Files
test/source/blender/gpu/vulkan/vk_command_buffer.cc

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

446 lines
14 KiB
C++
Raw Normal View History

/* SPDX-FileCopyrightText: 2023 Blender Authors
*
* SPDX-License-Identifier: GPL-2.0-or-later */
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
/** \file
* \ingroup gpu
*/
#include "vk_command_buffer.hh"
#include "vk_buffer.hh"
#include "vk_context.hh"
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
#include "vk_device.hh"
#include "vk_framebuffer.hh"
#include "vk_index_buffer.hh"
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
#include "vk_memory.hh"
#include "vk_pipeline.hh"
#include "vk_storage_buffer.hh"
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
#include "vk_texture.hh"
#include "vk_vertex_buffer.hh"
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
#include "BLI_assert.h"
namespace blender::gpu {
VKCommandBuffer::~VKCommandBuffer()
{
if (vk_device_ != VK_NULL_HANDLE) {
VK_ALLOCATION_CALLBACKS;
vkDestroyFence(vk_device_, vk_fence_, vk_allocation_callbacks);
vk_fence_ = VK_NULL_HANDLE;
}
}
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
bool VKCommandBuffer::is_initialized() const
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
{
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
return vk_command_buffer_ != VK_NULL_HANDLE;
}
void VKCommandBuffer::init(const VKDevice &device)
{
if (is_initialized()) {
return;
}
vk_device_ = device.device_get();
vk_queue_ = device.queue_get();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
Vulkan: Initial Graphics Pipeline Initial graphic pipeline targeting. The goal of this PR is to have an initial graphics pipeline with missing features. It should help identifying areas that requires engineering. Current state is that developers of the GPU module can help with the many smaller pieces that needs to be engineered in order to get it working. It is not intended for users or developers from other modules, but your welcome to learn and give feedback on the code and engineering part. We do expect that large parts of the code still needs to be re-engineered into a more future-proof implementation. **Some highlights**: - In Vulkan the state is kept in the pipeline. Therefore the state is tracked per pipeline. In the near future this could be used as a cache. More research is needed against the default pipeline cache that vulkan already provides. - This PR is based on the work that Kazashi Yoshioka already did. And include work from him in the next areas - Vertex attributes - Vertex data conversions - Pipeline state - Immediate support working. - This PR modifies the VKCommandBuffer to keep track of the framebuffer and its binding state(render pass). Some Vulkan commands require no render pass to be active, other require a render pass. As the order of our commands on API level can not be separated this PR introduces a state engine to keep track of the current state and desired state. This is a temporary solution, the final solution will be proposed when we have a pixel on the screen. At that time I expect that we can design a command encoder that supports all the cases we need. **Notices**: - This branch works on NVIDIA GPUs and has been validated on a Linux system. AMD is known not to work (stalls) and Intel GPUs have not been tested at all. Windows might work but hasn't been validated yet. - The graphics pipeline is implemented with pixels in mind, not with performance. Currently when a draw call is scheduled it is flushed and waited until it is finished drawing, before other draw calls can be scheduled. We expected the performance to be worse that it actually is, but we expect huge performance gains in the future. - Any advanced drawing (that is used by the image editor, compositor or 3d viewport) isn't implemented and might crash when used. - Using multiple windows or resizing of window isn't supported and will stall the system. Pull Request: https://projects.blender.org/blender/blender/pulls/106224
2023-05-11 13:01:56 +02:00
/* When a the last GHOST context is destroyed the device is deallocate. A moment later the GPU
* context is destroyed. The first step is to activate it. Activating would retrieve the device
* from GHOST which in that case is a #VK_NULL_HANDLE. */
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
if (vk_device_ == VK_NULL_HANDLE) {
