Files
test/source/blender/gpu/intern/gpu_framebuffer_private.hh
Clément Foucault 4a4b1482cd GL: Add subpass input emulation
This adds basic emulation of the subpass input feature
of vulkan and to a lower extend Raster Order Group on Metal.

This help test paths that might use this feature in the future
(like shadow rendering) on all platform and or simplify higher
level code for supporting older hardware.

This add clear description to the load/store ops and to the
new `GPUAttachementState`.

The OpenGL backend will correctly mask un-writable
attachments and will bind as texture readable attachments.

Even if possible by the vulkan standard, the GPU API prohibit
the read and write to the same attachment inside the same
subpass.

In the GL backend, this is implemented using `glTextureBarrier`
and `texelFetch` as it is described in the ARB_texture_barrier
extension.
https://registry.khronos.org/OpenGL/extensions/ARB/ARB_texture_barrier.txt

Pull Request: https://projects.blender.org/blender/blender/pulls/112051
2023-10-01 15:27:21 +02:00

257 lines
6.5 KiB
C++

/* SPDX-FileCopyrightText: 2020 Blender Authors
*
* SPDX-License-Identifier: GPL-2.0-or-later */
/** \file
* \ingroup gpu
*
* Private frame buffer API.
*/
#pragma once
#include "BLI_math_vector.h"
#include "BLI_span.hh"
#include "MEM_guardedalloc.h"
#include "GPU_framebuffer.h"
struct GPUTexture;
typedef enum GPUAttachmentType : int {
GPU_FB_DEPTH_ATTACHMENT = 0,
GPU_FB_DEPTH_STENCIL_ATTACHMENT,
GPU_FB_COLOR_ATTACHMENT0,
GPU_FB_COLOR_ATTACHMENT1,
GPU_FB_COLOR_ATTACHMENT2,
GPU_FB_COLOR_ATTACHMENT3,
GPU_FB_COLOR_ATTACHMENT4,
GPU_FB_COLOR_ATTACHMENT5,
GPU_FB_COLOR_ATTACHMENT6,
GPU_FB_COLOR_ATTACHMENT7,
/* Number of maximum output slots. */
/* Keep in mind that GL max is GL_MAX_DRAW_BUFFERS and is at least 8, corresponding to
* the maximum number of COLOR attachments specified by glDrawBuffers. */
GPU_FB_MAX_ATTACHMENT,
} GPUAttachmentType;
#define GPU_FB_MAX_COLOR_ATTACHMENT (GPU_FB_MAX_ATTACHMENT - GPU_FB_COLOR_ATTACHMENT0)
inline constexpr GPUAttachmentType operator-(GPUAttachmentType a, int b)
{
return static_cast<GPUAttachmentType>(int(a) - b);
}
inline constexpr GPUAttachmentType operator+(GPUAttachmentType a, int b)
{
return static_cast<GPUAttachmentType>(int(a) + b);
}
inline GPUAttachmentType &operator++(GPUAttachmentType &a)
{
a = a + 1;
return a;
}
inline GPUAttachmentType &operator--(GPUAttachmentType &a)
{
a = a - 1;
return a;
}
namespace blender {
namespace gpu {
#ifdef DEBUG
# define DEBUG_NAME_LEN 64
#else
# define DEBUG_NAME_LEN 16
#endif
class FrameBuffer {
protected:
/** Set of texture attachments to render to. DEPTH and DEPTH_STENCIL are mutually exclusive. */
GPUAttachment attachments_[GPU_FB_MAX_ATTACHMENT];
/** Is true if internal representation need to be updated. */
bool dirty_attachments_ = true;
/** Size of attachment textures. */
int width_ = 0, height_ = 0;
/** Debug name. */
char name_[DEBUG_NAME_LEN];
/** Frame-buffer state. */
int viewport_[GPU_MAX_VIEWPORTS][4] = {{0}};
int scissor_[4] = {0};
bool multi_viewport_ = false;
bool scissor_test_ = false;
bool dirty_state_ = true;
#ifndef GPU_NO_USE_PY_REFERENCES
public:
/**
* Reference of a pointer that needs to be cleaned when deallocating the frame-buffer.
* Points to #BPyGPUFrameBuffer.fb
*/
void **py_ref = nullptr;
#endif
public:
FrameBuffer(const char *name);
virtual ~FrameBuffer();
virtual void bind(bool enabled_srgb) = 0;
