[llvm] [AMDGPU] Introduce "amdgpu-sw-lower-lds" pass to lower LDS accesses. (PR #87265)

Sun Aug 25 21:46:33 PDT 2024

llvm-ci wrote:

LLVM Buildbot has detected a new failure on builder `clang-ppc64le-rhel` running on `ppc64le-clang-rhel-test` while building `llvm` at step 5 "build-unified-tree".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/145/builds/1462

<details>
<summary>Here is the relevant piece of the build log for the reference</summary>

```
Step 5 (build-unified-tree) failure: build (failure)
...
139.820 [203/52/5998] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIMemoryLegalizer.cpp.o
140.808 [203/51/5999] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUPerfHintAnalysis.cpp.o
140.823 [203/50/6000] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/R600TargetTransformInfo.cpp.o
140.905 [203/49/6001] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUPromoteAlloca.cpp.o
140.983 [203/48/6002] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMCInstLower.cpp.o
141.544 [203/47/6003] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPULibCalls.cpp.o
141.562 [203/46/6004] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIFixSGPRCopies.cpp.o
141.931 [203/45/6005] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SILowerControlFlow.cpp.o
141.935 [203/44/6006] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIAnnotateControlFlow.cpp.o
141.942 [203/43/6007] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSwLowerLDS.cpp.o
FAILED: lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSwLowerLDS.cpp.o 
ccache /home/docker/llvm-external-buildbots/clang.17.0.6/bin/clang++ --gcc-toolchain=/gcc-toolchain/usr -DGTEST_HAS_RTTI=0 -D_DEBUG -D_GLIBCXX_ASSERTIONS -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/build/lib/Target/AMDGPU -I/home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/llvm-project/llvm/lib/Target/AMDGPU -I/home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/build/include -I/home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/llvm-project/llvm/include -fPIC -fno-semantic-interposition -fvisibility-inlines-hidden -Werror -Werror=date-time -Werror=unguarded-availability-new -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wmissing-field-initializers -pedantic -Wno-long-long -Wc++98-compat-extra-semi -Wimplicit-fallthrough -Wcovered-switch-default -Wno-noexcept-type -Wnon-virtual-dtor -Wdelete-non-virtual-dtor -Wsuggest-override -Wstring-conversion -Wmisleading-indentation -Wctad-maybe-unsupported -fdiagnostics-color -ffunction-sections -fdata-sections -O3 -DNDEBUG -std=c++17 -fPIC  -fno-exceptions -funwind-tables -fno-rtti -UNDEBUG -MD -MT lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSwLowerLDS.cpp.o -MF lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSwLowerLDS.cpp.o.d -o lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSwLowerLDS.cpp.o -c /home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUSwLowerLDS.cpp
/home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUSwLowerLDS.cpp:260:10: error: moving a local object in a return statement prevents copy elision [-Werror,-Wpessimizing-move]
  260 |   return std::move(OrderedKernels);
      |          ^
/home/buildbots/llvm-external-buildbots/workers/ppc64le-clang-rhel-test/clang-ppc64le-rhel/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUSwLowerLDS.cpp:260:10: note: remove std::move call here
  260 |   return std::move(OrderedKernels);
      |          ^~~~~~~~~~              ~
1 error generated.
141.959 [203/42/6008] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIFormMemoryClauses.cpp.o
142.166 [203/41/6009] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNNSAReassign.cpp.o
142.236 [203/40/6010] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUPreLegalizerCombiner.cpp.o
142.391 [203/39/6011] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUAlwaysInlinePass.cpp.o
142.406 [203/38/6012] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPULateCodeGenPrepare.cpp.o
142.697 [203/37/6013] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SILateBranchLowering.cpp.o
142.701 [203/36/6014] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/R600ISelLowering.cpp.o
143.705 [203/35/6015] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstCombineIntrinsic.cpp.o
143.707 [203/34/6016] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUImageIntrinsicOptimizer.cpp.o
143.716 [203/33/6017] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUAttributor.cpp.o
143.753 [203/32/6018] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIWholeQuadMode.cpp.o
143.754 [203/31/6019] Building CXX object lib/Target/AMDGPU/Utils/CMakeFiles/LLVMAMDGPUUtils.dir/AMDGPUBaseInfo.cpp.o
143.873 [203/30/6020] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUArgumentUsageInfo.cpp.o
144.095 [203/29/6021] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUCallLowering.cpp.o
144.556 [203/28/6022] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNPreRALongBranchReg.cpp.o
144.994 [203/27/6023] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUTargetTransformInfo.cpp.o
145.250 [203/26/6024] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SILowerWWMCopies.cpp.o
145.364 [203/25/6025] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUCodeGenPrepare.cpp.o
145.404 [203/24/6026] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIPreAllocateWWMRegs.cpp.o
145.462 [203/23/6027] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUResourceUsageAnalysis.cpp.o
145.605 [203/22/6028] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPULowerBufferFatPointers.cpp.o
146.084 [203/21/6029] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUPostLegalizerCombiner.cpp.o
146.423 [203/20/6030] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SILowerSGPRSpills.cpp.o
146.519 [203/19/6031] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUISelLowering.cpp.o
146.662 [203/18/6032] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIOptimizeVGPRLiveRange.cpp.o
146.729 [203/17/6033] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUHSAMetadataStreamer.cpp.o
147.205 [203/16/6034] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIFrameLowering.cpp.o
147.214 [203/15/6035] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUAsmPrinter.cpp.o
147.992 [203/14/6036] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPULowerModuleLDSPass.cpp.o
148.008 [203/13/6037] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/SIFoldOperands.cpp.o

```

</details>

https://github.com/llvm/llvm-project/pull/87265