================
@@ -4149,6 +4181,39 @@ bool InterpretBuiltin(InterpState &S, CodePtr OpPC,
const CallExpr *Call,
case X86::BI__builtin_ia32_vec_set_v4di:
return interp__builtin_vec_set(S, OpPC, Call, BuiltinID);
+ case X86::BI__builtin_ia32_pslldqi128_byteshift:
+ case X86::BI__builtin_ia32_pslldqi256_byteshift:
+ // These SLLDQ intrinsics always operate on byte elements (8 bits).
+ // The lane width is hardcoded to 16 to match the SIMD register size,
+ // but the algorithm processes one byte per iteration,
+ // so APInt(8, ...) is correct and intentional.
+ return interp__builtin_x86_byteshift(
+ S, OpPC, Call, BuiltinID,
+ [](const Pointer &Src, unsigned Lane, unsigned I, unsigned Shift) {
+ if (I < Shift) {
+ return APInt(8, 0);
+ }
+ return APInt(
+ 8, static_cast<uint8_t>(Src.elem<uint8_t>(Lane + I - Shift)));
+ });
+
+ case X86::BI__builtin_ia32_psrldqi128_byteshift:
+ case X86::BI__builtin_ia32_psrldqi256_byteshift:
+ // These SRLDQ intrinsics always operate on byte elements (8 bits).
+ // The lane width is hardcoded to 16 to match the SIMD register size,
+ // but the algorithm processes one byte per iteration,
+ // so APInt(8, ...) is correct and intentional.
+ return interp__builtin_x86_byteshift(
+ S, OpPC, Call, BuiltinID,
+ [](const Pointer &Src, unsigned Lane, unsigned I, unsigned Shift) {
+ if (I + Shift < 16) {
+ return APInt(
+ 8, static_cast<uint8_t>(Src.elem<uint8_t>(Lane + I + Shift)));
----------------
RKSimon wrote:
Is the static_cast<uint8_t> necessary?
https://github.com/llvm/llvm-project/pull/164166
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
