blender
/
blender
134
398
Fork 580
Code Issues 6.3k Pull Requests 755 Projects 19 Wiki Activity

Compositor: Refactor File Output node #113982

Merged
Omar Emara merged 15 commits from OmarEmaraDev/blender:file-output-node-refactor into main 2023-12-13 11:08:12 +01:00
Conversation 9 Commits 15 Files Changed 36 +1294 -1271
Omar Emara commented 2023-10-20 20:41:26 +02:00
Member
Copy Link

This patches refactors the compositor File Output mechanism and
implements the file output node for the Realtime Compositor. The
refactor was done for the following reasons:

  1. The existing file output mechanism relied on a global EXR image
    resource where the result of each compositor execution for each
    view was accumulated and stored in the global resource, until the
    last view is executed, when the EXR is finally saved. Aside from
    relying on global resources, this can cause effective memory leaks
    since the compositor can be interrupted before the EXR is written and
    closed.
  2. We need common code to share between all compositors since we now
    have multiple compositor implementations.
  3. We needed to take the opportunity to fix some of the issues with the
    existing implementation, like the inability to save 4D vectors, lossy
    compression of data passes, and inability to save single values
    passes.

The refactor first introduced a new structure called the Compositor
Render Context. This context stores compositor information related to
the render pipeline and is persistent across all compositor executions
of all views. Its extended lifetime relative to a single compositor
execution lends itself well to store data that is accumulated across
views. The context currently has a map of File Output objects. Those
objects wrap a Render Result structure and can be used to construct
multi-view images which can then be saved after all views are executed
using the existing BKE_image_render_write function.

Minor adjustments were made to the BKE and RE modules to allow saving
using the BKE_image_render_write function. Namely, the function now
allows the use of a source image format for saving as well as the
ability to not save the render result as a render by introducing two new
default arguments. Further, for multi-layer EXR saving, the existent of
a single unnamed render layer will omit the layer name from the EXR
channel full name, and only the pass, view, and channel ID will remain.
Finally, the Render Result to Image Buffer conversion now take he number
of channels into account, instead of always assuming color channels.

The patch implements the File Output node in the Realtime Compositor
using the aforementioned mechanisms, replaces the implementation of the
CPU compositor using the same Realtime Compositor implementation, and
setup the necessary logic in the render pipeline code.

Depends on #114339.

This patches refactors the compositor File Output mechanism and implements the file output node for the Realtime Compositor. The refactor was done for the following reasons: 1. The existing file output mechanism relied on a global EXR image resource where the result of each compositor execution for each view was accumulated and stored in the global resource, until the last view is executed, when the EXR is finally saved. Aside from relying on global resources, this can cause effective memory leaks since the compositor can be interrupted before the EXR is written and closed. 2. We need common code to share between all compositors since we now have multiple compositor implementations. 3. We needed to take the opportunity to fix some of the issues with the existing implementation, like the inability to save 4D vectors, lossy compression of data passes, and inability to save single values passes. The refactor first introduced a new structure called the Compositor Render Context. This context stores compositor information related to the render pipeline and is persistent across all compositor executions of all views. Its extended lifetime relative to a single compositor execution lends itself well to store data that is accumulated across views. The context currently has a map of File Output objects. Those objects wrap a Render Result structure and can be used to construct multi-view images which can then be saved after all views are executed using the existing BKE_image_render_write function. Minor adjustments were made to the BKE and RE modules to allow saving using the BKE_image_render_write function. Namely, the function now allows the use of a source image format for saving as well as the ability to not save the render result as a render by introducing two new default arguments. Further, for multi-layer EXR saving, the existent of a single unnamed render layer will omit the layer name from the EXR channel full name, and only the pass, view, and channel ID will remain. Finally, the Render Result to Image Buffer conversion now take he number of channels into account, instead of always assuming color channels. The patch implements the File Output node in the Realtime Compositor using the aforementioned mechanisms, replaces the implementation of the CPU compositor using the same Realtime Compositor implementation, and setup the necessary logic in the render pipeline code. Depends on #114339.
❤️ 1
Omar Emara added 1 commit 2023-10-20 20:41:38 +02:00
Omar Emara added 1 commit 2023-10-23 20:47:34 +02:00
Omar Emara added 1 commit 2023-10-24 09:30:35 +02:00
Omar Emara changed title from WIP: Compositor: Refactor File Output node to Compositor: Refactor File Output node 2023-10-24 09:58:45 +02:00
Omar Emara requested review from Sergey Sharybin 2023-10-24 09:58:59 +02:00
Omar Emara added the
Module
VFX & Video
Interest
Compositing
 labels 2023-10-24 09:59:14 +02:00
Omar Emara added 1 commit 2023-10-24 10:19:55 +02:00
Omar Emara added 1 commit 2023-10-25 12:24:39 +02:00
Sergey Sharybin reviewed 2023-11-08 16:20:20 +01:00
Sergey Sharybin left a comment
Owner
Copy Link

Just a quick pass of code review. On a design description level it sounds reasonable. On the code side it nice to see that there is no duplicated tricky logic. You even managed to delete more lines than add!