Vulkan: Initial Graphics Pipeline Initial graphic pipeline targeting. The goal of this PR is to have an initial graphics pipeline with missing features. It should help identifying areas that requires engineering. Current state is that developers of the GPU module can help with the many smaller pieces that needs to be engineered in order to get it working. It is not intended for users or developers from other modules, but your welcome to learn and give feedback on the code and engineering part. We do expect that large parts of the code still needs to be re-engineered into a more future-proof implementation. **Some highlights**: - In Vulkan the state is kept in the pipeline. Therefore the state is tracked per pipeline. In the near future this could be used as a cache. More research is needed against the default pipeline cache that vulkan already provides. - This PR is based on the work that Kazashi Yoshioka already did. And include work from him in the next areas - Vertex attributes - Vertex data conversions - Pipeline state - Immediate support working. - This PR modifies the VKCommandBuffer to keep track of the framebuffer and its binding state(render pass). Some Vulkan commands require no render pass to be active, other require a render pass. As the order of our commands on API level can not be separated this PR introduces a state engine to keep track of the current state and desired state. This is a temporary solution, the final solution will be proposed when we have a pixel on the screen. At that time I expect that we can design a command encoder that supports all the cases we need. **Notices**: - This branch works on NVIDIA GPUs and has been validated on a Linux system. AMD is known not to work (stalls) and Intel GPUs have not been tested at all. Windows might work but hasn't been validated yet. - The graphics pipeline is implemented with pixels in mind, not with performance. Currently when a draw call is scheduled it is flushed and waited until it is finished drawing, before other draw calls can be scheduled. We expected the performance to be worse that it actually is, but we expect huge performance gains in the future. - Any advanced drawing (that is used by the image editor, compositor or 3d viewport) isn't implemented and might crash when used. - Using multiple windows or resizing of window isn't supported and will stall the system. Pull Request: https://projects.blender.org/blender/blender/pulls/106224
2023-05-11 13:01:56 +02:00
return;
}
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
VkCommandBufferAllocateInfo alloc_info = {};
alloc_info.sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO;
alloc_info.commandPool = device.vk_command_pool_get();
alloc_info.level = VK_COMMAND_BUFFER_LEVEL_PRIMARY;
alloc_info.commandBufferCount = 1;
vkAllocateCommandBuffers(vk_device_, &alloc_info, &vk_command_buffer_);
submission_id_.reset();
state.stage = Stage::Initial;
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
if (vk_fence_ == VK_NULL_HANDLE) {
VK_ALLOCATION_CALLBACKS;
VkFenceCreateInfo fenceInfo{};
fenceInfo.sType = VK_STRUCTURE_TYPE_FENCE_CREATE_INFO;
vkCreateFence(vk_device_, &fenceInfo, vk_allocation_callbacks, &vk_fence_);
}
else {
vkResetFences(vk_device_, 1, &vk_fence_);
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
}
void VKCommandBuffer::begin_recording()
{
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
ensure_no_active_framebuffer();
if (is_in_stage(Stage::Submitted)) {
vkWaitForFences(vk_device_, 1, &vk_fence_, VK_TRUE, FenceTimeout);
vkResetFences(vk_device_, 1, &vk_fence_);
stage_transfer(Stage::Submitted, Stage::Executed);
}
if (is_in_stage(Stage::Executed)) {
vkResetCommandBuffer(vk_command_buffer_, 0);
stage_transfer(Stage::Executed, Stage::Initial);
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
VkCommandBufferBeginInfo begin_info = {};
begin_info.sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO;
vkBeginCommandBuffer(vk_command_buffer_, &begin_info);
stage_transfer(Stage::Initial, Stage::Recording);
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
}
void VKCommandBuffer::end_recording()
{
ensure_no_active_framebuffer();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
vkEndCommandBuffer(vk_command_buffer_);
stage_transfer(Stage::Recording, Stage::BetweenRecordingAndSubmitting);
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
}
void VKCommandBuffer::bind(const VKPipeline &pipeline, VkPipelineBindPoint bind_point)
{
vkCmdBindPipeline(vk_command_buffer_, bind_point, pipeline.vk_handle());
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
void VKCommandBuffer::bind(const VKDescriptorSet &descriptor_set,
const VkPipelineLayout vk_pipeline_layout,
VkPipelineBindPoint bind_point)
{
VkDescriptorSet vk_descriptor_set = descriptor_set.vk_handle();
vkCmdBindDescriptorSets(
vk_command_buffer_, bind_point, vk_pipeline_layout, 0, 1, &vk_descriptor_set, 0, 0);
}
void VKCommandBuffer::bind(const uint32_t binding,
const VKVertexBuffer &vertex_buffer,
const VkDeviceSize offset)
{
bind(binding, vertex_buffer.vk_handle(), offset);
}
