for background renderings
* We do foreground rendering but use the preview kernels
* Check for the optimized kernels
*
* This works also the other way around, where we are using
* optimized kernels but new ones are being compiled due
* to other features that are needed */
if (background) {
/* The if-statements below would find the same result,
* But as the `finished` method uses a mutex we added
* this as an early exit */
return DEVICE_KERNEL_USING_FEATURE_KERNEL;
}
bool other_kernels_finished = load_kernel_num_compiling.load() == 0;
if (use_preview_kernels) {
if (other_kernels_finished) {
return DEVICE_KERNEL_FEATURE_KERNEL_AVAILABLE;
}
else {
return DEVICE_KERNEL_WAITING_FOR_FEATURE_KERNEL;
}
}
else {
if (other_kernels_finished) {
return DEVICE_KERNEL_USING_FEATURE_KERNEL;
}
else {
return DEVICE_KERNEL_FEATURE_KERNEL_INVALID;
}
}
return DEVICE_KERNEL_USING_FEATURE_KERNEL;
}
void OpenCLDevice::mem_alloc(device_memory &mem)

View File

@@ -195,31 +195,108 @@ ccl_device float2 regular_polygon_sample(float corners, float rotation, float u,
ccl_device float3 ensure_valid_reflection(float3 Ng, float3 I, float3 N)
{
float3 R;
float NI = dot(N, I);
float NgR, threshold;
float3 R = 2 * dot(N, I) * N - I;
/* Check if the incident ray is coming from behind normal N. */
if (NI > 0) {
/* Normal reflection */
R = (2 * NI) * N - I;
NgR = dot(Ng, R);
/* Reflection rays may always be at least as shallow as the incoming ray. */
float threshold = min(0.9f * dot(Ng, I), 0.01f);
if (dot(Ng, R) >= threshold) {
return N;
}
/* Reflection rays may always be at least as shallow as the incoming ray. */
threshold = min(0.9f * dot(Ng, I), 0.01f);
if (NgR >= threshold) {
return N;
/* Form coordinate system with Ng as the Z axis and N inside the X-Z-plane.
* The X axis is found by normalizing the component of N that's orthogonal to Ng.
* The Y axis isn't actually needed.
*/
float NdotNg = dot(N, Ng);
float3 X = normalize(N - NdotNg * Ng);
/* Keep math expressions. */
/* clang-format off */
/* Calculate N.z and N.x in the local coordinate system.
*
* The goal of this computation is to find a N' that is rotated towards Ng just enough
* to lift R' above the threshold (here called t), therefore dot(R', Ng) = t.
*
* According to the standard reflection equation,
* this means that we want dot(2*dot(N', I)*N' - I, Ng) = t.
*
* Since the Z axis of our local coordinate system is Ng, dot(x, Ng) is just x.z, so we get
* 2*dot(N', I)*N'.z - I.z = t.
*
* The rotation is simple to express in the coordinate system we formed -
* since N lies in the X-Z-plane, we know that N' will also lie in the X-Z-plane,
* so N'.y = 0 and therefore dot(N', I) = N'.x*I.x + N'.z*I.z .
*
* Furthermore, we want N' to be normalized, so N'.x = sqrt(1 - N'.z^2).
*
* With these simplifications,
* we get the final equation 2*(sqrt(1 - N'.z^2)*I.x + N'.z*I.z)*N'.z - I.z = t.
*
* The only unknown here is N'.z, so we can solve for that.
*
* The equation has four solutions in general:
*
* N'.z = +-sqrt(0.5*(+-sqrt(I.x^2*(I.x^2 + I.z^2 - t^2)) + t*I.z + I.x^2 + I.z^2)/(I.x^2 + I.z^2))
* We can simplify this expression a bit by grouping terms:
*
* a = I.x^2 + I.z^2
* b = sqrt(I.x^2 * (a - t^2))
* c = I.z*t + a
* N'.z = +-sqrt(0.5*(+-b + c)/a)
*
* Two solutions can immediately be discarded because they're negative so N' would lie in the
* lower hemisphere.
*/
/* clang-format on */
float Ix = dot(I, X), Iz = dot(I, Ng);
float Ix2 = sqr(Ix), Iz2 = sqr(Iz);
float a = Ix2 + Iz2;
float b = safe_sqrtf(Ix2 * (a - sqr(threshold)));
float c = Iz * threshold + a;
/* Evaluate both solutions.
* In many cases one can be immediately discarded (if N'.z would be imaginary or larger than
* one), so check for that first. If no option is viable (might happen in extreme cases like N
* being in the wrong hemisphere), give up and return Ng. */
float fac = 0.5f / a;
float N1_z2 = fac * (b + c), N2_z2 = fac * (-b + c);
bool valid1 = (N1_z2 > 1e-5f) && (N1_z2 <= (1.0f + 1e-5f));
bool valid2 = (N2_z2 > 1e-5f) && (N2_z2 <= (1.0f + 1e-5f));
float2 N_new;
if (valid1 && valid2) {
/* If both are possible, do the expensive reflection-based check. */
float2 N1 = make_float2(safe_sqrtf(1.0f - N1_z2), safe_sqrtf(N1_z2));
float2 N2 = make_float2(safe_sqrtf(1.0f - N2_z2), safe_sqrtf(N2_z2));
float R1 = 2 * (N1.x * Ix + N1.y * Iz) * N1.y - Iz;
float R2 = 2 * (N2.x * Ix + N2.y * Iz) * N2.y - Iz;
valid1 = (R1 >= 1e-5f);
valid2 = (R2 >= 1e-5f);
if (valid1 && valid2) {
/* If both solutions are valid, return the one with the shallower reflection since it will be
* closer to the input (if the original reflection wasn't shallow, we would not be in this
* part of the function). */
N_new = (R1 < R2) ? N1 : N2;
}
else {
/* If only one reflection is valid (= positive), pick that one. */
N_new = (R1 > R2) ? N1 : N2;
}
}
else if (valid1 || valid2) {
/* Only one solution passes the N'.z criterium, so pick that one. */
float Nz2 = valid1 ? N1_z2 : N2_z2;
N_new = make_float2(safe_sqrtf(1.0f - Nz2), safe_sqrtf(Nz2));
}
else {
/* Bad incident */
R = -I;
NgR = dot(Ng, R);
threshold = 0.01f;
return Ng;
}
R = R + Ng * (threshold - NgR); /* Lift the reflection above the threshold. */
return normalize(I * len(R) + R * len(I)); /* Find a bisector. */
return N_new.x * X + N_new.y * Ng;
}
CCL_NAMESPACE_END