One of the big things I am confused about is that it seems vector are stored as 4 elements now? Is my understanding correct? If yes, why is it so?

Just a quick pass of code review. On a design description level it sounds reasonable. On the code side it nice to see that there is no duplicated tricky logic. You even managed to delete more lines than add! One of the big things I am confused about is that it seems vector are stored as 4 elements now? Is my understanding correct? If yes, why is it so?
source/blender/blenkernel/BKE_image_save.h Outdated
@ -81,2 +81,3 @@
const bool stamp,
const char *filepath_basis);
const char *filepath_basis,
const struct ImageFormatData *source_format = nullptr,
Sergey Sharybin commented 2023-11-08 14:26:18 +01:00
Owner
Copy Link

Would be nice the explain what source_format is for in the comment above.

Would be nice the explain what `source_format` is for in the comment above.
OmarEmaraDev marked this conversation as resolved
source/blender/blenkernel/intern/image_save.cc Outdated
@ -785,6 +785,7 @@ bool BKE_image_render_write_exr(ReportList *reports,
/* Other render layers. */
int nr = (rr->have_combined) ? 1 : 0;
const int layers_count = BLI_listbase_count(&rr->layers);
Sergey Sharybin commented 2023-11-08 14:24:48 +01:00
Owner
Copy Link

Small optimization: const bool has_multiple_layers = BLI_listbase_count_at_most(&rr->layers, 2) > 1;

Small optimization: `const bool has_multiple_layers = BLI_listbase_count_at_most(&rr->layers, 2) > 1;`
OmarEmaraDev marked this conversation as resolved
source/blender/compositor/operations/COM_FileOutputOperation.cc Outdated
@ -0,0 +45,4 @@
{
int size = get_channels_count(datatype);
return static_cast<float *>(
MEM_callocN(width * height * size * sizeof(float), "File Output Buffer."));
Sergey Sharybin commented 2023-11-08 14:30:00 +01:00
Owner
Copy Link

MEM_malloc_array(size_t(width) * height * size, sizeof(float), "File Output Buffer").

This will reduce the possible overflow issues. Can also move size to the second argument.

`MEM_malloc_array(size_t(width) * height * size, sizeof(float), "File Output Buffer").` This will reduce the possible overflow issues. Can also move `size` to the second argument.
OmarEmaraDev marked this conversation as resolved
source/blender/compositor/operations/COM_FileOutputOperation.cc
@ -0,0 +160,4 @@
}
}
/* --------------------
Sergey Sharybin commented 2023-11-08 15:57:11 +01:00
Owner
Copy Link

Not sure if that is from an existing code. We have a different way in the guideline defining such code sections: https://wiki.blender.org/wiki/Style_Guide/C_Cpp#Comments (scroll down a bit to the Comment Sections)

Not sure if that is from an existing code. We have a different way in the guideline defining such code sections: https://wiki.blender.org/wiki/Style_Guide/C_Cpp#Comments (scroll down a bit to the Comment Sections)
OmarEmaraDev marked this conversation as resolved
Sergey Sharybin reviewed 2023-11-08 16:20:21 +01:00
Sergey Sharybin left a comment
Owner
Copy Link

Just a quick pass of code review. On a design description level it sounds reasonable. On the code side it nice to see that there is no duplicated tricky logic. You even managed to delete more lines than add!

One of the big things I am confused about is that it seems vector are stored as 4 elements now? Is my understanding correct? If yes, why is it so?

Just a quick pass of code review. On a design description level it sounds reasonable. On the code side it nice to see that there is no duplicated tricky logic. You even managed to delete more lines than add! One of the big things I am confused about is that it seems vector are stored as 4 elements now? Is my understanding correct? If yes, why is it so?
Omar Emara commented 2023-11-09 08:40:36 +01:00
Author
Member
Copy Link

@Sergey They are stored as 4 elements but not necessarily saved as such. The user can choose the RGB color format to save them as three element vectors or the user can choose RGBA color format to save them as four element vectors.