Vulkan: Initial Graphics Pipeline Initial graphic pipeline targeting. The goal of this PR is to have an initial graphics pipeline with missing features. It should help identifying areas that requires engineering. Current state is that developers of the GPU module can help with the many smaller pieces that needs to be engineered in order to get it working. It is not intended for users or developers from other modules, but your welcome to learn and give feedback on the code and engineering part. We do expect that large parts of the code still needs to be re-engineered into a more future-proof implementation. **Some highlights**: - In Vulkan the state is kept in the pipeline. Therefore the state is tracked per pipeline. In the near future this could be used as a cache. More research is needed against the default pipeline cache that vulkan already provides. - This PR is based on the work that Kazashi Yoshioka already did. And include work from him in the next areas - Vertex attributes - Vertex data conversions - Pipeline state - Immediate support working. - This PR modifies the VKCommandBuffer to keep track of the framebuffer and its binding state(render pass). Some Vulkan commands require no render pass to be active, other require a render pass. As the order of our commands on API level can not be separated this PR introduces a state engine to keep track of the current state and desired state. This is a temporary solution, the final solution will be proposed when we have a pixel on the screen. At that time I expect that we can design a command encoder that supports all the cases we need. **Notices**: - This branch works on NVIDIA GPUs and has been validated on a Linux system. AMD is known not to work (stalls) and Intel GPUs have not been tested at all. Windows might work but hasn't been validated yet. - The graphics pipeline is implemented with pixels in mind, not with performance. Currently when a draw call is scheduled it is flushed and waited until it is finished drawing, before other draw calls can be scheduled. We expected the performance to be worse that it actually is, but we expect huge performance gains in the future. - Any advanced drawing (that is used by the image editor, compositor or 3d viewport) isn't implemented and might crash when used. - Using multiple windows or resizing of window isn't supported and will stall the system. Pull Request: https://projects.blender.org/blender/blender/pulls/106224
2023-05-11 13:01:56 +02:00
void VKCommandBuffer::bind(const uint32_t binding, const VKBufferWithOffset &vertex_buffer)
{
bind(binding, vertex_buffer.buffer.vk_handle(), vertex_buffer.offset);
}
void VKCommandBuffer::bind(const uint32_t binding,
const VkBuffer &vk_vertex_buffer,
const VkDeviceSize offset)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdBindVertexBuffers(vk_command_buffer_, binding, 1, &vk_vertex_buffer, &offset);
}
void VKCommandBuffer::bind(const VKBufferWithOffset &index_buffer, VkIndexType index_type)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdBindIndexBuffer(
vk_command_buffer_, index_buffer.buffer.vk_handle(), index_buffer.offset, index_type);
}
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
void VKCommandBuffer::begin_render_pass(VKFrameBuffer &framebuffer)
{
validate_framebuffer_not_exists();
state.framebuffer_ = &framebuffer;
}
void VKCommandBuffer::end_render_pass(const VKFrameBuffer &framebuffer)
{
UNUSED_VARS_NDEBUG(framebuffer);
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
BLI_assert(state.framebuffer_ == nullptr || state.framebuffer_ == &framebuffer);
ensure_no_active_framebuffer();
state.framebuffer_ = nullptr;
}
void VKCommandBuffer::push_constants(const VKPushConstants &push_constants,
const VkPipelineLayout vk_pipeline_layout,
const VkShaderStageFlags vk_shader_stages)
{
BLI_assert(push_constants.layout_get().storage_type_get() ==
VKPushConstants::StorageType::PUSH_CONSTANTS);
vkCmdPushConstants(vk_command_buffer_,
vk_pipeline_layout,
vk_shader_stages,
push_constants.offset(),
push_constants.layout_get().size_in_bytes(),
push_constants.data());
}
void VKCommandBuffer::fill(VKBuffer &buffer, uint32_t clear_data)
{
ensure_no_active_framebuffer();
vkCmdFillBuffer(vk_command_buffer_, buffer.vk_handle(), 0, buffer.size_in_bytes(), clear_data);
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
void VKCommandBuffer::copy(VKBuffer &dst_buffer,
VKTexture &src_texture,
Span<VkBufferImageCopy> regions)
{
ensure_no_active_framebuffer();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
vkCmdCopyImageToBuffer(vk_command_buffer_,
src_texture.vk_image_handle(),
src_texture.current_layout_get(),
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
dst_buffer.vk_handle(),
regions.size(),