View File

@@ -99,27 +99,23 @@ CCL_NAMESPACE_BEGIN
#define __AO__
#define __PASSES__
#define __HAIR__
/* Without these we get an AO render, used by OpenCL preview kernel. */
#ifndef __KERNEL_AO_PREVIEW__
# define __SVM__
# define __EMISSION__
# define __HOLDOUT__
# define __MULTI_CLOSURE__
# define __TRANSPARENT_SHADOWS__
# define __BACKGROUND_MIS__
# define __LAMP_MIS__
# define __CAMERA_MOTION__
# define __OBJECT_MOTION__
# define __BAKING__
# define __PRINCIPLED__
# define __SUBSURFACE__
# define __VOLUME__
# define __VOLUME_SCATTER__
# define __CMJ__
# define __SHADOW_RECORD_ALL__
# define __BRANCHED_PATH__
#endif
#define __SVM__
#define __EMISSION__
#define __HOLDOUT__
#define __MULTI_CLOSURE__
#define __TRANSPARENT_SHADOWS__
#define __BACKGROUND_MIS__
#define __LAMP_MIS__
#define __CAMERA_MOTION__
#define __OBJECT_MOTION__
#define __BAKING__
#define __PRINCIPLED__
#define __SUBSURFACE__
#define __VOLUME__
#define __VOLUME_SCATTER__
#define __CMJ__
#define __SHADOW_RECORD_ALL__
#define __BRANCHED_PATH__
/* Device specific features */
#ifdef __KERNEL_CPU__

View File

@@ -25,7 +25,7 @@
* coordinates to act as a seed since the noise functions don't have seed values.
* A seed value is needed for generating distortion textures and color outputs.
* The offset's components are in the range [100, 200], not too high to cause
* bad precision and not to small to be noticeable. We use float seed because
* bad precision and not too small to be noticeable. We use float seed because
* OSL only support float hashes.
*/

View File

@@ -84,30 +84,67 @@ closure color principled_hair(normal N,
closure color henyey_greenstein(float g) BUILTIN;
closure color absorption() BUILTIN;
normal ensure_valid_reflection(normal Ng, normal I, normal N)
normal ensure_valid_reflection(normal Ng, vector I, normal N)
{
/* The implementation here mirrors the one in kernel_montecarlo.h,
* check there for an explanation of the algorithm. */
vector R;
float NI = dot(N, I);
float NgR, threshold;
if (NI > 0) {
R = (2 * NI) * N - I;
NgR = dot(Ng, R);
threshold = min(0.9 * dot(Ng, I), 0.01);
if (NgR >= threshold) {
return N;
float sqr(float x)
{
return x * x;
}
vector R = 2 * dot(N, I) * N - I;
float threshold = min(0.9 * dot(Ng, I), 0.01);
if (dot(Ng, R) >= threshold) {
return N;
}
float NdotNg = dot(N, Ng);
vector X = normalize(N - NdotNg * Ng);
float Ix = dot(I, X), Iz = dot(I, Ng);
float Ix2 = sqr(Ix), Iz2 = sqr(Iz);
float a = Ix2 + Iz2;
float b = sqrt(Ix2 * (a - sqr(threshold)));
float c = Iz * threshold + a;
float fac = 0.5 / a;
float N1_z2 = fac * (b + c), N2_z2 = fac * (-b + c);
int valid1 = (N1_z2 > 1e-5) && (N1_z2 <= (1.0 + 1e-5));
int valid2 = (N2_z2 > 1e-5) && (N2_z2 <= (1.0 + 1e-5));
float N_new_x, N_new_z;
if (valid1 && valid2) {
float N1_x = sqrt(1.0 - N1_z2), N1_z = sqrt(N1_z2);
float N2_x = sqrt(1.0 - N2_z2), N2_z = sqrt(N2_z2);
float R1 = 2 * (N1_x * Ix + N1_z * Iz) * N1_z - Iz;
float R2 = 2 * (N2_x * Ix + N2_z * Iz) * N2_z - Iz;
valid1 = (R1 >= 1e-5);
valid2 = (R2 >= 1e-5);
if (valid1 && valid2) {
N_new_x = (R1 < R2) ? N1_x : N2_x;
N_new_z = (R1 < R2) ? N1_z : N2_z;
}
else {
N_new_x = (R1 > R2) ? N1_x : N2_x;
N_new_z = (R1 > R2) ? N1_z : N2_z;
}
}
else if (valid1 || valid2) {
float Nz2 = valid1 ? N1_z2 : N2_z2;
N_new_x = sqrt(1.0 - Nz2);
N_new_z = sqrt(Nz2);
}
else {
R = -I;
NgR = dot(Ng, R);
threshold = 0.01;
return Ng;
}
R = R + Ng * (threshold - NgR);
return normalize(I * length(R) + R * length(I));
return N_new_x * X + N_new_z * Ng;
}
#endif /* CCL_STDOSL_H */

