[InstCombine] remove unneeded one-use checks for icmp fold

Related folds were added in:
rL125734
...the code comment about register pressure is discussed in
more detail in:
https://bugs.llvm.org/show_bug.cgi?id=2698

But 10 years later, perf testing bzip2 with this change now
shows a slight (0.2% average) improvement on Haswell although
that's probably within test noise.

Given that this is IR canonicalization, we shouldn't be worried
about register pressure though; the backend should be able to
adjust for that as needed.

This is part of solving PR43310 the theoretically right way:
https://bugs.llvm.org/show_bug.cgi?id=43310
...ie, if we don't cripple basic transforms, then we won't
need to add special-case code to detect larger patterns.

rL371940 and rL371981 are related patches in this series.

llvm-svn: 372007

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
index 5c443fa..e569ae8 100644 (file)
--- a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
@@ -3697,11 +3697,8 @@ Instruction *InstCombiner::foldICmpBinOp(ICmpInst &I) {
                         C == Op0 ? D : C);
 
   // icmp (A+B), (A+D) -> icmp B, D for equalities or if there is no overflow.
-  // TODO: The one-use checks should not be necessary.
   if (A && C && (A == C || A == D || B == C || B == D) && NoOp0WrapProblem &&
-      NoOp1WrapProblem &&
-      // Try not to increase register pressure.
-      BO0->hasOneUse() && BO1->hasOneUse()) {
+      NoOp1WrapProblem) {
     // Determine Y and Z in the form icmp (X+Y), (X+Z).
     Value *Y, *Z;
     if (A == C) {

diff --git a/llvm/test/Transforms/InstCombine/icmp-add.ll b/llvm/test/Transforms/InstCombine/icmp-add.ll
index ffbfbab..1e3875b 100644 (file)
--- a/llvm/test/Transforms/InstCombine/icmp-add.ll
+++ b/llvm/test/Transforms/InstCombine/icmp-add.ll
@@ -483,7 +483,7 @@ define i1 @common_op_nsw_extra_uses(i32 %x, i32 %y, i32 %z) {
 ; CHECK-NEXT:    call void @use(i32 [[LHS]])
 ; CHECK-NEXT:    [[RHS:%.*]] = add nsw i32 [[Y:%.*]], [[Z]]
 ; CHECK-NEXT:    call void @use(i32 [[RHS]])
-; CHECK-NEXT:    [[C:%.*]] = icmp sgt i32 [[LHS]], [[RHS]]
+; CHECK-NEXT:    [[C:%.*]] = icmp sgt i32 [[X]], [[Y]]
 ; CHECK-NEXT:    ret i1 [[C]]
 ;
   %lhs = add nsw i32 %x, %z
@@ -512,7 +512,7 @@ define i1 @common_op_nuw_extra_uses(i32 %x, i32 %y, i32 %z) {
 ; CHECK-NEXT:    call void @use(i32 [[LHS]])
 ; CHECK-NEXT:    [[RHS:%.*]] = add nuw i32 [[Z]], [[Y:%.*]]
 ; CHECK-NEXT:    call void @use(i32 [[RHS]])
-; CHECK-NEXT:    [[C:%.*]] = icmp ugt i32 [[LHS]], [[RHS]]
+; CHECK-NEXT:    [[C:%.*]] = icmp ugt i32 [[X]], [[Y]]
 ; CHECK-NEXT:    ret i1 [[C]]
 ;
   %lhs = add nuw i32 %x, %z
@@ -609,8 +609,7 @@ define void @bzip1(i8 %a, i8 %b, i8 %x) {
 define void @bzip2(i8 %a, i8 %b, i8 %x) {
 ; CHECK-LABEL: @bzip2(
 ; CHECK-NEXT:    [[ADD1:%.*]] = add i8 [[A:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[ADD2:%.*]] = add i8 [[B:%.*]], [[X]]
-; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[ADD1]], [[ADD2]]
+; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[A]], [[B:%.*]]
 ; CHECK-NEXT:    call void @use1(i1 [[CMP]])
 ; CHECK-NEXT:    call void @use8(i8 [[ADD1]])
 ; CHECK-NEXT:    ret void