regions.data());
}
Vulkan: Texture Data Conversions This PR adds basic support for texture update, read back and clearing for Vulkan. In Vulkan we need to convert each data type ourselves as vulkan buffers are untyped. Therefore this change mostly is about data conversions. Considerations: - Use a compute shader to do the conversions: - Leads to performance regression as compute pipeline can stall graphics pipeline - Lead to additional memory usage as two staging buffers are needed one to hold the CPU data, and one to hold the converted data. - Do inline conversion when sending the data to Vulkan using `eGPUDataFormat` - Additional CPU cycles required and not easy to optimize as it the implementation requires many branches. - Do inline conversion when sending the data to Vulkan (optimized for CPU) For this solution it was chosen to implement the 3rd option as it is fast and doesn't require additional memory what the other options do. **Use Imath/half.h** This patch uses `Imath/half.h` (dependency of OpenEXR) similar to alembic. But this makes vulkan dependent of the availability of OpenEXR. For now this isn't checked, but when we are closer to a working Vulkan backend we have to make a decision how to cope with this dependency. **Missing Features** *Framebuffer textures* This doesn't include all possible data transformations. Some of those transformation can only be tested after the VKFramebuffer has been implemented. Some texture types are only available when created for a framebuffer. These include the depth and stencil variations. *Component format* Is more relevant when implementing VKVertexBuffer. *SRGB textures* SRGB encoded textures aren't natively supported on all platforms, in all usages and might require workarounds. This should be done in a separate PR in a later stage when we are required to use SRGB textures. **Test cases** The added test cases gives an overview of the missing bits and pieces of the patch. When the implementation/direction is accepted more test cases can be enabled/implemented. Some of these test cases will skip depending on the actual support of platform the tests are running on. For example OpenGL/NVidia will skip the next test as it doesn't support the texture format on OpenGL, although it does support it on Vulkan. ``` [ RUN ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ SKIPPED ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ RUN ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ OK ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI ``` Pull Request: https://projects.blender.org/blender/blender/pulls/105762
2023-03-24 08:09:19 +01:00
void VKCommandBuffer::copy(VKTexture &dst_texture,
VKBuffer &src_buffer,
Span<VkBufferImageCopy> regions)
{
ensure_no_active_framebuffer();
Vulkan: Texture Data Conversions This PR adds basic support for texture update, read back and clearing for Vulkan. In Vulkan we need to convert each data type ourselves as vulkan buffers are untyped. Therefore this change mostly is about data conversions. Considerations: - Use a compute shader to do the conversions: - Leads to performance regression as compute pipeline can stall graphics pipeline - Lead to additional memory usage as two staging buffers are needed one to hold the CPU data, and one to hold the converted data. - Do inline conversion when sending the data to Vulkan using `eGPUDataFormat` - Additional CPU cycles required and not easy to optimize as it the implementation requires many branches. - Do inline conversion when sending the data to Vulkan (optimized for CPU) For this solution it was chosen to implement the 3rd option as it is fast and doesn't require additional memory what the other options do. **Use Imath/half.h** This patch uses `Imath/half.h` (dependency of OpenEXR) similar to alembic. But this makes vulkan dependent of the availability of OpenEXR. For now this isn't checked, but when we are closer to a working Vulkan backend we have to make a decision how to cope with this dependency. **Missing Features** *Framebuffer textures* This doesn't include all possible data transformations. Some of those transformation can only be tested after the VKFramebuffer has been implemented. Some texture types are only available when created for a framebuffer. These include the depth and stencil variations. *Component format* Is more relevant when implementing VKVertexBuffer. *SRGB textures* SRGB encoded textures aren't natively supported on all platforms, in all usages and might require workarounds. This should be done in a separate PR in a later stage when we are required to use SRGB textures. **Test cases** The added test cases gives an overview of the missing bits and pieces of the patch. When the implementation/direction is accepted more test cases can be enabled/implemented. Some of these test cases will skip depending on the actual support of platform the tests are running on. For example OpenGL/NVidia will skip the next test as it doesn't support the texture format on OpenGL, although it does support it on Vulkan. ``` [ RUN ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ SKIPPED ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ RUN ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ OK ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI ``` Pull Request: https://projects.blender.org/blender/blender/pulls/105762