@Sergey They are stored as 4 elements but not necessarily saved as such. The user can choose the RGB color format to save them as three element vectors or the user can choose RGBA color format to save them as four element vectors.
Omar Emara added 3 commits 2023-11-13 13:43:02 +01:00
Omar Emara added 2 commits 2023-11-15 16:52:53 +01:00
Omar Emara added 1 commit 2023-11-15 17:38:34 +01:00
Sergey Sharybin commented 2023-11-20 18:24:41 +01:00
Owner
Copy Link

The user can choose the RGB color format to save them as three element vectors or the user can choose RGBA color format to save them as four element vectors.

But this is not something that is possible to control on a per-pass level when the File Output is using Multilayer EXR?

I've prepared a simple file. Before this change the vector socket is stored as Image_001.{X,Y,Z} it is stored as Image_001.{r,g,b,a}. It is nice to better support Vector pass output, but having extra component for Normal pass sounds a bit wasteful. The change also "breaks" channel naming in the EXR (not immediately sure if it is really bad, but it feels so from the interop perspective).

Not sure we should be doing such changes as part of a refactor, and adding support of the node to the GPU compositor. It feels like more proper would be to have 2,3, and 4- element vectors as socket types, and derive channel naming from them. So that Normal is stored as {x,y,z}, Vector is stored as {r,g,b,a}, and UV is stored as {x,y}.

P.S. The alpha handling for Vector is different in tiled and full-frame and GPU: tiled initializes alpha to 1, the latter two initialize it to 0.

> The user can choose the RGB color format to save them as three element vectors or the user can choose RGBA color format to save them as four element vectors. But this is not something that is possible to control on a per-pass level when the File Output is using Multilayer EXR? I've prepared a simple file. Before this change the vector socket is stored as `Image_001.{X,Y,Z}` it is stored as `Image_001.{r,g,b,a}`. It is nice to better support `Vector` pass output, but having extra component for `Normal` pass sounds a bit wasteful. The change also "breaks" channel naming in the EXR (not immediately sure if it is really bad, but it feels so from the interop perspective). Not sure we should be doing such changes as part of a refactor, and adding support of the node to the GPU compositor. It feels like more proper would be to have 2,3, and 4- element vectors as socket types, and derive channel naming from them. So that Normal is stored as {x,y,z}, Vector is stored as {r,g,b,a}, and UV is stored as {x,y}. P.S. The alpha handling for Vector is different in tiled and full-frame and GPU: tiled initializes alpha to 1, the latter two initialize it to 0.
file_output.blend
999 KiB
Omar Emara commented 2023-11-20 18:39:18 +01:00
Author
Member
Copy Link

But this is not something that is possible to control on a per-pass level when the File Output is using Multilayer EXR?

I see. I missed that case.

@Sergey Okay, so I will restore three-components XYZ vectors and split the patch into a refactor and a GPU implementation.

> But this is not something that is possible to control on a per-pass level when the File Output is using Multilayer EXR? I see. I missed that case. @Sergey Okay, so I will restore three-components XYZ vectors and split the patch into a refactor and a GPU implementation.
Omar Emara added 2 commits 2023-11-22 17:27:08 +01:00
buildbot/vexp-code-patch-lint Build done. Details
buildbot/vexp-code-patch-linux-x86_64 Build done. Details
buildbot/vexp-code-patch-darwin-x86_64 Build done. Details
buildbot/vexp-code-patch-windows-amd64 Build done. Details
buildbot/vexp-code-patch-darwin-arm64 Build done. Details
buildbot/vexp-code-patch-coordinator Build done. Details
60fa0d5bfd Revert to using XYZ vector passes
Sergey Sharybin requested review from Brecht Van Lommel 2023-12-01 17:48:10 +01:00
Brecht Van Lommel approved these changes 2023-12-12 21:00:18 +01:00
Brecht Van Lommel left a comment
Owner
Copy Link

I tested various different types of passes, file formats, stereo, all seems to be working well.

And almost the same lines of code as before, nice.

I tested various different types of passes, file formats, stereo, all seems to be working well. And almost the same lines of code as before, nice.
source/blender/blenkernel/intern/image_save.cc Outdated
@ -922,2 +933,3 @@
const bool stamp,
const char *filepath_basis)
const char *filepath_basis,
const ImageFormatData *source_format,
Brecht Van Lommel commented 2023-12-12 16:23:26 +01:00
Owner
Copy Link

I would name this just format. It is the file format that will be written, not of the source image or something like that.

I guess this was taken from BKE_image_format_init_for_write, but in that context it means that the source data initialized image_format from.

