================
@@ -504,6 +772,24 @@ define <4 x i32> @scmp_normal_vectors(<4 x i32> %x, <4 x
i32> %y) nounwind {
; X86-NEXT: popl %edi
; X86-NEXT: popl %ebx
; X86-NEXT: retl $4
+;
+; SETZUCC-LABEL: scmp_normal_vectors:
+; SETZUCC: # %bb.0:
+; SETZUCC-NEXT: movdqa %xmm0, %xmm2
+; SETZUCC-NEXT: pcmpgtd %xmm1, %xmm2
+; SETZUCC-NEXT: pcmpgtd %xmm0, %xmm1
+; SETZUCC-NEXT: psubd %xmm2, %xmm1
+; SETZUCC-NEXT: movdqa %xmm1, %xmm0
+; SETZUCC-NEXT: retq
+;
+; NO-SETZUCC-LABEL: scmp_normal_vectors:
+; NO-SETZUCC: # %bb.0:
+; NO-SETZUCC-NEXT: movdqa %xmm0, %xmm2
+; NO-SETZUCC-NEXT: pcmpgtd %xmm1, %xmm2
+; NO-SETZUCC-NEXT: pcmpgtd %xmm0, %xmm1
+; NO-SETZUCC-NEXT: psubd %xmm2, %xmm1
+; NO-SETZUCC-NEXT: movdqa %xmm1, %xmm0
+; NO-SETZUCC-NEXT: retq
----------------
phoebewang wrote:
These are the same as SSE. We can add a common SSE prefix in the RUN.
https://github.com/llvm/llvm-project/pull/173965
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