virtual bool check(char err_out[256]) = 0;
virtual void clear(eGPUFrameBufferBits buffers,
const float clear_col[4],
float clear_depth,
uint clear_stencil) = 0;
virtual void clear_multi(const float (*clear_col)[4]) = 0;
virtual void clear_attachment(GPUAttachmentType type,
eGPUDataFormat data_format,
const void *clear_value) = 0;
virtual void attachment_set_loadstore_op(GPUAttachmentType type, GPULoadStore ls) = 0;
virtual void read(eGPUFrameBufferBits planes,
eGPUDataFormat format,
const int area[4],
int channel_len,
int slot,
void *r_data) = 0;
virtual void blit_to(eGPUFrameBufferBits planes,
int src_slot,
FrameBuffer *dst,
int dst_slot,
int dst_offset_x,
int dst_offset_y) = 0;
virtual void subpass_transition(const GPUAttachmentState depth_attachment_state,
Span<GPUAttachmentState> color_attachment_states) = 0;
void load_store_config_array(const GPULoadStore *load_store_actions, uint actions_len);
void attachment_set(GPUAttachmentType type, const GPUAttachment &new_attachment);
void attachment_remove(GPUAttachmentType type);
void recursive_downsample(int max_lvl,
void (*callback)(void *user_data, int level),
void *user_data);
uint get_bits_per_pixel();
/* Sets the size after creation. */
inline void size_set(int width, int height)
{
width_ = width;
height_ = height;
dirty_state_ = true;
}
/* Sets the size for frame-buffer with no attachments. */
inline void default_size_set(int width, int height)
{
width_ = width;
height_ = height;
dirty_attachments_ = true;
dirty_state_ = true;
}
inline void viewport_set(const int viewport[4])
{
if (!equals_v4v4_int(viewport_[0], viewport)) {
copy_v4_v4_int(viewport_[0], viewport);
dirty_state_ = true;
}
multi_viewport_ = false;
}
inline void viewport_multi_set(const int viewports[GPU_MAX_VIEWPORTS][4])
{
for (size_t i = 0; i < GPU_MAX_VIEWPORTS; i++) {
if (!equals_v4v4_int(viewport_[i], viewports[i])) {
copy_v4_v4_int(viewport_[i], viewports[i]);
dirty_state_ = true;
}
}
multi_viewport_ = true;
}
inline void scissor_set(const int scissor[4])
{
if (!equals_v4v4_int(scissor_, scissor)) {
copy_v4_v4_int(scissor_, scissor);
dirty_state_ = true;
}
}
inline void scissor_test_set(bool test)
{
scissor_test_ = test;
}
inline void viewport_get(int r_viewport[4]) const
{
copy_v4_v4_int(r_viewport, viewport_[0]);
}
inline void scissor_get(int r_scissor[4]) const
{
copy_v4_v4_int(r_scissor, scissor_);
}
inline bool scissor_test_get() const
{
return scissor_test_;
}
inline void viewport_reset()
{
int viewport_rect[4] = {0, 0, width_, height_};
viewport_set(viewport_rect);
}
inline void scissor_reset()
{
int scissor_rect[4] = {0, 0, width_, height_};
scissor_set(scissor_rect);
}
inline GPUTexture *depth_tex() const
{
if (attachments_[GPU_FB_DEPTH_ATTACHMENT].tex) {
return attachments_[GPU_FB_DEPTH_ATTACHMENT].tex;
}
return attachments_[GPU_FB_DEPTH_STENCIL_ATTACHMENT].tex;
};
inline GPUTexture *color_tex(int slot) const
{
return attachments_[GPU_FB_COLOR_ATTACHMENT0 + slot].tex;
};
inline const char *const name_get() const
{
return name_;
};
};
/* Syntactic sugar. */
static inline GPUFrameBuffer *wrap(FrameBuffer *vert)
{
return reinterpret_cast<GPUFrameBuffer *>(vert);
}
static inline FrameBuffer *unwrap(GPUFrameBuffer *vert)
{
return reinterpret_cast<FrameBuffer *>(vert);
}
static inline const FrameBuffer *unwrap(const GPUFrameBuffer *vert)
{
return reinterpret_cast<const FrameBuffer *>(vert);
}
#undef DEBUG_NAME_LEN
} // namespace gpu
} // namespace blender