| Issue |
152749
|
| Summary |
[Clang][X86] Replace F16C __builtin_ia32_vcvtph2ps/256 intrinsics with __builtin_convertvector
|
| Labels |
good first issue,
backend:X86,
clang:frontend,
clang:headers
|
| Assignees |
|
| Reporter |
RKSimon
|
Similar to #152461, there doesn't appear to be any reason to have these when we can do the same with __builtin_shufflevector + __builtin_convertvector
Note this is just for PH2PS - not PS2PH which requires the rounding control immediate
![]()
_______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