View File

@@ -20,7 +20,7 @@ CCL_NAMESPACE_BEGIN
* coordinates to act as a seed since the noise functions don't have seed values.
* A seed value is needed for generating distortion textures and color outputs.
* The offset's components are in the range [100, 200], not too high to cause
* bad precision and not to small to be noticeable. We use float seed because
* bad precision and not too small to be noticeable. We use float seed because
* OSL only support float hashes.
*/

View File

@@ -498,12 +498,8 @@ void AlembicObject::load_data_in_cache(CachedData &cached_data,
/* Use the schema as the base compound property to also be able to look for top level properties.
*/
read_attributes(proc,
cached_data,
schema.getArbGeomParams(),
schema.getUVsParam(),
get_requested_attributes(),
progress);
read_attributes(
proc, cached_data, schema, schema.getUVsParam(), get_requested_attributes(), progress);
cached_data.invalidate_last_loaded_time(true);
data_loaded = true;
@@ -658,8 +654,7 @@ static void update_attributes(AttributeSet &attributes, CachedData &cached_data,
list<Attribute>::iterator it;
for (it = attributes.attributes.begin(); it != attributes.attributes.end();) {
if (cached_attributes.find(&(*it)) == cached_attributes.end()) {
attributes.attributes.erase(it++);
attributes.modified = true;
attributes.remove(it++);
continue;
}

View File

@@ -606,7 +606,8 @@ void read_geometry_data(AlembicProcedural *proc,
template<typename T> struct value_type_converter {
using cycles_type = float;
static constexpr TypeDesc type_desc = TypeFloat;
/* Use `TypeDesc::FLOAT` instead of `TypeFloat` to work around a compiler bug in gcc 11. */
static constexpr TypeDesc type_desc = TypeDesc::FLOAT;
static constexpr const char *type_name = "float (default)";
static cycles_type convert_value(T value)
@@ -894,6 +895,10 @@ static void parse_requested_attributes_recursive(const AttributeRequestSet &requ
const ICompoundProperty &arb_geom_params,
vector<PropHeaderAndParent> &requested_properties)
{
if (!arb_geom_params.valid()) {
return;
}
for (const AttributeRequest &req : requested_attributes.requests) {
const PropertyHeader *property_header = arb_geom_params.getPropertyHeader(req.name.c_str());

