[llvm] [X86] Prefer sub-of-not for vector inc-of-add (PR #207695)
via llvm-commits
llvm-commits at lists.llvm.org
Mon Jul 6 04:54:10 PDT 2026
================
@@ -53,19 +53,23 @@ define <1 x i128> @add_v1i128(<1 x i128> %x, <1 x i128> %y) nounwind {
; X86: # %bb.0:
; X86-NEXT: pushl %edi
; X86-NEXT: pushl %esi
-; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X86-NEXT: movl {{[0-9]+}}(%esp), %edx
; X86-NEXT: movl {{[0-9]+}}(%esp), %esi
; X86-NEXT: movl {{[0-9]+}}(%esp), %edi
-; X86-NEXT: addl {{[0-9]+}}(%esp), %esi
-; X86-NEXT: adcl {{[0-9]+}}(%esp), %edi
-; X86-NEXT: adcl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT: adcl {{[0-9]+}}(%esp), %edx
-; X86-NEXT: addl $1, %esi
-; X86-NEXT: adcl $0, %edi
-; X86-NEXT: adcl $0, %ecx
-; X86-NEXT: adcl $0, %edx
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: notl %eax
+; X86-NEXT: subl %eax, %esi
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: notl %eax
+; X86-NEXT: sbbl %eax, %edi
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: notl %eax
+; X86-NEXT: sbbl %eax, %ecx
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: notl %eax
+; X86-NEXT: sbbl %eax, %edx
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: movl %edi, 4(%eax)
; X86-NEXT: movl %esi, (%eax)
; X86-NEXT: movl %ecx, 8(%eax)
----------------
222rohan wrote:
Thank you, added a check for single element vectors as well.
https://github.com/llvm/llvm-project/pull/207695
More information about the llvm-commits
mailing list