2023-03-24 08:09:19 +01:00
vkCmdCopyBufferToImage(vk_command_buffer_,
src_buffer.vk_handle(),
dst_texture.vk_image_handle(),
dst_texture.current_layout_get(),
Vulkan: Texture Data Conversions This PR adds basic support for texture update, read back and clearing for Vulkan. In Vulkan we need to convert each data type ourselves as vulkan buffers are untyped. Therefore this change mostly is about data conversions. Considerations: - Use a compute shader to do the conversions: - Leads to performance regression as compute pipeline can stall graphics pipeline - Lead to additional memory usage as two staging buffers are needed one to hold the CPU data, and one to hold the converted data. - Do inline conversion when sending the data to Vulkan using `eGPUDataFormat` - Additional CPU cycles required and not easy to optimize as it the implementation requires many branches. - Do inline conversion when sending the data to Vulkan (optimized for CPU) For this solution it was chosen to implement the 3rd option as it is fast and doesn't require additional memory what the other options do. **Use Imath/half.h** This patch uses `Imath/half.h` (dependency of OpenEXR) similar to alembic. But this makes vulkan dependent of the availability of OpenEXR. For now this isn't checked, but when we are closer to a working Vulkan backend we have to make a decision how to cope with this dependency. **Missing Features** *Framebuffer textures* This doesn't include all possible data transformations. Some of those transformation can only be tested after the VKFramebuffer has been implemented. Some texture types are only available when created for a framebuffer. These include the depth and stencil variations. *Component format* Is more relevant when implementing VKVertexBuffer. *SRGB textures* SRGB encoded textures aren't natively supported on all platforms, in all usages and might require workarounds. This should be done in a separate PR in a later stage when we are required to use SRGB textures. **Test cases** The added test cases gives an overview of the missing bits and pieces of the patch. When the implementation/direction is accepted more test cases can be enabled/implemented. Some of these test cases will skip depending on the actual support of platform the tests are running on. For example OpenGL/NVidia will skip the next test as it doesn't support the texture format on OpenGL, although it does support it on Vulkan. ``` [ RUN ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ SKIPPED ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ RUN ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ OK ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI ``` Pull Request: https://projects.blender.org/blender/blender/pulls/105762
2023-03-24 08:09:19 +01:00
regions.size(),
regions.data());
}
void VKCommandBuffer::copy(VKTexture &dst_texture,
VKTexture &src_texture,
Span<VkImageCopy> regions)
{
ensure_no_active_framebuffer();
vkCmdCopyImage(vk_command_buffer_,
src_texture.vk_image_handle(),
src_texture.current_layout_get(),
dst_texture.vk_image_handle(),
dst_texture.current_layout_get(),
regions.size(),
regions.data());
}
void VKCommandBuffer::copy(VKBuffer &dst_buffer, VkBuffer src_buffer, Span<VkBufferCopy> regions)
{
ensure_no_active_framebuffer();
vkCmdCopyBuffer(
vk_command_buffer_, src_buffer, dst_buffer.vk_handle(), regions.size(), regions.data());
}
void VKCommandBuffer::blit(VKTexture &dst_texture,
VKTexture &src_texture,
Span<VkImageBlit> regions)
{
blit(dst_texture,
dst_texture.current_layout_get(),
src_texture,
src_texture.current_layout_get(),
regions);
}
void VKCommandBuffer::blit(VKTexture &dst_texture,
VkImageLayout dst_layout,
VKTexture &src_texture,
VkImageLayout src_layout,
Span<VkImageBlit> regions)
{
ensure_no_active_framebuffer();
vkCmdBlitImage(vk_command_buffer_,
src_texture.vk_image_handle(),
src_layout,
dst_texture.vk_image_handle(),
dst_layout,
regions.size(),
regions.data(),
VK_FILTER_NEAREST);
}
