[clang] [llvm] Fix/172104 clang cl simd intrinsics (PR #172116)

[Priyanshu Singh via llvm-commits]

================
@@ -153,9 +153,13 @@ bool X86TargetInfo::initFeatureMap(
     llvm::StringMap<bool> &Features, DiagnosticsEngine &Diags, StringRef CPU,
     const std::vector<std::string> &FeaturesVec) const {
   // FIXME: This *really* should not be here.
-  // X86_64 always has SSE2.
-  if (getTriple().getArch() == llvm::Triple::x86_64)
+  // X86_64 always has SSE2 and SSE4.1 to support common SIMD intrinsics.
----------------
dev-priyanshu15 wrote:

Thanks for the feedback. I've updated the fix to address this concern. The SSE4.1 enablement is now scoped specifically to Windows MSVC targets using getTriple().isWindowsMSVCEnvironment(), which ensures it doesn't affect Linux x86_64 targets. The updated commit has been force-pushed to the branch.

https://github.com/llvm/llvm-project/pull/172116