View File

@@ -383,6 +383,23 @@ AttributeStandard Attribute::name_standard(const char *name)
return ATTR_STD_NONE;
}
AttrKernelDataType Attribute::kernel_type(const Attribute &attr)
{
if (attr.element == ATTR_ELEMENT_CORNER) {
return AttrKernelDataType::UCHAR4;
}
if (attr.type == TypeDesc::TypeFloat) {
return AttrKernelDataType::FLOAT;
}
if (attr.type == TypeFloat2) {
return AttrKernelDataType::FLOAT2;
}
return AttrKernelDataType::FLOAT3;
}
void Attribute::get_uv_tiles(Geometry *geom,
AttributePrimitive prim,
unordered_set<int> &tiles) const
@@ -417,7 +434,7 @@ void Attribute::get_uv_tiles(Geometry *geom,
/* Attribute Set */
AttributeSet::AttributeSet(Geometry *geometry, AttributePrimitive prim)
: geometry(geometry), prim(prim)
: modified_flag(~0u), geometry(geometry), prim(prim)
{
}
@@ -440,7 +457,7 @@ Attribute *AttributeSet::add(ustring name, TypeDesc type, AttributeElement eleme
Attribute new_attr(name, type, element, geometry, prim);
attributes.emplace_back(std::move(new_attr));
modified = true;
tag_modified(attributes.back());
return &attributes.back();
}
@@ -462,8 +479,7 @@ void AttributeSet::remove(ustring name)
for (it = attributes.begin(); it != attributes.end(); it++) {
if (&*it == attr) {
modified = true;
attributes.erase(it);
remove(it);
return;
}
}
@@ -608,8 +624,7 @@ void AttributeSet::remove(AttributeStandard std)
for (it = attributes.begin(); it != attributes.end(); it++) {
if (&*it == attr) {
modified = true;
attributes.erase(it);
remove(it);
return;
}
}
@@ -634,6 +649,12 @@ void AttributeSet::remove(Attribute *attribute)
}
}
void AttributeSet::remove(list<Attribute>::iterator it)
{
tag_modified(*it);
attributes.erase(it);
}
void AttributeSet::resize(bool reserve_only)
{
foreach (Attribute &attr, attributes) {
@@ -674,15 +695,13 @@ void AttributeSet::update(AttributeSet &&new_attributes)
for (it = attributes.begin(); it != attributes.end();) {
if (it->std != ATTR_STD_NONE) {
if (new_attributes.find(it->std) == nullptr) {
modified = true;
attributes.erase(it++);
remove(it++);
continue;
}
}
else if (it->name != "") {
if (new_attributes.find(it->name) == nullptr) {
modified = true;
attributes.erase(it++);
remove(it++);
continue;
}
}
@@ -699,7 +718,27 @@ void AttributeSet::clear_modified()
foreach (Attribute &attr, attributes) {
attr.modified = false;
}
modified = false;
modified_flag = 0;
}
void AttributeSet::tag_modified(const Attribute &attr)
{
/* Some attributes are not stored in the various kernel attribute arrays
* (DeviceScene::attribute_*), so the modified flags are only set if the associated standard
* corresponds to an attribute which will be stored in the kernel's attribute arrays. */
const bool modifies_device_array = (attr.std != ATTR_STD_FACE_NORMAL &&
attr.std != ATTR_STD_VERTEX_NORMAL);
if (modifies_device_array) {
AttrKernelDataType kernel_type = Attribute::kernel_type(attr);
modified_flag |= (1u << kernel_type);
}
}
bool AttributeSet::modified(AttrKernelDataType kernel_type) const
{
return (modified_flag & (1u << kernel_type)) != 0;
}
/* AttributeRequest */

View File

@@ -39,6 +39,21 @@ class Hair;
class Mesh;
struct Transform;
/* AttrKernelDataType.
*
* The data type of the device arrays storing the attribute's data. Those data types are different
* than the ones for attributes as some attribute types are stored in the same array, e.g. Point,
* Vector, and Transform are all stored as float3 in the kernel.
*
* The values of this enumeration are also used as flags to detect changes in AttributeSet. */
enum AttrKernelDataType {
FLOAT = 0,
FLOAT2 = 1,
FLOAT3 = 2,
UCHAR4 = 3,
};
/* Attribute
*
* Arbitrary data layers on meshes.
@@ -167,6 +182,8 @@ class Attribute {
static const char *standard_name(AttributeStandard std);
static AttributeStandard name_standard(const char *name);
static AttrKernelDataType kernel_type(const Attribute &attr);
void get_uv_tiles(Geometry *geom, AttributePrimitive prim, unordered_set<int> &tiles) const;
};
@@ -175,11 +192,12 @@ class Attribute {
* Set of attributes on a mesh. */
class AttributeSet {
uint32_t modified_flag;
public:
Geometry *geometry;
AttributePrimitive prim;
list<Attribute> attributes;
bool modified = true;
AttributeSet(Geometry *geometry, AttributePrimitive prim);
AttributeSet(AttributeSet &&) = default;
@@ -197,6 +215,8 @@ class AttributeSet {
void remove(Attribute *attribute);
void remove(list<Attribute>::iterator it);
void resize(bool reserve_only = false);
void clear(bool preserve_voxel_data = false);
@@ -204,7 +224,18 @@ class AttributeSet {
* and remove any attribute not found on the new set from this. */
void update(AttributeSet &&new_attributes);
/* Return whether the attributes of the given kernel_type are modified, where "modified" means
* that some attributes of the given type were added or removed from this AttributeSet. This does
* not mean that the data of the remaining attributes in this AttributeSet were also modified. To
* check this, use Attribute.modified. */
bool modified(AttrKernelDataType kernel_type) const;
void clear_modified();
private:
/* Set the relevant modified flag for the attribute. Only attributes that are stored in device
* arrays will be considered for tagging this AttributeSet as modified. */
void tag_modified(const Attribute &attr);
};
/* AttributeRequest

