kbsmith1 added a comment. In D107004#2911654 <https://reviews.llvm.org/D107004#2911654>, @lebedev.ri wrote:
> Patch description states what the change is, but not why the change is what > it is. > I don't think this is a good direction to second-guess the user's intentions. I added the reason for the change to the description. I think when a user programs t = _mm_add_ps(a, b); t = _mm_sub_ps(t, b); that they truly did want those intrinsics evaluated like they were written, even with fast math flags. This change to the intrinsic headers makes programming with intrinsics more accurately reflect what the user wrote, even when fast math flags are used. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D107004/new/ https://reviews.llvm.org/D107004 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
