[all-commits] [llvm/llvm-project] b00c4f: [Matrix][IR] Cap stride bitwidth at 64 (#163729)

[Nathan Corbyn via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b00c4ff4b9d6e55a00bb2fcc71dbcd37e1305188
      https://github.com/llvm/llvm-project/commit/b00c4ff4b9d6e55a00bb2fcc71dbcd37e1305188
  Author: Nathan Corbyn <n_corbyn at apple.com>
  Date:   2025-10-17 (Fri, 17 Oct 2025)

  Changed paths:
    M llvm/docs/LangRef.rst
    M llvm/lib/IR/Verifier.cpp
    M llvm/test/Transforms/LowerMatrixIntrinsics/data-layout-multiply-fused.ll
    M llvm/test/Transforms/LowerMatrixIntrinsics/data-layout.ll
    M llvm/test/Verifier/matrix-intrinsics.ll

  Log Message:
  -----------
  [Matrix][IR] Cap stride bitwidth at 64 (#163729)

a1ef81d added overloads for `llvm.matrix.column.major.store` and
`llvm.matrix.column.major.load` that allow strides to occupy an
arbitrary bitwidth. This change wasn't reflected in the verifier,
causing an assertion to trip when given strides overflowing 64-bit. This
patch explicitly caps the bitwidth at 64, repairing the crash and
avoiding future complexity dealing with strides that overflow 64 bits.

PR: https://github.com/llvm/llvm-project/pull/163729



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