Files
test/source/blender/compositor/COM_context.hh

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

178 lines
7.2 KiB
C++
Raw Permalink Normal View History

/* SPDX-FileCopyrightText: 2023 Blender Authors
*
* SPDX-License-Identifier: GPL-2.0-or-later */
#pragma once
#include <cstdint>
#include "BLI_bounds_types.hh"
#include "BLI_enum_flags.hh"
#include "BLI_math_vector_types.hh"
#include "BLI_string_ref.hh"
#include "DNA_scene_types.h"
#include "GPU_shader.hh"
#include "COM_domain.hh"
#include "COM_meta_data.hh"
#include "COM_profiler.hh"
Compositor: Refactor File Output node 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 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. Pull Request: https://projects.blender.org/blender/blender/pulls/113982
2023-12-13 11:08:03 +01:00
#include "COM_render_context.hh"
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
#include "COM_result.hh"
#include "COM_static_cache_manager.hh"
namespace blender::compositor {
/* Enumerates the possible outputs that the compositor can compute. */
enum class OutputTypes : uint8_t {
None = 0,
Composite = 1 << 0,
Viewer = 1 << 1,
FileOutput = 1 << 2,
Previews = 1 << 3,
};
ENUM_OPERATORS(OutputTypes)
/* ------------------------------------------------------------------------------------------------
* Context
*
* A Context is an abstract class that is implemented by the caller of the evaluator to provide the
* necessary data and functionalities for the correct operation of the evaluator. This includes
* providing input data like render passes and the active scene, as well as references to the data
* where the output of the evaluator will be written. Finally, the class have an instance of a
* static resource manager for acquiring cached resources efficiently. */
class Context {
private:
/* A static cache manager that can be used to acquire cached resources for the compositor
* efficiently. */
StaticCacheManager cache_manager_;
public:
/* Get the compositing scene. */
virtual const Scene &get_scene() const = 0;
/* Get the node tree used for compositing. */
virtual const bNodeTree &get_node_tree() const = 0;
/* Returns all output types that should be computed. */
virtual OutputTypes needed_outputs() const = 0;
/* Get the rectangular region representing the area of the input that the compositor will operate
* on. Conversely, the compositor will only update the region of the output that corresponds to
* the compositing region. In the base case, the compositing region covers the entirety of the
* render region. In other cases, the compositing region might be a subset of the render region.
* Callers should check the validity of the region through is_valid_compositing_region(), since
* the region can be zero sized. */
virtual Bounds<int2> get_compositing_region() const = 0;
/* Get the result where the result of the compositor should be written. */
virtual Result get_output(Domain domain) = 0;
/* Get the result where the result of the compositor viewer should be written, given the domain
* of the result to be viewed, its precision, and whether the output is a non-color data image
* that should be displayed without view transform. */
virtual Result get_viewer_output(Domain domain, bool is_data, ResultPrecision precision) = 0;
/* Get the result where the given input is stored. */
virtual Result get_input(StringRef name) = 0;
/* True if the compositor should use GPU acceleration. */
virtual bool use_gpu() const = 0;
/* Get the result where the given pass is stored. */
virtual Result get_pass(const Scene *scene, int view_layer, const char *name);
/* Get the render settings for compositing. This could be different from scene->r render settings
* in case the render size or other settings needs to be overwritten. */
virtual const RenderData &get_render_data() const;
/* Get the name of the view currently being rendered. If the context is not multi-view, return an
* empty string. */
virtual StringRef get_view_name() const;
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
/* Get the precision of the intermediate results of the compositor. */
virtual ResultPrecision get_precision() const;
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
/* Set an info message. This is called by the compositor evaluator to inform or warn the user
* about something, typically an error. The implementation should display the message in an
* appropriate place, which can be directly in the UI or just logged to the output stream. */
virtual void set_info_message(StringRef message) const;
/* True if the compositor should treat viewers as composite outputs because it has no concept of
* or support for viewers. */
virtual bool treat_viewer_as_compositor_output() const;
/* True if the compositor input/output should use output region/bounds setup in the context. */
virtual bool use_context_bounds_for_input_output() const
{
return true;
}
/* Populates the given meta data from the render stamp information of the given render pass. */
virtual void populate_meta_data_for_pass(const Scene *scene,
