simon_tatham added inline comments.
================ Comment at: clang/include/clang/Basic/arm_mve.td:68 +def "": Intrinsic<Vector, (args Vector:$inactive, Vector:$a, Vector:$b, Predicate:$pred), + (bitcast (IRInt<int_op, [UVector, Predicate]> + (bitcast $a, UVector), ---------------- I think it's worth adding a comment here explaining why this bitcasting isn't overcomplicating the IR in the integer case. (Because `Vector` and `UVector` may be different vector types at the C level – vectors of signed/unsigned ints of the same size – but once they're lowered to IR, they're just vectors of `i16` or `i32` or whatever, with no sign at all, so the bitcasts will be automatically elided by `IRBuilder`.) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D70547/new/ https://reviews.llvm.org/D70547 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