Vulkan: Texture Data Conversions This PR adds basic support for texture update, read back and clearing for Vulkan. In Vulkan we need to convert each data type ourselves as vulkan buffers are untyped. Therefore this change mostly is about data conversions. Considerations: - Use a compute shader to do the conversions: - Leads to performance regression as compute pipeline can stall graphics pipeline - Lead to additional memory usage as two staging buffers are needed one to hold the CPU data, and one to hold the converted data. - Do inline conversion when sending the data to Vulkan using `eGPUDataFormat` - Additional CPU cycles required and not easy to optimize as it the implementation requires many branches. - Do inline conversion when sending the data to Vulkan (optimized for CPU) For this solution it was chosen to implement the 3rd option as it is fast and doesn't require additional memory what the other options do. **Use Imath/half.h** This patch uses `Imath/half.h` (dependency of OpenEXR) similar to alembic. But this makes vulkan dependent of the availability of OpenEXR. For now this isn't checked, but when we are closer to a working Vulkan backend we have to make a decision how to cope with this dependency. **Missing Features** *Framebuffer textures* This doesn't include all possible data transformations. Some of those transformation can only be tested after the VKFramebuffer has been implemented. Some texture types are only available when created for a framebuffer. These include the depth and stencil variations. *Component format* Is more relevant when implementing VKVertexBuffer. *SRGB textures* SRGB encoded textures aren't natively supported on all platforms, in all usages and might require workarounds. This should be done in a separate PR in a later stage when we are required to use SRGB textures. **Test cases** The added test cases gives an overview of the missing bits and pieces of the patch. When the implementation/direction is accepted more test cases can be enabled/implemented. Some of these test cases will skip depending on the actual support of platform the tests are running on. For example OpenGL/NVidia will skip the next test as it doesn't support the texture format on OpenGL, although it does support it on Vulkan. ``` [ RUN ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ SKIPPED ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ RUN ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ OK ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI ``` Pull Request: https://projects.blender.org/blender/blender/pulls/105762
2023-03-24 08:09:19 +01:00
void VKCommandBuffer::clear(VkImage vk_image,
VkImageLayout vk_image_layout,
const VkClearColorValue &vk_clear_color,
Span<VkImageSubresourceRange> ranges)
{
ensure_no_active_framebuffer();
Vulkan: Texture Data Conversions This PR adds basic support for texture update, read back and clearing for Vulkan. In Vulkan we need to convert each data type ourselves as vulkan buffers are untyped. Therefore this change mostly is about data conversions. Considerations: - Use a compute shader to do the conversions: - Leads to performance regression as compute pipeline can stall graphics pipeline - Lead to additional memory usage as two staging buffers are needed one to hold the CPU data, and one to hold the converted data. - Do inline conversion when sending the data to Vulkan using `eGPUDataFormat` - Additional CPU cycles required and not easy to optimize as it the implementation requires many branches. - Do inline conversion when sending the data to Vulkan (optimized for CPU) For this solution it was chosen to implement the 3rd option as it is fast and doesn't require additional memory what the other options do. **Use Imath/half.h** This patch uses `Imath/half.h` (dependency of OpenEXR) similar to alembic. But this makes vulkan dependent of the availability of OpenEXR. For now this isn't checked, but when we are closer to a working Vulkan backend we have to make a decision how to cope with this dependency. **Missing Features** *Framebuffer textures* This doesn't include all possible data transformations. Some of those transformation can only be tested after the VKFramebuffer has been implemented. Some texture types are only available when created for a framebuffer. These include the depth and stencil variations. *Component format* Is more relevant when implementing VKVertexBuffer. *SRGB textures* SRGB encoded textures aren't natively supported on all platforms, in all usages and might require workarounds. This should be done in a separate PR in a later stage when we are required to use SRGB textures. **Test cases** The added test cases gives an overview of the missing bits and pieces of the patch. When the implementation/direction is accepted more test cases can be enabled/implemented. Some of these test cases will skip depending on the actual support of platform the tests are running on. For example OpenGL/NVidia will skip the next test as it doesn't support the texture format on OpenGL, although it does support it on Vulkan. ``` [ RUN ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ SKIPPED ] GPUOpenGLTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ RUN ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI [ OK ] GPUVulkanTest.texture_roundtrip__GPU_DATA_2_10_10_10_REV__GPU_RGB10_A2UI ``` Pull Request: https://projects.blender.org/blender/blender/pulls/105762