int view_layer_id,
const char *pass_name,
MetaData &meta_data) const;
Compositor: Refactor File Output node 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 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. Pull Request: https://projects.blender.org/blender/blender/pulls/113982
2023-12-13 11:08:03 +01:00
/* Get a pointer to the render context of this context. A render context stores information about
* the current render. It might be null if the compositor is not being evaluated as part of a
* render pipeline. */
virtual RenderContext *render_context() const;
/* Get a pointer to the profiler of this context. It might be null if the compositor context does
* not support profiling. */
virtual Profiler *profiler() const;
/* Gets called after the evaluation of each compositor operation. See overrides for possible
* uses. */
virtual void evaluate_operation_post() const;
/* Returns true if the compositor evaluation is canceled and that the evaluator should stop
* executing as soon as possible. */
virtual bool is_canceled() const;
/* Resets the context's internal structures like the cache manager. This should be called before
* every evaluation. */
void reset();
/* Get the size of the compositing region. See get_compositing_region(). The output size is
* sanitized such that it is at least 1 in both dimensions. However, the developer is expected to
* gracefully handled zero sizes regions by checking the is_valid_compositing_region method. */
int2 get_compositing_region_size() const;
/* Returns true if the compositing region has a valid size, that is, has at least one pixel in
* both dimensions, returns false otherwise. */
bool is_valid_compositing_region() const;
/* Get the normalized render percentage of the active scene. */
float get_render_percentage() const;
/* Get the current frame number of the active scene. */
int get_frame_number() const;
/* Get the current time in seconds of the active scene. */
float get_time() const;
/* Get the OIDN denoiser quality which should be used if the user doesn't explicitly set
* denoising quality on a node. */
eCompositorDenoiseQaulity get_denoise_quality() const;
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
/* Get a GPU shader with the given info name and precision. */
gpu::Shader *get_shader(const char *info_name, ResultPrecision precision);
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
/* Get a GPU shader with the given info name and context's precision. */
gpu::Shader *get_shader(const char *info_name);
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
/* Create a result of the given type and precision. */
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
Result create_result(ResultType type, ResultPrecision precision);
/* Create a result of the given type using the context's precision. */
Realtime Compositor: Support full precision compositing This patch adds support for full precision compositing for the Realtime Compositor. A new precision option was added to the compositor to change between half and full precision compositing, where the Auto option uses half for the viewport compositor and the interactive render compositor, while full is used for final renders. The compositor context now need to implement the get_precision() method to indicate its preferred precision. Intermediate results will be stored using the context's precision, with a number of exceptions that can use a different precision regardless of the context's precision. For instance, summed area tables are always stored in full float results even if the context specified half float. Conversely, jump flooding tables are always stored in half integer results even if the context specified full. The former requires full float while the latter has no use for it. Since shaders are created for a specific precision, we need two variants of each compositor shader to account for the context's possible precision. However, to avoid doubling the shader info count and reduce boilerplate code and development time, an automated mechanism was employed. A single shader info of whatever precision needs to be added, then, at runtime, the shader info can be adjusted to change the precision of the outputs. That shader variant is then cached in the static cache manager for future processing-free shader retrieval. Therefore, the shader manager was removed in favor of a cached shader container in the static cache manager. A number of utilities were added to make the creation of results as well as the retrieval of shader with the target precision easier. Further, a number of precision-specific shaders were removed in favor of more generic ones that utilizes the aforementioned shader retrieval mechanism. Pull Request: https://projects.blender.org/blender/blender/pulls/113476
2023-11-08 08:32:00 +01:00
Result create_result(ResultType type);
/* Get a reference to the static cache manager of this context. */
StaticCacheManager &cache_manager();
};
} // namespace blender::compositor