Crafter.Graphics/interfaces/Crafter.Graphics-ShaderBindingTableVulkan.cppm

56 lines
2.2 KiB
Text
Raw Permalink Normal View History

2026-01-31 21:08:42 +01:00
/*
Crafter®.Graphics
Copyright (C) 2026 Catcrafts®
catcrafts.net
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License version 3.0 as published by the Free Software Foundation;
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
module;
2026-05-18 02:07:48 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-03-02 23:53:13 +01:00
#include "vulkan/vulkan.h"
2026-05-18 02:07:48 +02:00
#endif // !CRAFTER_GRAPHICS_WINDOW_DOM
2026-01-31 21:08:42 +01:00
export module Crafter.Graphics:ShaderBindingTableVulkan;
2026-05-18 02:07:48 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-01-31 21:08:42 +01:00
import std;
2026-03-09 20:10:19 +01:00
import :Device;
2026-01-31 21:08:42 +01:00
import :VulkanBuffer;
2026-02-22 00:46:38 +01:00
import :ShaderVulkan;
2026-01-31 21:08:42 +01:00
import :Types;
export namespace Crafter {
class ShaderBindingTableVulkan {
2026-02-22 00:46:38 +01:00
public:
std::vector<VkPipelineShaderStageCreateInfo> shaderStages;
fix(vulkan-rt): configurable recursion depth + per-shader TLAS push for compute (#21) Two gaps in the Vulkan RT path that fault the device on the NVIDIA proprietary driver with a non-trivial pipeline (simple VulkanTriangle never hit them): 1. maxPipelineRayRecursionDepth was hardcoded to 1, so any closest-hit shader that traces a secondary ray (shadow ray — a very common pattern) recursed past the pipeline limit (UB → device fault). PipelineRTVulkan::Init now takes a maxRecursionDepth parameter (default 1, clamped to the device's maxRayRecursionDepth). 2. The NVIDIA descriptor-heap AS-read workaround rewrites every shader that reads an accelerationStructureEXT from the heap — including compute shaders — to read the TLAS device address from a push constant, but only RTPass pushed that address. A compute shader that ray-queries the TLAS (rayQueryEXT) therefore ran against an unwritten push slot → garbage AS handle → VK_ERROR_DEVICE_LOST. WorkaroundNvidiaAS::Patch now returns a per-shader PatchResult {patched, tlasPushOffset} instead of writing the clobber-prone global Device::workaroundTlasPushOffset (removed). VulkanShader stores it; ShaderBindingTableVulkan/PipelineRTVulkan carry it for RTPass, and ComputeShader tracks its own offset and pushes the caller-supplied TLAS address in Dispatch (new defaulted tlasAddress parameter), mirroring RTPass::Record. The PushConstantRewrite regression test now asserts Patch's returned patched/offset and adds two ray-querying compute-shader cases, proving the rewrite is stage-agnostic and the per-shader offset is correct. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-03 18:35:39 +00:00
// NVIDIA descriptor-heap AS-read workaround (issue #15 / #7): true when
// any stage in this table reads an acceleration structure and was
// rewritten to fetch the TLAS address from a push constant, with the
// byte offset that stage expects it at. PipelineRTVulkan copies these so
// RTPass can push the address without consulting a clobber-prone global.
// Both inert (false/0) on every other driver.
bool workaroundNeedsTlas = false;
std::uint32_t workaroundTlasPushOffset = 0;
2026-04-10 20:53:17 +02:00
void Init(const std::span<const VulkanShader> shaders) {
2026-02-22 00:46:38 +01:00
shaderStages.reserve(shaders.size());
for(const VulkanShader& shader: shaders) {
2026-04-05 22:53:59 +02:00
shaderStages.emplace_back(VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO, nullptr, 0, shader.stage, shader.shader, shader.entrypoint.c_str(), shader.specilizationInfo);
fix(vulkan-rt): configurable recursion depth + per-shader TLAS push for compute (#21) Two gaps in the Vulkan RT path that fault the device on the NVIDIA proprietary driver with a non-trivial pipeline (simple VulkanTriangle never hit them): 1. maxPipelineRayRecursionDepth was hardcoded to 1, so any closest-hit shader that traces a secondary ray (shadow ray — a very common pattern) recursed past the pipeline limit (UB → device fault). PipelineRTVulkan::Init now takes a maxRecursionDepth parameter (default 1, clamped to the device's maxRayRecursionDepth). 2. The NVIDIA descriptor-heap AS-read workaround rewrites every shader that reads an accelerationStructureEXT from the heap — including compute shaders — to read the TLAS device address from a push constant, but only RTPass pushed that address. A compute shader that ray-queries the TLAS (rayQueryEXT) therefore ran against an unwritten push slot → garbage AS handle → VK_ERROR_DEVICE_LOST. WorkaroundNvidiaAS::Patch now returns a per-shader PatchResult {patched, tlasPushOffset} instead of writing the clobber-prone global Device::workaroundTlasPushOffset (removed). VulkanShader stores it; ShaderBindingTableVulkan/PipelineRTVulkan carry it for RTPass, and ComputeShader tracks its own offset and pushes the caller-supplied TLAS address in Dispatch (new defaulted tlasAddress parameter), mirroring RTPass::Record. The PushConstantRewrite regression test now asserts Patch's returned patched/offset and adds two ray-querying compute-shader cases, proving the rewrite is stage-agnostic and the per-shader offset is correct. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-03 18:35:39 +00:00
if (shader.patchedAS) {
workaroundNeedsTlas = true;
workaroundTlasPushOffset = shader.tlasPushOffset;
}
2026-02-22 00:46:38 +01:00
}
}
};
2026-05-18 02:07:48 +02:00
}
#endif // !CRAFTER_GRAPHICS_WINDOW_DOM