2023-03-24 08:09:19 +01:00
vkCmdClearColorImage(vk_command_buffer_,
vk_image,
vk_image_layout,
&vk_clear_color,
ranges.size(),
ranges.data());
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
void VKCommandBuffer::clear(VkImage vk_image,
VkImageLayout vk_image_layout,
const VkClearDepthStencilValue &vk_clear_value,
Span<VkImageSubresourceRange> ranges)
{
ensure_no_active_framebuffer();
vkCmdClearDepthStencilImage(vk_command_buffer_,
vk_image,
vk_image_layout,
&vk_clear_value,
ranges.size(),
ranges.data());
}
void VKCommandBuffer::clear(Span<VkClearAttachment> attachments, Span<VkClearRect> areas)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdClearAttachments(
vk_command_buffer_, attachments.size(), attachments.data(), areas.size(), areas.data());
}
void VKCommandBuffer::draw(int v_first, int v_count, int i_first, int i_count)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdDraw(vk_command_buffer_, v_count, i_count, v_first, i_first);
state.draw_counts++;
}
void VKCommandBuffer::draw_indexed(
int index_count, int instance_count, int first_index, int vertex_offset, int first_instance)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdDrawIndexed(
vk_command_buffer_, index_count, instance_count, first_index, vertex_offset, first_instance);
state.draw_counts++;
}
void VKCommandBuffer::draw_indirect(const VKStorageBuffer &buffer,
VkDeviceSize offset,
uint32_t draw_count,
uint32_t stride)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdDrawIndirect(vk_command_buffer_, buffer.vk_handle(), offset, draw_count, stride);
state.draw_counts++;
}
void VKCommandBuffer::draw_indexed_indirect(const VKStorageBuffer &buffer,
VkDeviceSize offset,
uint32_t draw_count,
uint32_t stride)
{
validate_framebuffer_exists();
ensure_active_framebuffer();
vkCmdDrawIndexedIndirect(vk_command_buffer_, buffer.vk_handle(), offset, draw_count, stride);
state.draw_counts++;
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
void VKCommandBuffer::pipeline_barrier(VkPipelineStageFlags source_stages,
VkPipelineStageFlags destination_stages)
{
if (state.framebuffer_) {
ensure_active_framebuffer();
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
vkCmdPipelineBarrier(vk_command_buffer_,
source_stages,
destination_stages,
0,
0,
nullptr,
0,
nullptr,
0,
nullptr);
}
void VKCommandBuffer::pipeline_barrier(Span<VkImageMemoryBarrier> image_memory_barriers)
{
ensure_no_active_framebuffer();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
vkCmdPipelineBarrier(vk_command_buffer_,
VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT,
VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT,
VK_DEPENDENCY_BY_REGION_BIT,
0,
nullptr,
0,
nullptr,
image_memory_barriers.size(),
image_memory_barriers.data());
}
void VKCommandBuffer::dispatch(int groups_x_len, int groups_y_len, int groups_z_len)
{
ensure_no_active_framebuffer();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
vkCmdDispatch(vk_command_buffer_, groups_x_len, groups_y_len, groups_z_len);
}
void VKCommandBuffer::dispatch(VKStorageBuffer &command_buffer)
{
ensure_no_active_framebuffer();
vkCmdDispatchIndirect(vk_command_buffer_, command_buffer.vk_handle(), 0);
}
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
void VKCommandBuffer::submit()
{
ensure_no_active_framebuffer();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
end_recording();
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
submit_commands();
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
begin_recording();
}
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
void VKCommandBuffer::submit_commands()
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
{
VkSubmitInfo submit_info = {};
submit_info.sType = VK_STRUCTURE_TYPE_SUBMIT_INFO;
submit_info.commandBufferCount = 1;
submit_info.pCommandBuffers = &vk_command_buffer_;
vkQueueSubmit(vk_queue_, 1, &submit_info, vk_fence_);
submission_id_.next();
stage_transfer(Stage::BetweenRecordingAndSubmitting, Stage::Submitted);
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
}
/* -------------------------------------------------------------------- */
2023-04-26 12:01:23 +10:00
/** \name FrameBuffer/RenderPass state tracking
* \{ */
