https://github.com/andykaylor requested changes to this pull request.
Thank you for your interest in contributing to ClangIR! Unfortunately, this PR is entirely wrong. It appears you did not build and test with ClangIR enabled and so your changes were not built and your test was not run. In order to build with ClangIR enabled, you will need to build with `-DCLANG_ENABLE_CIR=ON` on your CMake line. It's a good idea to try running your test before making the code change to verify that the test fails without your change. This also proves that your test is being run as expected. The change you have made here is not the correct change. What you need to do is add is handling for the `BI__builtin_ia32_vfmadd*` builtins here: https://github.com/llvm/llvm-project/blob/159dcbfcba88c9d3c7760b804f9b916b2a6bc7ca/clang/lib/CIR/CodeGen/CIRGenBuiltinX86.cpp#L1394-L1429 You can use the classic codegen implementation in clang/lib/CodeGen/TargetBuiltins/X86.cpp as a guide to the correct handling, but you will need to adapt the code to generate CIR instructions. https://github.com/llvm/llvm-project/pull/217326 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
