[all-commits] [llvm/llvm-project] 90a2e5: [msan] Handle AVX512 mask.compress (#209654)

[Thurston Dang via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 90a2e5010fa88627cd21bd0311833c786be8a213
      https://github.com/llvm/llvm-project/commit/90a2e5010fa88627cd21bd0311833c786be8a213
  Author: Thurston Dang <thurston at google.com>
  Date:   2026-07-15 (Wed, 15 Jul 2026)

  Changed paths:
    M llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
    M llvm/test/Instrumentation/MemorySanitizer/X86/avx512-intrinsics-upgrade.ll
    M llvm/test/Instrumentation/MemorySanitizer/X86/avx512-intrinsics.ll
    M llvm/test/Instrumentation/MemorySanitizer/X86/avx512vl-intrinsics.ll

  Log Message:
  -----------
  [msan] Handle AVX512 mask.compress (#209654)

mask.compress selects elements from the first two operands according to
the mask; thus, we can propagate the shadow by applying the intrinsic to
the input shadows (with the mask unchanged i.e., trailingVerbatimArgs ==
1), using the forceIntegerIntrinsic option that was added in
https://github.com/llvm/llvm-project/pull/207053



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