-
Notifications
You must be signed in to change notification settings - Fork 116
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Promotion from AMD internal branch for 2024.Q3.1 #2903
Merged
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[Continuations] Remove `CONTINUATIONS_USE_DUMMY_RET_ADDR` define [Continuations] Remove references to `_cont_SetupRayGen` [RT] Fix IR verifier error when reading GPURT SPIRVReader: Add LayoutMode::None, remove isParentPointer, m_typeMap, make recordTypeWithPad return void [LGC] Remove descriptor type argument from getDescPtr Fix broken spvasm tests Add support for SPV_EXT_fragment_shader_interlock compilerutils: Put TypeLowering into a namespace Remove RT lit test dependency on SPIR-V binary Fix and cleanup barrier translation Add an check of dual source blend in full pipeline [LGC] Fix deprecation warnings SPIRVReader: Unify OpType{Runtime}Array conversion Fix OpControlBarrier mapping Fix compiling warning of switch cases [LGC] Handle load and build buffer descriptor in PatchBufferOp SPIRVReader: refactor image support [Continuations] Pass padding and payload as argument [Continuations] Fix irreducible cycle during continuation transform Unify the namespace compilerutils with CompilerUtils Rework mesh shader export slot assignment [Continuations] Remove RegisterBuffer pass. [RT] Rayquery refactor to lgc [LGC] Fix cast error when forceRawView buffer load is used [Continuations] Remove `DISABLE_CPS_PAYLOADREG` define. Stop replacing atomicrmw fadd instructions with intrinsics [LGC] Missing index for LoadTfeOp structured load [Continations] Remove comment on register buffer Remove assertion and update tests to align with recent LLVM changes amdllpc: Fix amdllpc test failure PipelineRays_TestStaticCompile.pipe [Continuations] Move `handleGetFuncAddr` amdllpc: Flush SPIR-V output to aid in debugging Make readlane call mangling-agnostic [LGC] Fix deprecated StringRef::equals warning [LGC] Remove use of RepeatedPass [LGC] Include FormatVariadic.h [Continuations] Reformat code SPIRVReader: workaround for some invalid SPIR-V Fix traversal module pop back Extract LowerRayQuery from LGC and move to RT [Continuations] Dialectify `continuation.continue` and `continuation.waitContinue`. Update intrinsic names for upstream change Fix use-after-free in LowerRayQuery refactor Add debug printer for ModuleBunch Stop replacing atomicrmw fmin/fmax instructions with intrinsics Update tests for non-i8 gep of mul change [Continuations] Tidy up assertion in LowerRaytracingPipeline Prefer mdconst::extract over mdconst::dyn_extract Update includes and fwd decls Fix some Clang warnings Enable intrinsic instead of using named call Use range-based set() for SmallBitVector [LGC] Add dereferenceable info for descriptor load [Continuations] Ignore single unreachable in block [CompilerUtils] Handle phi in replaceAllPointerUses Use CreateIntrinsic for building intrinsic calls llvmraytracing: Add getShaderSubtypeForRtShaderStage [Continuations] Refactor metadata handling [RT] Support GpurtGetKnown(Un)SetRayFlagsOp, and add a lit test Fix build and test issue when bumping LLPC version to 73 Move the Dx9MulZero match from llpcSpirvLowerMath into PatchMulDx9Zero which is added after Scalarize Pass llvmraytracing: Add getFuncReturnPtrElementType [RT] Avoid unnecessary compilations for traversal module Fix RTIP-overriding for GPURT rayQuery functions Fix #include file ordering [LGC] Set new .uses_cps PAL metadata where applicable [CompilerUtils] Simplify pointee types metadata [Continuations] Use `lgc.cps.jump` in `LegacyCleanupContinuations` [CompilerUtils] Move promotePointerArguments etc to compilerutils Use function linkage setting from spirv [CompilerUtils] Fix cross-module-inliner tests, and move to subdir [LGC] Add option to force user data spilling Improve continuations heuristic [Continuations] Fix clang-format Remove SpirvLowerRayQuery from lower passes [Continuations] Cleanup continuations analyses, remove registercount + statesize metadata [Continuations] Prevent zero-size `lgc.cps.free` calls in `CleanupContinuations` [LGC] Only insert non null components in reassembleOutputExportCalls [Continuations] Use new `replaceVarArgs` helper from dialects [Continuations] Add constant jump 'inliner' pass [Continuations] Basic Continuations 'linter' pass [Continuations] Introduce lgc.cps.jump in `LowerRaytracingPipeline` Remove explicit LLVM benchmark targets from CMake [Continuations] Handle _AmdValue* intrinsics [LGC] Fix issues of FS without exports Update llvm-dialects submodule Pack mesh shader outputs in LDS space
Test summary for commit 5278a3fCTS tests (Failed: 11/148731)
Ubuntu navi3x, SrdcvkUbuntu navi2x, Srdcvk |
test this please |
Test summary for commit 5278a3fCTS tests (Failed: 11/148731)
Ubuntu navi3x, SrdcvkUbuntu navi2x, Srdcvk |
Test summary for commit 5278a3fCTS tests (Failed: 0/148731)
Ubuntu navi3x, SrdcvkUbuntu navi2x, Srdcvk |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
[Continuations] Remove
CONTINUATIONS_USE_DUMMY_RET_ADDR
define [Continuations] Remove references to_cont_SetupRayGen
[RT] Fix IR verifier error when reading GPURTSPIRVReader: Add LayoutMode::None, remove isParentPointer, m_typeMap, make recordTypeWithPad return void [LGC] Remove descriptor type argument from getDescPtr Fix broken spvasm tests
Add support for SPV_EXT_fragment_shader_interlock
compilerutils: Put TypeLowering into a namespace
Remove RT lit test dependency on SPIR-V binary
Fix and cleanup barrier translation
Add an check of dual source blend in full pipeline [LGC] Fix deprecation warnings
SPIRVReader: Unify OpType{Runtime}Array conversion Fix OpControlBarrier mapping
Fix compiling warning of switch cases
[LGC] Handle load and build buffer descriptor in PatchBufferOp SPIRVReader: refactor image support
[Continuations] Pass padding and payload as argument [Continuations] Fix irreducible cycle during continuation transform Unify the namespace compilerutils with CompilerUtils Rework mesh shader export slot assignment
[Continuations] Remove RegisterBuffer pass.
