shifathosn
/
AnyPS5
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #1408 from hashaam1217/fix/pin-wide-compute-subgroups 

fix(agc): require the widest subgroup for compute programs wider than the device
This commit is contained in:
Adrià 2026-10-08 15:25:25 +02:00 committed by GitHub
parent 02e741060e f91e329140
commit b3fcd6d6c0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
core/libs/prx/libSceAgcDriver/Execution/src/VulkanDevice.cpp
View File

@ -220,6 +220,7 @@ struct VulkanDevice::State {
// VK_KHR_timeline_semaphore enabled: the recorder's unlocked waits are available.
bool timelineSemaphores = false;
bool computeWave32 = false;
std::uint32_t maxComputeSubgroupSize = 0;
// Indirect draw features enabled (see Graphics::Context).
bool drawIndirectFirstInstance = false;
bool multiDrawIndirect = false;
@ -1066,6 +1067,7 @@ VulkanDevice::VulkanDevice(const PresentationWindow* window) : state(std::make_u
state->InstanceFunction<PFN_vkGetPhysicalDeviceProperties2>("vkGetPhysicalDeviceProperties2")(selected, &sizeProperties);
state->computeWave32 = subgroupSizeFeatures.subgroupSizeControl == VK_TRUE && subgroupSize.minSubgroupSize <= 32u && subgroupSize.maxSubgroupSize >= 32u &&
(subgroupSize.requiredSubgroupSizeStages & VK_SHADER_STAGE_COMPUTE_BIT) != 0;
state->maxComputeSubgroupSize = subgroupSize.maxSubgroupSize;
}
if (state->computeWave32) {
subgroupSizeFeatures = {VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_SUBGROUP_SIZE_CONTROL_FEATURES_EXT};
@ -3528,8 +3530,8 @@ VulkanDevice::IndirectOutcome VulkanDevice::dispatch(const ShaderRecompiler::Rec
pipelineInfo.stage.module = objects->module;
pipelineInfo.stage.pName = "main";
VkPipelineShaderStageRequiredSubgroupSizeCreateInfoEXT requiredSubgroup{VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_REQUIRED_SUBGROUP_SIZE_CREATE_INFO_EXT};
requiredSubgroup.requiredSubgroupSize = shader.hostSubgroupSize;
if (state->computeWave32 && shader.hostSubgroupSize == 32u) pipelineInfo.stage.pNext = &requiredSubgroup;
requiredSubgroup.requiredSubgroupSize = std::min(shader.hostSubgroupSize, state->maxComputeSubgroupSize);
if (state->computeWave32 && (shader.hostSubgroupSize == 32u || shader.hostSubgroupSize > state->maxComputeSubgroupSize)) pipelineInfo.stage.pNext = &requiredSubgroup;
pipelineInfo.layout = objects->layout;
if (profile && shader.spirv.size() > 100000) std::fprintf(stderr, "[dispatch] creating a pipeline for %zu SPIR-V words (program 0x%llx)\n", shader.spirv.size(), static_cast<unsigned long long>(programAddress));
check(state->DeviceFunction<PFN_vkCreateComputePipelines>("vkCreateComputePipelines")(state->device, context.pipelineCache, 1, &pipelineInfo, nullptr, &objects->pipeline), "vkCreateComputePipelines");