FailedChanges

Summary

  1. Revert "[amdgpu] Lower SGPR-to-VGPR copy in the final phase of ISel." (details)
  2. emacs: Add nofree and willreturn to list of attributes (details)
  3. IR: Move denormal mode parsing from MachineFunction to Function (details)
  4. [InstSimplify] fix fmin/fmax miscompile for partial undef vectors (PR47567) (details)
  5. [mlir][StandardToSPIRV] Handle vector of i1 case for lowering zexti to SPIR-V. (details)
Commit 27df1652709ba83d6b07f313297e7c796e36dce1 by Matthew.Arsenault
Revert "[amdgpu] Lower SGPR-to-VGPR copy in the final phase of ISel."

This reverts commit c3492a1aa1b98c8d81b0969d52cea7681f0624c2.

I think this is the wrong strategy and wrong place to do this
transform anyway. Also reverts follow up commit
7d593d0d6905b55ca1124fca5e4d1ebb17203138.
The file was modifiedllvm/test/CodeGen/AMDGPU/waitcnt-vscnt.ll (diff)
The file was removedllvm/test/CodeGen/AMDGPU/sgpr-copy-cse.ll
The file was modifiedllvm/test/CodeGen/AMDGPU/fabs.ll (diff)
The file was modifiedllvm/test/CodeGen/AMDGPU/fneg-fabs.ll (diff)
The file was modifiedllvm/lib/Target/AMDGPU/SIFoldOperands.cpp (diff)
The file was modifiedllvm/lib/Target/AMDGPU/SIISelLowering.cpp (diff)
The file was modifiedllvm/test/CodeGen/AMDGPU/wqm.ll (diff)
Commit 05c02eda4552076dc08ce34866b3d8ee33bbf842 by Matthew.Arsenault
emacs: Add nofree and willreturn to list of attributes
The file was modifiedllvm/utils/emacs/llvm-mode.el (diff)
Commit 751a6c5760b8de591cf241effbdad1b1cae67814 by Matthew.Arsenault
IR: Move denormal mode parsing from MachineFunction to Function

This was just inspecting the IR to begin with, and is useful to check
in some places in the IR.
The file was modifiedllvm/lib/CodeGen/MachineFunction.cpp (diff)
The file was modifiedllvm/include/llvm/IR/Function.h (diff)
The file was modifiedllvm/lib/IR/Function.cpp (diff)
Commit 3f100e64b429b6468e9a2c5b3e7ef7757a06dc48 by spatel
[InstSimplify] fix fmin/fmax miscompile for partial undef vectors (PR47567)

It would also be correct to return the variable operand in these cases,
but eliminating a variable use is probably better for optimization.
The file was modifiedllvm/test/Transforms/InstSimplify/fminmax-folds.ll (diff)
The file was modifiedllvm/lib/Analysis/InstructionSimplify.cpp (diff)
Commit 1909b6ac0dbc2f1306103a5ea7f5e59f2232b133 by hanchung
[mlir][StandardToSPIRV] Handle vector of i1 case for lowering zexti to SPIR-V.

Reviewed By: mravishankar

Differential Revision: https://reviews.llvm.org/D87887
The file was modifiedmlir/test/Conversion/StandardToSPIRV/std-ops-to-spirv.mlir (diff)
The file was modifiedmlir/lib/Conversion/StandardToSPIRV/ConvertStandardToSPIRV.cpp (diff)