[llvm] [AMDGPU][True16][CodeGen] FLAT_load using D16 pseudo instruction (PR #114500)

Tue Feb 18 08:12:09 PST 2025

llvm-ci wrote:

LLVM Buildbot has detected a new failure on builder `ml-opt-devrel-x86-64` running on `ml-opt-devrel-x86-64-b2` while building `llvm` at step 5 "build-unified-tree".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/175/builds/13380

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

```
Step 5 (build-unified-tree) failure: build (failure)
...
121.738 [282/64/3488] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/R600OpenCLImageTypeLoweringPass.cpp.o
121.957 [281/64/3489] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSetWavePriority.cpp.o
123.520 [280/64/3490] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUWaitSGPRHazards.cpp.o
124.197 [279/64/3491] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUCombinerHelper.cpp.o
125.082 [278/64/3492] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPULowerKernelArguments.cpp.o
129.077 [277/64/3493] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPURegBankLegalizeRules.cpp.o
129.493 [276/64/3494] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUAtomicOptimizer.cpp.o
129.650 [275/64/3495] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNDPPCombine.cpp.o
131.491 [274/64/3496] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPULibCalls.cpp.o
138.304 [273/64/3497] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMCInstLower.cpp.o
FAILED: lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMCInstLower.cpp.o 
ccache /usr/bin/c++ -DCPUINFO_SUPPORTED_PLATFORM=1 -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/b/ml-opt-devrel-x86-64-b1/build/lib/Target/AMDGPU -I/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU -I/var/lib/buildbot/.local/lib/python3.7/site-packages/tensorflow/include -I/b/ml-opt-devrel-x86-64-b1/build/include -I/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/include -isystem /tmp/tflitebuild/tensorflow/include -isystem /tmp/tflitebuild/eigen/include/eigen3 -isystem /tmp/tflitebuild/abseil-cpp/include -isystem /tmp/tflitebuild/flatbuffers/include -isystem /tmp/tflitebuild/gemmlowp/include/gemmlowp -isystem /tmp/tflitebuild/ml_dtypes/src/ml_dtypes -isystem /tmp/tflitebuild/ml_dtypes/src/ml_dtypes/ml_dtypes -isystem /tmp/tflitebuild/ruy/include -isystem /tmp/tflitebuild/cpuinfo/include -isystem /tmp/tflitebuild/ARM_NEON_2_x86_SSE/include -fPIC -fno-semantic-interposition -fvisibility-inlines-hidden -Werror=date-time -fno-lifetime-dse -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wno-missing-field-initializers -pedantic -Wno-long-long -Wimplicit-fallthrough -Wno-uninitialized -Wno-nonnull -Wno-class-memaccess -Wno-redundant-move -Wno-pessimizing-move -Wno-noexcept-type -Wdelete-non-virtual-dtor -Wsuggest-override -Wno-comment -Wno-misleading-indentation -fdiagnostics-color -ffunction-sections -fdata-sections -O3 -DNDEBUG -fvisibility=hidden  -fno-exceptions -funwind-tables -fno-rtti -UNDEBUG -DEIGEN_NEON_GEBP_NR=4 -DTFL_STATIC_LIBRARY_BUILD -std=c++17 -MD -MT lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMCInstLower.cpp.o -MF lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMCInstLower.cpp.o.d -o lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMCInstLower.cpp.o -c /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp
/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp: In member function ‘void AMDGPUMCInstLower::lowerT16D16Helper(const llvm::MachineInstr*, llvm::MCInst&) const’:
/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:126: error: ‘OPERAND_LAST’ is not a member of ‘llvm::AMDGPU::OpName’
  126 |   uint16_t OpName = AMDGPU::OpName::OPERAND_LAST;
      | 
/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:129: error: cannot convert ‘llvm::AMDGPU::OpName’ to ‘uint16_t’ {aka ‘short unsigned int’} in assignment
  129 |       OpName = llvm::AMDGPU::OpName::vdst;
      | 
/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:131: error: cannot convert ‘llvm::AMDGPU::OpName’ to ‘uint16_t’ {aka ‘short unsigned int’} in assignment
  131 |       OpName = llvm::AMDGPU::OpName::data0;
      | 
/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:136: error: cannot convert ‘llvm::AMDGPU::OpName’ to ‘uint16_t’ {aka ‘short unsigned int’} in assignment
  135 |     OpName = AMDGPU::hasNamedOperand(Opcode, llvm::AMDGPU::OpName::vdata)
      |     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  136 |                  ? llvm::AMDGPU::OpName::vdata
      |                  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  137 |                  : llvm::AMDGPU::OpName::vdst;
      | 
/b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:141: error: cannot convert ‘uint16_t’ {aka ‘short unsigned int’} to ‘llvm::AMDGPU::OpName’
  141 |   int VDstOrVDataIdx = AMDGPU::getNamedOperandIdx(Opcode, OpName);
      | 
In file included from /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h:24,
                 from /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/SIInstrInfo.h:20,
                 from /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/GCNSubtarget.h:22,
                 from /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h:17,
                 from /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.h:18,
                 from /b/ml-opt-devrel-x86-64-b1/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:15:
/b/ml-opt-devrel-x86-64-b1/build/lib/Target/AMDGPU/AMDGPUGenInstrInfo.inc:135914: note:   initializing argument 2 of ‘int16_t llvm::AMDGPU::getNamedOperandIdx(uint16_t, llvm::AMDGPU::OpName)’
135914 | int16_t getNamedOperandIdx(uint16_t Opcode, OpName Name);
       | 
140.241 [273/63/3498] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUPromoteAlloca.cpp.o
140.337 [273/62/3499] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNRewritePartialRegUses.cpp.o
140.808 [273/61/3500] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/R600ClauseMergePass.cpp.o
141.212 [273/60/3501] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUMIRFormatter.cpp.o
141.314 [273/59/3502] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUAlwaysInlinePass.cpp.o
143.357 [273/58/3503] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/R600AsmPrinter.cpp.o
143.661 [273/57/3504] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstCombineIntrinsic.cpp.o
144.502 [273/56/3505] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNPreRAOptimizations.cpp.o

```

</details>

https://github.com/llvm/llvm-project/pull/114500