[all-commits] [llvm/llvm-project] db79fb: [msan] Add handlers for AVX masked load/store intr...

[Thurston Dang via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: db79fb2a91df31a07f312f8e061936927ac5c506
      https://github.com/llvm/llvm-project/commit/db79fb2a91df31a07f312f8e061936927ac5c506
  Author: Thurston Dang <thurston at google.com>
  Date:   2025-01-26 (Sun, 26 Jan 2025)

  Changed paths:
    M llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
    M llvm/test/Instrumentation/MemorySanitizer/X86/avx-intrinsics-x86.ll
    M llvm/test/Instrumentation/MemorySanitizer/X86/avx2-intrinsics-x86.ll
    M llvm/test/Instrumentation/MemorySanitizer/i386/avx-intrinsics-i386.ll
    M llvm/test/Instrumentation/MemorySanitizer/i386/avx2-intrinsics-i386.ll

  Log Message:
  -----------
  [msan] Add handlers for AVX masked load/store intrinsics (#123857)

This patch adds explicit support for AVX masked load/store intrinsics,
largely by applying the intrinsics to the shadows (but subtly different
to handleIntrinsicByApplyingToShadow()).

We do not reuse the handleMaskedLoad/Store functions. The key challenge
is that the LLVM masked intrinsics require a vector of booleans, while
AVX masked intrinsics use the MSBs of a vector of integers.
X86InstCombineIntrinsic.cpp::simplifyX86MaskedLoad mentions that the x86
backend does not know how to efficiently convert from a vector of
booleans back into the AVX mask format; therefore, they (and we) do not
reduce AVX masked intrinsics into LLVM masked intrinsics.



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