I would name this just `format`. It is the file format that will be written, not of the source image or something like that. I guess this was taken from `BKE_image_format_init_for_write`, but in that context it means that the source data initialized `image_format` from.
source/blender/nodes/composite/nodes/node_composite_file_output.cc
@ -0,0 +15,4 @@
#include "BLI_string.h"
#include "BLI_string_utf8.h"
#include "BLI_string_utils.hh"
#include "BLI_task.hh"
Brecht Van Lommel commented 2023-12-12 20:51:13 +01:00
Owner
Copy Link

This include seems unnecessary, no multithreading so far.

This include seems unnecessary, no multithreading so far.
Brecht Van Lommel commented 2023-12-12 21:01:09 +01:00
Owner
Copy Link

Oops, I meant to delete this comment.

Oops, I meant to delete this comment.
brecht marked this conversation as resolved
Omar Emara commented 2023-12-13 09:21:30 +01:00
Author
Member
Copy Link

@blender-bot build

@blender-bot build
Omar Emara added 1 commit 2023-12-13 10:21:36 +01:00
buildbot/vexp-code-patch-lint Build done. Details
buildbot/vexp-code-patch-linux-x86_64 Build done. Details
buildbot/vexp-code-patch-darwin-arm64 Build done. Details
buildbot/vexp-code-patch-darwin-x86_64 Build done. Details
buildbot/vexp-code-patch-windows-amd64 Build done. Details
buildbot/vexp-code-patch-coordinator Build done. Details
9a75f85e5a Merge branch 'main' into file-output-node-refactor
Omar Emara commented 2023-12-13 10:23:00 +01:00
Author
Member
Copy Link

@blender-bot build

@blender-bot build
Omar Emara added 1 commit 2023-12-13 11:06:58 +01:00
Omar Emara merged commit 931c188ce5 into main 2023-12-13 11:08:12 +01:00
Omar Emara deleted branch file-output-node-refactor 2023-12-13 11:08:14 +01:00
Lsscpp commented 2023-12-20 10:06:33 +01:00
First-time contributor
Copy Link

I wonder if (speaking about refactoring) it wouldn't make more sense to refactor also the output panel giving the chance to add multiple formats and not just one

I wonder if (speaking about refactoring) it wouldn't make more sense to refactor also the output panel giving the chance to add multiple formats and not just one
Omar Emara commented 2023-12-20 12:41:42 +01:00
Author
Member
Copy Link

@lsscpp This is already possible, just uncheck "Use Node Format" for the inputs and control each of their formats independently.

@lsscpp This is already possible, just uncheck "Use Node Format" for the inputs and control each of their formats independently.
Lsscpp commented 2023-12-20 13:45:58 +01:00
First-time contributor
Copy Link

I meant in the property editor, which I know is not related to nodes

I meant in the property editor, which I know is not related to nodes
Omar Emara commented 2023-12-20 14:16:29 +01:00
Author
Member
Copy Link

I see. Not sure if that would be a desirable feature. In any case, it is not really part of the compositor module so I am not going to personally look into it in the near future. Also, RCS and DevTalk are better suited for such requests/discussions.

I see. Not sure if that would be a desirable feature. In any case, it is not really part of the compositor module so I am not going to personally look into it in the near future. Also, RCS and DevTalk are better suited for such requests/discussions.
Sign in to join this conversation.
Reviewers
No reviewers
Sergey Sharybin
Brecht Van Lommel
Labels
Clear labels   
Interest
Alembic

  
Interest
Animation & Rigging

  
Interest
Asset Browser

  
Interest
Asset Browser Project Overview

  
Interest
Audio

  
Interest
Automated Testing

  
Interest
Blender Asset Bundle

  
Interest
BlendFile

  
Interest
Collada

  
Interest
Compatibility
This issue affects/is about backward or forward compatibility

  
Interest
Compositing

  
Interest
Core

  
Interest
Cycles

  
Interest
Dependency Graph

  
Interest
Development Management

  
Interest
EEVEE

  
Interest
EEVEE & Viewport

  
Interest
Freestyle

  
Interest
Geometry Nodes

  
Interest
Grease Pencil

  
Interest
ID Management

  
Interest
Images & Movies

  
Interest
Import Export

  
Interest
Line Art

  
Interest
Masking

  
Interest
Metal

  
Interest
Modeling

  
Interest
Modifiers

  
Interest
Motion Tracking

  
Interest
Nodes & Physics

  
Interest
OpenGL

  
Interest
Overlay

  
Interest
Overrides

  
Interest
Performance

  
Interest
Physics

  
Interest
Pipeline, Assets & IO

  
Interest
Platforms, Builds & Tests

  
Interest
Python API

  
Interest
Render & Cycles

  
Interest
Render Pipeline

  
Interest
Sculpt, Paint & Texture

  
Interest
Text Editor

  
Interest
Translations

  
Interest
Triaging

  
Interest
Undo

  
Interest
USD

  
Interest
User Interface

  
Interest
UV Editing

  
Interest
VFX & Video

  
Interest
Video Sequencer

  
Interest
Virtual Reality

  
Interest
Vulkan

  
Interest
Wayland
Issues relating to Wayland windowing support.