[RT] Rayquery refactor to lgc
[LGC] Fix cast error when forceRawView buffer load is used [Continuations] Remove
DISABLE_CPS_PAYLOADREG
define. Stop replacing atomicrmw fadd instructions with intrinsics [LGC] Missing index for LoadTfeOp structured load[Continations] Remove comment on register buffer
Remove assertion and update tests to align with recent LLVM changes amdllpc: Fix amdllpc test failure PipelineRays_TestStaticCompile.pipe [Continuations] Move
handleGetFuncAddr
amdllpc: Flush SPIR-V output to aid in debugging
Make readlane call mangling-agnostic
[LGC] Fix deprecated StringRef::equals warning
[LGC] Remove use of RepeatedPass
[LGC] Include FormatVariadic.h
[Continuations] Reformat code
SPIRVReader: workaround for some invalid SPIR-V
Fix traversal module pop back
Extract LowerRayQuery from LGC and move to RT
[Continuations] Dialectify
continuation.continue
andcontinuation.waitContinue
. Update intrinsic names for upstream changeFix use-after-free in LowerRayQuery refactor
Add debug printer for ModuleBunch
Stop replacing atomicrmw fmin/fmax instructions with intrinsics Update tests for non-i8 gep of mul change
[Continuations] Tidy up assertion in LowerRaytracingPipeline Prefer mdconst::extract over mdconst::dyn_extract
Update includes and fwd decls
Fix some Clang warnings
Enable intrinsic instead of using named call
Use range-based set() for SmallBitVector
[LGC] Add dereferenceable info for descriptor load [Continuations] Ignore single unreachable in block [CompilerUtils] Handle phi in replaceAllPointerUses Use CreateIntrinsic for building intrinsic calls
llvmraytracing: Add getShaderSubtypeForRtShaderStage [Continuations] Refactor metadata handling
[RT] Support GpurtGetKnown(Un)SetRayFlagsOp, and add a lit test Fix build and test issue when bumping LLPC version to 73 Move the Dx9MulZero match from llpcSpirvLowerMath into PatchMulDx9Zero which is added after Scalarize Pass llvmraytracing: Add getFuncReturnPtrElementType
[RT] Avoid unnecessary compilations for traversal module Fix RTIP-overriding for GPURT rayQuery functions
Fix #include file ordering
[LGC] Set new .uses_cps PAL metadata where applicable [CompilerUtils] Simplify pointee types metadata
[Continuations] Use
lgc.cps.jump
inLegacyCleanupContinuations
[CompilerUtils] Move promotePointerArguments etc to compilerutils Use function linkage setting from spirv[CompilerUtils] Fix cross-module-inliner tests, and move to subdir [LGC] Add option to force user data spilling
Improve continuations heuristic
[Continuations] Fix clang-format
Remove SpirvLowerRayQuery from lower passes
[Continuations] Cleanup continuations analyses, remove registercount + statesize metadata [Continuations] Prevent zero-size
lgc.cps.free
calls inCleanupContinuations
[LGC] Replace threadMaskedSelect function to use inverseballot [LGC] Only insert non null components in reassembleOutputExportCalls [Continuations] Use newreplaceVarArgs
helper from dialects [Continuations] Add constant jump 'inliner' pass[Continuations] Basic Continuations 'linter' pass
[Continuations] Introduce lgc.cps.jump in
LowerRaytracingPipeline
Remove explicit LLVM benchmark targets from CMake[Continuations] Handle _AmdValue* intrinsics
[LGC] Fix issues of FS without exports
Fix dEQP-VK.subgroups.arithmetic.ray_tracing* crash Update llvm-dialects submodule
Pack mesh shader outputs in LDS space