View File

@@ -46,6 +46,12 @@ CCL_NAMESPACE_BEGIN
/* Geometry */
PackFlags operator|=(PackFlags &pack_flags, uint32_t value)
{
pack_flags = (PackFlags)((uint32_t)pack_flags | value);
return pack_flags;
}
NODE_ABSTRACT_DEFINE(Geometry)
{
NodeType *type = NodeType::add("geometry_base", NULL);
@@ -824,10 +830,13 @@ void GeometryManager::device_update_attributes(Device *device,
dscene->attributes_float3.alloc(attr_float3_size);
dscene->attributes_uchar4.alloc(attr_uchar4_size);
const bool copy_all_data = dscene->attributes_float.need_realloc() ||
dscene->attributes_float2.need_realloc() ||
dscene->attributes_float3.need_realloc() ||
dscene->attributes_uchar4.need_realloc();
/* The order of those flags needs to match that of AttrKernelDataType. */
const bool attributes_need_realloc[4] = {
dscene->attributes_float.need_realloc(),
dscene->attributes_float2.need_realloc(),
dscene->attributes_float3.need_realloc(),
dscene->attributes_uchar4.need_realloc(),
};
size_t attr_float_offset = 0;
size_t attr_float2_offset = 0;
@@ -846,7 +855,7 @@ void GeometryManager::device_update_attributes(Device *device,
if (attr) {
/* force a copy if we need to reallocate all the data */
attr->modified |= copy_all_data;
attr->modified |= attributes_need_realloc[Attribute::kernel_type(*attr)];
}
update_attribute_element_offset(geom,
@@ -869,7 +878,7 @@ void GeometryManager::device_update_attributes(Device *device,
if (subd_attr) {
/* force a copy if we need to reallocate all the data */
subd_attr->modified |= copy_all_data;
subd_attr->modified |= attributes_need_realloc[Attribute::kernel_type(*subd_attr)];
}
update_attribute_element_offset(mesh,
@@ -900,6 +909,10 @@ void GeometryManager::device_update_attributes(Device *device,
foreach (AttributeRequest &req, attributes.requests) {
Attribute *attr = values.find(req);
if (attr) {
attr->modified |= attributes_need_realloc[Attribute::kernel_type(*attr)];
}
update_attribute_element_offset(object->geometry,
dscene->attributes_float,
attr_float_offset,
@@ -935,10 +948,10 @@ void GeometryManager::device_update_attributes(Device *device,
/* copy to device */
progress.set_status("Updating Mesh", "Copying Attributes to device");
dscene->attributes_float.copy_to_device();
dscene->attributes_float2.copy_to_device();
dscene->attributes_float3.copy_to_device();
dscene->attributes_uchar4.copy_to_device();
dscene->attributes_float.copy_to_device_if_modified();
dscene->attributes_float2.copy_to_device_if_modified();
dscene->attributes_float3.copy_to_device_if_modified();
dscene->attributes_uchar4.copy_to_device_if_modified();
if (progress.get_cancel())
return;
@@ -1236,7 +1249,16 @@ void GeometryManager::device_update_bvh(Device *device,
const bool can_refit = scene->bvh != nullptr &&
(bparams.bvh_layout == BVHLayout::BVH_LAYOUT_OPTIX);
const bool pack_all = scene->bvh == nullptr;
PackFlags pack_flags = PackFlags::PACK_NONE;
if (scene->bvh == nullptr) {
pack_flags |= PackFlags::PACK_ALL;
}
if (dscene->prim_visibility.is_modified()) {
pack_flags |= PackFlags::PACK_VISIBILITY;
}
BVH *bvh = scene->bvh;
if (!scene->bvh) {
@@ -1274,10 +1296,14 @@ void GeometryManager::device_update_bvh(Device *device,
pack.root_index = -1;
if (!pack_all) {
if (pack_flags != PackFlags::PACK_ALL) {
/* if we do not need to recreate the BVH, then only the vertices are updated, so we can
* safely retake the memory */
dscene->prim_tri_verts.give_data(pack.prim_tri_verts);
if ((pack_flags & PackFlags::PACK_VISIBILITY) != 0) {
dscene->prim_visibility.give_data(pack.prim_visibility);
}
}
else {
/* It is not strictly necessary to skip those resizes we if do not have to repack, as the OS
@@ -1306,13 +1332,21 @@ void GeometryManager::device_update_bvh(Device *device,
// Iterate over scene mesh list instead of objects, since 'optix_prim_offset' was calculated
// based on that list, which may be ordered differently from the object list.
foreach (Geometry *geom, scene->geometry) {
if (!pack_all && !geom->is_modified()) {
/* Make a copy of the pack_flags so the current geometry's flags do not pollute the others'.
*/
PackFlags geom_pack_flags = pack_flags;
if (geom->is_modified()) {
geom_pack_flags |= PackFlags::PACK_VERTICES;
}
if (geom_pack_flags == PACK_NONE) {
continue;
}
const pair<int, uint> &info = geometry_to_object_info[geom];
pool.push(function_bind(
&Geometry::pack_primitives, geom, &pack, info.first, info.second, pack_all));
&Geometry::pack_primitives, geom, &pack, info.first, info.second, geom_pack_flags));
}
pool.wait_work();
}
@@ -1347,7 +1381,7 @@ void GeometryManager::device_update_bvh(Device *device,
dscene->prim_type.steal_data(pack.prim_type);
dscene->prim_type.copy_to_device();
}
if (pack.prim_visibility.size() && (dscene->prim_visibility.need_realloc() || has_bvh2_layout)) {
if (pack.prim_visibility.size() && (dscene->prim_visibility.is_modified() || has_bvh2_layout)) {
dscene->prim_visibility.steal_data(pack.prim_visibility);
dscene->prim_visibility.copy_to_device();
}
@@ -1404,24 +1438,46 @@ static void update_device_flags_attribute(uint32_t &device_update_flags,
continue;
}
if (attr.element == ATTR_ELEMENT_CORNER) {
device_update_flags |= ATTR_UCHAR4_MODIFIED;
}
else if (attr.type == TypeDesc::TypeFloat) {
device_update_flags |= ATTR_FLOAT_MODIFIED;
}
else if (attr.type == TypeFloat2) {
device_update_flags |= ATTR_FLOAT2_MODIFIED;
}
else if (attr.type == TypeDesc::TypeMatrix) {
device_update_flags |= ATTR_FLOAT3_MODIFIED;
}
else if (attr.element != ATTR_ELEMENT_VOXEL) {
device_update_flags |= ATTR_FLOAT3_MODIFIED;
AttrKernelDataType kernel_type = Attribute::kernel_type(attr);
switch (kernel_type) {
case AttrKernelDataType::FLOAT: {
device_update_flags |= ATTR_FLOAT_MODIFIED;
break;
}
case AttrKernelDataType::FLOAT2: {
device_update_flags |= ATTR_FLOAT2_MODIFIED;
break;
}
case AttrKernelDataType::FLOAT3: {
device_update_flags |= ATTR_FLOAT3_MODIFIED;
break;
}
case AttrKernelDataType::UCHAR4: {
device_update_flags |= ATTR_UCHAR4_MODIFIED;
break;
}
}
}
}
static void update_attribute_realloc_flags(uint32_t &device_update_flags,
const AttributeSet &attributes)
{
if (attributes.modified(AttrKernelDataType::FLOAT)) {
device_update_flags |= ATTR_FLOAT_NEEDS_REALLOC;
}
if (attributes.modified(AttrKernelDataType::FLOAT2)) {
device_update_flags |= ATTR_FLOAT2_NEEDS_REALLOC;
}
if (attributes.modified(AttrKernelDataType::FLOAT3)) {
device_update_flags |= ATTR_FLOAT3_NEEDS_REALLOC;
}
if (attributes.modified(AttrKernelDataType::UCHAR4)) {
device_update_flags |= ATTR_UCHAR4_NEEDS_REALLOC;
}
}
void GeometryManager::device_update_preprocess(Device *device, Scene *scene, Progress &progress)
{
if (!need_update() && !need_flags_update) {
@@ -1444,16 +1500,11 @@ void GeometryManager::device_update_preprocess(Device *device, Scene *scene, Pro
foreach (Geometry *geom, scene->geometry) {
geom->has_volume = false;
if (geom->attributes.modified) {
device_update_flags |= ATTRS_NEED_REALLOC;
}
update_attribute_realloc_flags(device_update_flags, geom->attributes);
if (geom->is_mesh()) {
Mesh *mesh = static_cast<Mesh *>(geom);
if (mesh->subd_attributes.modified) {
device_update_flags |= ATTRS_NEED_REALLOC;
}
update_attribute_realloc_flags(device_update_flags, mesh->subd_attributes);
}
foreach (Node *node, geom->get_used_shaders()) {
@@ -1595,6 +1646,10 @@ void GeometryManager::device_update_preprocess(Device *device, Scene *scene, Pro
}
}
if ((update_flags & VISIBILITY_MODIFIED) != 0) {
dscene->prim_visibility.tag_modified();
}
if (device_update_flags & ATTR_FLOAT_NEEDS_REALLOC) {
dscene->attributes_map.tag_realloc();
dscene->attributes_float.tag_realloc();
@@ -1921,7 +1976,8 @@ void GeometryManager::device_update(Device *device,
* Also update the BVH if the transformations change, we cannot rely on tagging the Geometry
* as modified in this case, as we may accumulate displacement if the vertices do not also
* change. */
bool need_update_scene_bvh = (scene->bvh == nullptr || (update_flags & TRANSFORM_MODIFIED) != 0);
bool need_update_scene_bvh = (scene->bvh == nullptr ||
(update_flags & (TRANSFORM_MODIFIED | VISIBILITY_MODIFIED)) != 0);
{
scoped_callback_timer timer([scene](double time) {
if (scene->update_stats) {
@@ -2007,7 +2063,7 @@ void GeometryManager::device_update(Device *device,
* for meshes with correct bounding boxes.
*
* This wouldn't cause wrong results, just true
* displacement might be less optimal ot calculate.
* displacement might be less optimal to calculate.
*/
scene->object_manager->need_flags_update = old_need_object_flags_update;
}

