================ ---------------- RKSimon wrote:
I'm hesitant to use `__builtin_convertvector` as although it looks like we're consistent with fp32 -> bf16 using round rmNearestTiesToEven in dag, I don't know the code well enough to be sure that we do this everywhere - do you think its ok? https://github.com/llvm/llvm-project/pull/169841 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
