[DAGCombiner][X86] Fold (not (neg X)) -> (add X, -1)

This is a special case of a more general transform (not (sub Y, X)) -> (add X, ~Y). InstCombine knows the general form. I've restricted to the special case to fix the motivating case PR42118. I tried handling any case where Y was constant, but got some changes on some Mips tests that I couldn't quickly prove where beneficial.

Fixes PR42118

Differential Revision: https://reviews.llvm.org/D62828

llvm-svn: 362533

2 files changed, 20 insertions, 17 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index 33ef68c2f1f..b69936d462c 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -6789,6 +6789,16 @@ SDValue DAGCombiner::visitXOR(SDNode *N) {
       return DAG.getNode(NewOpcode, DL, VT, LHS, RHS);
     }
   }
+
+  // fold (not (neg x)) -> (add X, -1)
+  // FIXME: This can be generalized to (not (sub Y, X)) -> (add X, ~Y) if
+  // Y is a constant or the subtract has a single use.
+  if (isAllOnesConstant(N1) && N0.getOpcode() == ISD::SUB &&
+      isNullConstant(N0.getOperand(0))) {
+    return DAG.getNode(ISD::ADD, DL, VT, N0.getOperand(1),
+                       DAG.getAllOnesConstant(DL, VT));
+  }
+
   // fold (xor (and x, y), y) -> (and (not x), y)
   if (N0Opcode == ISD::AND && N0.hasOneUse() && N0->getOperand(1) == N1) {
     SDValue X = N0.getOperand(0);
diff --git a/llvm/test/CodeGen/X86/bmi.ll b/llvm/test/CodeGen/X86/bmi.ll
index c48be66705f..c2d13c29710 100644
--- a/llvm/test/CodeGen/X86/bmi.ll
+++ b/llvm/test/CodeGen/X86/bmi.ll
@@ -1153,10 +1153,7 @@ declare void @bar()
 define void @pr42118_i32(i32 %x) {
 ; X86-LABEL: pr42118_i32:
 ; X86:       # %bb.0:
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl %eax, %ecx
-; X86-NEXT:    negl %ecx
-; X86-NEXT:    andnl %eax, %ecx, %eax
+; X86-NEXT:    blsrl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    jne .LBB48_1
 ; X86-NEXT:  # %bb.2:
 ; X86-NEXT:    jmp bar # TAILCALL
@@ -1165,9 +1162,7 @@ define void @pr42118_i32(i32 %x) {
 ;
 ; X64-LABEL: pr42118_i32:
 ; X64:       # %bb.0:
-; X64-NEXT:    movl %edi, %eax
-; X64-NEXT:    negl %eax
-; X64-NEXT:    andnl %edi, %eax, %eax
+; X64-NEXT:    blsrl %edi, %eax
 ; X64-NEXT:    jne .LBB48_1
 ; X64-NEXT:  # %bb.2:
 ; X64-NEXT:    jmp bar # TAILCALL
@@ -1192,13 +1187,13 @@ define void @pr42118_i64(i64 %x) {
 ; X86-NEXT:    .cfi_offset %esi, -8
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    xorl %edx, %edx
-; X86-NEXT:    movl %eax, %esi
-; X86-NEXT:    negl %esi
-; X86-NEXT:    sbbl %ecx, %edx
-; X86-NEXT:    andnl %ecx, %edx, %ecx
-; X86-NEXT:    andnl %eax, %esi, %eax
-; X86-NEXT:    orl %ecx, %eax
+; X86-NEXT:    movl %eax, %edx
+; X86-NEXT:    addl $-1, %edx
+; X86-NEXT:    movl %ecx, %esi
+; X86-NEXT:    adcl $-1, %esi
+; X86-NEXT:    andl %eax, %edx
+; X86-NEXT:    andl %ecx, %esi
+; X86-NEXT:    orl %edx, %esi
 ; X86-NEXT:    jne .LBB49_1
 ; X86-NEXT:  # %bb.2:
 ; X86-NEXT:    popl %esi
@@ -1212,9 +1207,7 @@ define void @pr42118_i64(i64 %x) {
 ;
 ; X64-LABEL: pr42118_i64:
 ; X64:       # %bb.0:
-; X64-NEXT:    movq %rdi, %rax
-; X64-NEXT:    negq %rax
-; X64-NEXT:    andnq %rdi, %rax, %rax
+; X64-NEXT:    blsrq %rdi, %rax
 ; X64-NEXT:    jne .LBB49_1
 ; X64-NEXT:  # %bb.2:
 ; X64-NEXT:    jmp bar # TAILCALL