void VKCommandBuffer::validate_framebuffer_not_exists()
{
BLI_assert_msg(state.framebuffer_ == nullptr && state.framebuffer_active_ == false,
"State error: expected no framebuffer being tracked.");
}
void VKCommandBuffer::validate_framebuffer_exists()
{
BLI_assert_msg(state.framebuffer_, "State error: expected framebuffer being tracked.");
}
void VKCommandBuffer::ensure_no_active_framebuffer()
{
state.checks_++;
if (state.framebuffer_ && state.framebuffer_active_) {
vkCmdEndRenderPass(vk_command_buffer_);
state.framebuffer_active_ = false;
state.switches_++;
}
}
void VKCommandBuffer::ensure_active_framebuffer()
{
BLI_assert(state.framebuffer_);
state.checks_++;
if (!state.framebuffer_active_) {
VkRenderPassBeginInfo render_pass_begin_info = {};
render_pass_begin_info.sType = VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO;
Vulkan: Rewrite GHOST_ContextVK This is a rewrite of GHOST_ContextVK to align with Metal backend as described in #111389 - solution 3 with the adaptation that GHOST is still responsible for presenting the swap chain image and a post callback is still needed in case the swapchain is recreated. This PR also includes some smaller improvements in stability. Technical documentation: https://developer.blender.org/docs/eevee_and_viewport/gpu/vulkan/swap_chain/ * Renderpasses and framebuffers are not owned anymore by GHOST_ContextVK * VKFramebuffer doesn't contain a swap chain image. * Swapchain images can only be used as a blit destination or present source. Not as an attachment. * GHOST_ContextVK::swapBuffers would call a callback with the image the GPU module needs to blit the results to. * Clearing of depth/stencil attachments when no depth write state is set. * Enable VK_KHR_maintenance4 to relax the stage interface mapping. * Removes most vulkan validation warnings/errors. * Detection of frame buffer changes that needs to be applied before performing a command requiring render pass (draw/clear attachment) **Benefits** * Late retrieval of a swap chain image results in better overall performance as Blender doesn't need to wait until the image is presented on the screen. * Easier API and clearer state (transitions) * More control over Image layouts and command buffer states. (Better alignment with Vulkan API) Pull Request: https://projects.blender.org/blender/blender/pulls/111473
2023-08-29 15:05:08 +02:00
state.framebuffer_->vk_render_pass_ensure();
render_pass_begin_info.renderPass = state.framebuffer_->vk_render_pass_get();
render_pass_begin_info.framebuffer = state.framebuffer_->vk_framebuffer_get();
render_pass_begin_info.renderArea = state.framebuffer_->vk_render_areas_get()[0];
/* We don't use clear ops, but vulkan wants to have at least one. */
VkClearValue clear_value = {};
render_pass_begin_info.clearValueCount = 1;
render_pass_begin_info.pClearValues = &clear_value;
vkCmdBeginRenderPass(vk_command_buffer_, &render_pass_begin_info, VK_SUBPASS_CONTENTS_INLINE);
state.framebuffer_active_ = true;
state.switches_++;
}
}
/** \} */
Vulkan: Initial Compute Shaders support This patch adds initial support for compute shaders to the vulkan backend. As the development is oriented to the test- cases we have the implementation is limited to what is used there. It has been validated that with this patch that the following test cases are running as expected - `GPUVulkanTest.gpu_shader_compute_vbo` - `GPUVulkanTest.gpu_shader_compute_ibo` - `GPUVulkanTest.gpu_shader_compute_ssbo` - `GPUVulkanTest.gpu_storage_buffer_create_update_read` - `GPUVulkanTest.gpu_shader_compute_2d` This patch includes: - Allocating VkBuffer on device. - Uploading data from CPU to VkBuffer. - Binding VkBuffer as SSBO to a compute shader. - Execute compute shader and altering VkBuffer. - Download the VkBuffer to CPU ram. - Validate that it worked. - Use device only vertex buffer as SSBO - Use device only index buffer as SSBO - Use device only image buffers GHOST API has been changed as the original design was created before we even had support for compute shaders in blender. The function `GHOST_getVulkanBackbuffer` has been separated to retrieve the command buffer without a backbuffer (`GHOST_getVulkanCommandBuffer`). In order to do correct command buffer processing we needed access to the queue owned by GHOST. This is returned as part of the `GHOST_getVulkanHandles` function. Open topics (not considered part of this patch) - Memory barriers & command buffer encoding - Indirect compute dispatching - Rest of the test cases - Data conversions when requested data format is different than on device. - GPUVulkanTest.gpu_shader_compute_1d is supported on AMD devices. NVIDIA doesn't seem to support 1d textures. Pull-request: #104518
2023-02-21 15:03:12 +01:00
} // namespace blender::gpu