  
Interest
Workbench

  
Interest: X11

Issues relating to Xorg/X11 support.

  
Legacy
Blender 2.8 Project

  
Legacy
Milestone 1: Basic, Local Asset Browser

  
Legacy
OpenGL Error

  
Meta
Good First Issue

  
Meta
Papercut

  
Meta
Retrospective

  
Meta
Security
Issues relating to security: https://wiki.blender.org/wiki/Process/Vulnerability_Reports

  
Module
Animation & Rigging

  
Module
Core

  
Module
Development Management

  
Module
EEVEE & Viewport

  
Module
Grease Pencil

  
Module
Modeling

  
Module
Nodes & Physics

  
Module
Pipeline, Assets & IO

  
Module
Platforms, Builds & Tests

  
Module
Python API

  
Module
Render & Cycles

  
Module
Sculpt, Paint & Texture

  
Module
Triaging

  
Module
User Interface

  
Module
VFX & Video

  
Platform
FreeBSD

  
Platform
Linux

  
Platform
macOS

  
Platform
Windows

  
Priority
High

  
Priority
Low

  
Priority
Normal

  
Priority
Unbreak Now!

  
Status
Archived

  
Status
Confirmed

  
Status
Duplicate

  
Status
Needs Info from Developers

  
Status
Needs Information from User

  
Status
Needs Triage

  
Status
Resolved

  
Type
Bug

  
Type
Design

  
Type
Known Issue

  
Type
Patch

  
Type
Report

  
Type
To Do

No Label
Interest
Alembic
Interest
Animation & Rigging
Interest
Asset Browser
Interest
Asset Browser Project Overview
Interest
Audio
Interest
Automated Testing
Interest
Blender Asset Bundle
Interest
BlendFile
Interest
Collada
Interest
Compatibility
Interest
Compositing
Interest
Core
Interest
Cycles
Interest
Dependency Graph
Interest
Development Management
Interest
EEVEE
Interest
EEVEE & Viewport
Interest
Freestyle
Interest
Geometry Nodes
Interest
Grease Pencil
Interest
ID Management
Interest
Images & Movies
Interest
Import Export
Interest
Line Art
Interest
Masking
Interest
Metal
Interest
Modeling
Interest
Modifiers
Interest
Motion Tracking
Interest
Nodes & Physics
Interest
OpenGL
Interest
Overlay
Interest
Overrides
Interest
Performance
Interest
Physics
Interest
Pipeline, Assets & IO
Interest
Platforms, Builds & Tests
Interest
Python API
Interest
Render & Cycles
Interest
Render Pipeline
Interest
Sculpt, Paint & Texture
Interest
Text Editor
Interest
Translations
Interest
Triaging
Interest
Undo
Interest
USD
Interest
User Interface
Interest
UV Editing
Interest
VFX & Video
Interest
Video Sequencer
Interest
Virtual Reality
Interest
Vulkan
Interest
Wayland
Interest
Workbench
Interest: X11
Legacy
Blender 2.8 Project
Legacy
Milestone 1: Basic, Local Asset Browser
Legacy
OpenGL Error
Meta
Good First Issue
Meta
Papercut
Meta
Retrospective
Meta
Security
Module
Animation & Rigging
Module
Core
Module
Development Management
Module
EEVEE & Viewport
Module
Grease Pencil
Module
Modeling
Module
Nodes & Physics
Module
Pipeline, Assets & IO
Module
Platforms, Builds & Tests
Module
Python API
Module
Render & Cycles
Module
Sculpt, Paint & Texture
Module
Triaging
Module
User Interface
Module
VFX & Video
Platform
FreeBSD
Platform
Linux
Platform
macOS
Platform
Windows
Priority
High
Priority
Low
Priority
Normal
Priority
Unbreak Now!
Status
Archived
Status
Confirmed
Status
Duplicate
Status
Needs Info from Developers
Status
Needs Information from User
Status
Needs Triage
Status
Resolved
Type
Bug
Type
Design
Type
Known Issue
Type
Patch
Type
Report
Type
To Do
Milestone
Clear milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
No Assignees
4 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: blender/blender#113982
No description provided.
Delete Branch "OmarEmaraDev/blender:file-output-node-refactor"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?