View File

@@ -43,6 +43,24 @@ class Shader;
class Volume;
struct PackedBVH;
/* Flags used to determine which geometry data need to be packed. */
enum PackFlags : uint32_t {
PACK_NONE = 0u,
/* Pack the geometry information (e.g. triangle or curve keys indices). */
PACK_GEOMETRY = (1u << 0),
/* Pack the vertices, for Meshes and Volumes' bounding meshes. */
PACK_VERTICES = (1u << 1),
/* Pack the visibility flags for each triangle or curve. */
PACK_VISIBILITY = (1u << 2),
PACK_ALL = (PACK_GEOMETRY | PACK_VERTICES | PACK_VISIBILITY),
};
PackFlags operator|=(PackFlags &pack_flags, uint32_t value);
/* Geometry
*
* Base class for geometric types like Mesh and Hair. */
@@ -126,7 +144,10 @@ class Geometry : public Node {
int n,
int total);
virtual void pack_primitives(PackedBVH *pack, int object, uint visibility, bool pack_all) = 0;
virtual void pack_primitives(PackedBVH *pack,
int object,
uint visibility,
PackFlags pack_flags) = 0;
/* Check whether the geometry should have own BVH built separately. Briefly,
* own BVH is needed for geometry, if:
@@ -191,6 +212,8 @@ class GeometryManager {
TRANSFORM_MODIFIED = (1 << 10),
VISIBILITY_MODIFIED = (1 << 11),
/* tag everything in the manager for an update */
UPDATE_ALL = ~0u,

