swung0x48/MobileGL
1
0
Fork 0
mirror of https://github.com/MobileGL-Dev/MobileGL synced 2026-11-07 18:13:23 +09:00
Code Issues Packages Projects Releases Wiki Activity

[Fix, Test] (DirectVulkan, ShaderTranspiler, TraceReplay): repair iterationRP's missing reduction barrier

Browse Source 
Program 203 reuses prefixSumCache for a second subgroup reduction before every workgroup invocation has consumed the first result. Add a fingerprint-gated SPIR-V pass that inserts the missing Workgroup acquire-release barrier while preserving native subgroup operations.

Keep the repair opt-in behind MOBILEGL_ITERATIONRP_FIX_BARRIER, cover insertion, pass-through, and idempotence, and enable it together with the existing iterationRP subgroup repairs for the matching Linux and Android CI retraces.
This commit is contained in:
swung0x48
2026-08-19 23:42:17 -04:00
parent 5bd8ef01e5
commit c09045fe59
20 changed files with 583 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
MobileGL/MG_Backend/DirectVulkan/Renderer/ProgramFactory.cpp
+16 -1
View File
@@ -3190,10 +3190,25 @@ namespace MobileGL::MG_Backend::DirectVulkan {
}
// GL_KHR_shader_subgroup handling (SubgroupSupportPolicy.h). Native subgroup
// operations execute natively; two module repairs keep the GL contract intact
// operations execute natively; module repairs keep the GL contract intact
// around them. The opt-in emulation path replaces them only on devices with no
// subgroup support at all (MOBILEGL_MAGMA_EMULATE_SUBGROUP).
if (shaders[i] && shaders[i]->GetShaderStage() == ShaderStage::Compute) {
// Program 203 broadcasts the first reduction through
// prefixSumCache[0], then lets the second reduction overwrite that
// scratch without first rendezvousing all readers. Patch that exact
// fingerprint before either native or emulated subgroup lowering.
if (m_subgroupPolicy.fixIterationRPBarrier) {
Vector<Uint> patchedSpirv;
if (MG_Util::ShaderTranspiler::ShaderCompiler::FixIterationRPBarrierForVulkan(
moduleSpirvs[i], patchedSpirv, enableSpirvValidation)) {
moduleSpirvs[i] = std::move(patchedSpirv);
} else {
MGLOG_E("ProgramFactory: iterationRP barrier patch failed for program %u; "
"Program 203 keeps its shared-scratch race",
program.GetExternalIndex());
}
}
if (m_subgroupPolicy.emulateSubgroups) {
Vector<Uint> emulatedSpirv;
if (MG_Util::ShaderTranspiler::ShaderCompiler::EmulateSubgroupsForVulkan(