[llvm] [X86] lowerShuffleAsSplitOrBlend - prefer splitting AVX1-only shuffles if the operands can be freely split (PR #141055)

[via llvm-commits]

github-actions[bot] wrote:

<!--LLVM CODE FORMAT COMMENT: {clang-format}-->


:warning: C/C++ code formatter, clang-format found issues in your code. :warning:

<details>
<summary>
You can test this locally with the following command:
</summary>

``````````bash
git-clang-format --diff HEAD~1 HEAD --extensions cpp -- llvm/lib/Target/X86/X86ISelLowering.cpp
``````````

</details>

<details>
<summary>
View the diff from clang-format here.
</summary>

``````````diff
diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index da9cd4d6c..7abc85445 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -15253,7 +15253,7 @@ static SDValue lowerShuffleAsSplitOrBlend(const SDLoc &DL, MVT VT, SDValue V1,
                                 /*SimpleOnly*/ false);
 
   // Without AVX2, if we can freely split the subvectors then we're better off
-  // performing half width shuffles. 
+  // performing half width shuffles.
   if (!Subtarget.hasAVX2()) {
     SDValue BC1 = peekThroughBitcasts(V1);
     SDValue BC2 = peekThroughBitcasts(V2);

``````````

</details>


https://github.com/llvm/llvm-project/pull/141055