View File

@@ -494,38 +494,47 @@ void Hair::pack_curves(Scene *scene,
}
}
void Hair::pack_primitives(PackedBVH *pack, int object, uint visibility, bool pack_all)
void Hair::pack_primitives(PackedBVH *pack, int object, uint visibility, PackFlags pack_flags)
{
if (curve_first_key.empty())
return;
/* If the BVH does not have to be recreated, we can bail out. */
if (!pack_all) {
return;
/* Separate loop as other arrays are not initialized if their packing is not required. */
if ((pack_flags & PACK_VISIBILITY) != 0) {
unsigned int *prim_visibility = &pack->prim_visibility[optix_prim_offset];
size_t index = 0;
for (size_t j = 0; j < num_curves(); ++j) {
Curve curve = get_curve(j);
for (size_t k = 0; k < curve.num_segments(); ++k, ++index) {
prim_visibility[index] = visibility;
}
}
}
unsigned int *prim_tri_index = &pack->prim_tri_index[optix_prim_offset];
int *prim_type = &pack->prim_type[optix_prim_offset];
unsigned int *prim_visibility = &pack->prim_visibility[optix_prim_offset];
int *prim_index = &pack->prim_index[optix_prim_offset];
int *prim_object = &pack->prim_object[optix_prim_offset];
// 'pack->prim_time' is unused by Embree and OptiX
if ((pack_flags & PACK_GEOMETRY) != 0) {
unsigned int *prim_tri_index = &pack->prim_tri_index[optix_prim_offset];
int *prim_type = &pack->prim_type[optix_prim_offset];
int *prim_index = &pack->prim_index[optix_prim_offset];
int *prim_object = &pack->prim_object[optix_prim_offset];
// 'pack->prim_time' is unused by Embree and OptiX
uint type = has_motion_blur() ?
((curve_shape == CURVE_RIBBON) ? PRIMITIVE_MOTION_CURVE_RIBBON :
PRIMITIVE_MOTION_CURVE_THICK) :
((curve_shape == CURVE_RIBBON) ? PRIMITIVE_CURVE_RIBBON : PRIMITIVE_CURVE_THICK);
uint type = has_motion_blur() ?
((curve_shape == CURVE_RIBBON) ? PRIMITIVE_MOTION_CURVE_RIBBON :
PRIMITIVE_MOTION_CURVE_THICK) :
((curve_shape == CURVE_RIBBON) ? PRIMITIVE_CURVE_RIBBON :
PRIMITIVE_CURVE_THICK);
size_t index = 0;
for (size_t j = 0; j < num_curves(); ++j) {
Curve curve = get_curve(j);
for (size_t k = 0; k < curve.num_segments(); ++k, ++index) {
prim_tri_index[index] = -1;
prim_type[index] = PRIMITIVE_PACK_SEGMENT(type, k);
prim_visibility[index] = visibility;
// Each curve segment points back to its curve index
prim_index[index] = j + prim_offset;
prim_object[index] = object;
size_t index = 0;
for (size_t j = 0; j < num_curves(); ++j) {
Curve curve = get_curve(j);
for (size_t k = 0; k < curve.num_segments(); ++k, ++index) {
prim_tri_index[index] = -1;
prim_type[index] = PRIMITIVE_PACK_SEGMENT(type, k);
// Each curve segment points back to its curve index
prim_index[index] = j + prim_offset;
prim_object[index] = object;
}
}
}
}

View File

@@ -146,7 +146,10 @@ class Hair : public Geometry {
/* BVH */
void pack_curves(Scene *scene, float4 *curve_key_co, float4 *curve_data, size_t curvekey_offset);
void pack_primitives(PackedBVH *pack, int object, uint visibility, bool pack_all) override;
void pack_primitives(PackedBVH *pack,
int object,
uint visibility,
PackFlags pack_flags) override;
};
CCL_NAMESPACE_END

