[all-commits] [llvm/llvm-project] 2e3f4e: [SPIRV] Legalize Cast Instructions for Matrix Type...

[Farzon Lotfi via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2e3f4ef5855e07f4bd2b0a1270e6101beceb1f8c
      https://github.com/llvm/llvm-project/commit/2e3f4ef5855e07f4bd2b0a1270e6101beceb1f8c
  Author: Farzon Lotfi <farzonlotfi at microsoft.com>
  Date:   2026-07-30 (Thu, 30 Jul 2026)

  Changed paths:
    M llvm/lib/Target/SPIRV/SPIRVLegalizerInfo.cpp
    M llvm/lib/Target/SPIRV/SPIRVPostLegalizer.cpp
    A llvm/test/CodeGen/SPIRV/legalization/matrix-wide-vector-shader.ll

  Log Message:
  -----------
  [SPIRV] Legalize Cast Instructions for Matrix Types (#212619)

fixes #186864

This is a more limited fix. that doesn't require us to change the whole
legalization strategy for SPIR-V.

To get the last fix to work it looks like I would have to have changed
every global opcode to no longer use power of 2 based legalization.

This change with power of 2 fixed the transpose case in the offload
testsuite when we do the conversion of bools to i32s before we call
transpose.

There remains an issue when a i1 makes it into a transpose intrinsic.
That said I can only trigger that case via llvmir and not HLSL so will
move on for now.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications