================
@@ -4191,6 +4250,31 @@ bool InterpretBuiltin(InterpState &S, CodePtr OpPC,
const CallExpr *Call,
case X86::BI__builtin_ia32_selectpd_512:
return interp__builtin_select(S, OpPC, Call);
+ case X86::BI__builtin_ia32_shufps:
+ case X86::BI__builtin_ia32_shufps256:
+ case X86::BI__builtin_ia32_shufps512:
+ case X86::BI__builtin_ia32_shufpd:
+ case X86::BI__builtin_ia32_shufpd256:
+ case X86::BI__builtin_ia32_shufpd512:
+ return interp__builtin_ia32_shuffle_generic(
+ S, OpPC, Call,
+ [](unsigned DstIdx, unsigned ShuffleMask, unsigned ElementSize) {
+ unsigned NumElemPerLane = 128 / ElementSize;
----------------
RKSimon wrote:
We could drop the ElementSize argument and simplify this callback code if we
split the shufps/shufpd callbacks
https://github.com/llvm/llvm-project/pull/164078
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