View File

@@ -805,7 +805,7 @@ void Mesh::pack_patches(uint *patch_data, uint vert_offset, uint face_offset, ui
}
}
void Mesh::pack_primitives(ccl::PackedBVH *pack, int object, uint visibility, bool pack_all)
void Mesh::pack_primitives(ccl::PackedBVH *pack, int object, uint visibility, PackFlags pack_flags)
{
if (triangles.empty())
return;
@@ -819,28 +819,38 @@ void Mesh::pack_primitives(ccl::PackedBVH *pack, int object, uint visibility, bo
uint type = has_motion_blur() ? PRIMITIVE_MOTION_TRIANGLE : PRIMITIVE_TRIANGLE;
if (pack_all) {
/* Use optix_prim_offset for indexing as those arrays also contain data for Hair geometries. */
unsigned int *prim_tri_index = &pack->prim_tri_index[optix_prim_offset];
int *prim_type = &pack->prim_type[optix_prim_offset];
/* Separate loop as other arrays are not initialized if their packing is not required. */
if ((pack_flags & PackFlags::PACK_VISIBILITY) != 0) {
unsigned int *prim_visibility = &pack->prim_visibility[optix_prim_offset];
int *prim_index = &pack->prim_index[optix_prim_offset];
int *prim_object = &pack->prim_object[optix_prim_offset];
for (size_t k = 0; k < num_prims; ++k) {
prim_tri_index[k] = (prim_offset + k) * 3;
prim_type[k] = type;
prim_index[k] = prim_offset + k;
prim_object[k] = object;
prim_visibility[k] = visibility;
}
}
for (size_t k = 0; k < num_prims; ++k) {
const Mesh::Triangle t = get_triangle(k);
prim_tri_verts[k * 3] = float3_to_float4(verts[t.v[0]]);
prim_tri_verts[k * 3 + 1] = float3_to_float4(verts[t.v[1]]);
prim_tri_verts[k * 3 + 2] = float3_to_float4(verts[t.v[2]]);
if ((pack_flags & PackFlags::PACK_GEOMETRY) != 0) {
/* Use optix_prim_offset for indexing as those arrays also contain data for Hair geometries. */
unsigned int *prim_tri_index = &pack->prim_tri_index[optix_prim_offset];
int *prim_type = &pack->prim_type[optix_prim_offset];
int *prim_index = &pack->prim_index[optix_prim_offset];
int *prim_object = &pack->prim_object[optix_prim_offset];
for (size_t k = 0; k < num_prims; ++k) {
if ((pack_flags & PackFlags::PACK_GEOMETRY) != 0) {
prim_tri_index[k] = (prim_offset + k) * 3;
prim_type[k] = type;
prim_index[k] = prim_offset + k;
prim_object[k] = object;
}
}
}
if ((pack_flags & PackFlags::PACK_VERTICES) != 0) {
for (size_t k = 0; k < num_prims; ++k) {
const Mesh::Triangle t = get_triangle(k);
prim_tri_verts[k * 3] = float3_to_float4(verts[t.v[0]]);
prim_tri_verts[k * 3 + 1] = float3_to_float4(verts[t.v[1]]);
prim_tri_verts[k * 3 + 2] = float3_to_float4(verts[t.v[2]]);
}
}
}

View File

@@ -232,7 +232,10 @@ class Mesh : public Geometry {
size_t tri_offset);
void pack_patches(uint *patch_data, uint vert_offset, uint face_offset, uint corner_offset);
void pack_primitives(PackedBVH *pack, int object, uint visibility, bool pack_all) override;
void pack_primitives(PackedBVH *pack,
int object,
uint visibility,
PackFlags pack_flags) override;
void tessellate(DiagSplit *split);

View File

@@ -220,6 +220,10 @@ void Object::tag_update(Scene *scene)
flag |= ObjectManager::TRANSFORM_MODIFIED;
}
if (visibility_is_modified()) {
flag |= ObjectManager::VISIBILITY_MODIFIED;
}
foreach (Node *node, geometry->get_used_shaders()) {
Shader *shader = static_cast<Shader *>(node);
if (shader->get_use_mis() && shader->has_surface_emission)
@@ -914,6 +918,10 @@ void ObjectManager::tag_update(Scene *scene, uint32_t flag)
geometry_flag |= GeometryManager::TRANSFORM_MODIFIED;
}
if ((flag & VISIBILITY_MODIFIED) != 0) {
geometry_flag |= GeometryManager::VISIBILITY_MODIFIED;
}
scene->geometry_manager->tag_update(scene, geometry_flag);
}

View File

@@ -134,6 +134,7 @@ class ObjectManager {
OBJECT_MODIFIED = (1 << 5),
HOLDOUT_MODIFIED = (1 << 6),
TRANSFORM_MODIFIED = (1 << 7),
VISIBILITY_MODIFIED = (1 << 8),
/* tag everything in the manager for an update */
UPDATE_ALL = ~0u,

View File

@@ -542,9 +542,6 @@ bool Scene::update(Progress &progress, bool &kernel_switch_needed)
DeviceKernelStatus kernel_switch_status = device->get_active_kernel_switch_state();
kernel_switch_needed = kernel_switch_status == DEVICE_KERNEL_FEATURE_KERNEL_AVAILABLE ||
kernel_switch_status == DEVICE_KERNEL_FEATURE_KERNEL_INVALID;
if (kernel_switch_status == DEVICE_KERNEL_WAITING_FOR_FEATURE_KERNEL) {
progress.set_kernel_status("Compiling render kernels");
}
if (new_kernels_needed || kernel_switch_needed) {
progress.set_kernel_status("Compiling render kernels");
device->wait_for_availability(loaded_kernel_features);

Some files were not shown because too many files have changed in this diff Show More