[llvm] [SCEV] Match both (-1)b + a and a + (-1)b as a - b (PR #84247)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-llvm-analysis

Author: Philip Reames (preames)

<details>
<summary>Changes</summary>

In our analysis of guarding conditions, we were converting a-b == 0 into a == b alternate form, but we were only checking for one of the two forms for the sub.  There's no requirement that the multiply only be on the LHS of the add.

---
Full diff: https://github.com/llvm/llvm-project/pull/84247.diff


2 Files Affected:

- (modified) llvm/lib/Analysis/ScalarEvolution.cpp (+21-11) 
- (modified) llvm/test/Analysis/ScalarEvolution/trip-count.ll (+1-1) 


``````````diff
diff --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp
index 15c2965aede1a0..acc0aa23107bb5 100644
--- a/llvm/lib/Analysis/ScalarEvolution.cpp
+++ b/llvm/lib/Analysis/ScalarEvolution.cpp
@@ -10577,6 +10577,25 @@ static bool HasSameValue(const SCEV *A, const SCEV *B) {
   return false;
 }
 
+static bool MatchBinarySub(const SCEV *S, const SCEV *&LHS, const SCEV *&RHS) {
+  const SCEVAddExpr *Add = dyn_cast<SCEVAddExpr>(S);
+  if (!Add || Add->getNumOperands() != 2)
+    return false;
+  if (auto *ME = dyn_cast<SCEVMulExpr>(Add->getOperand(0));
+      ME && ME->getNumOperands() == 2 && ME->getOperand(0)->isAllOnesValue()) {
+    LHS = Add->getOperand(1);
+    RHS = ME->getOperand(1);
+    return true;
+  }
+  if (auto *ME = dyn_cast<SCEVMulExpr>(Add->getOperand(1));
+      ME && ME->getNumOperands() == 2 && ME->getOperand(0)->isAllOnesValue()) {
+    LHS = Add->getOperand(0);
+    RHS = ME->getOperand(1);
+    return true;
+  }
+  return false;
+}
+
 bool ScalarEvolution::SimplifyICmpOperands(ICmpInst::Predicate &Pred,
                                            const SCEV *&LHS, const SCEV *&RHS,
                                            unsigned Depth) {
@@ -10652,19 +10671,10 @@ bool ScalarEvolution::SimplifyICmpOperands(ICmpInst::Predicate &Pred,
       case ICmpInst::ICMP_EQ:
       case ICmpInst::ICMP_NE:
         // Fold ((-1) * %a) + %b == 0 (equivalent to %b-%a == 0) into %a == %b.
-        if (!RA)
-          if (const SCEVAddExpr *AE = dyn_cast<SCEVAddExpr>(LHS))
-            if (const SCEVMulExpr *ME =
-                    dyn_cast<SCEVMulExpr>(AE->getOperand(0)))
-              if (AE->getNumOperands() == 2 && ME->getNumOperands() == 2 &&
-                  ME->getOperand(0)->isAllOnesValue()) {
-                RHS = AE->getOperand(1);
-                LHS = ME->getOperand(1);
-                Changed = true;
-              }
+        if (RA.isZero() && MatchBinarySub(LHS, LHS, RHS))
+          Changed = true;
         break;
 
-
         // The "Should have been caught earlier!" messages refer to the fact
         // that the ExactCR.isFullSet() or ExactCR.isEmptySet() check above
         // should have fired on the corresponding cases, and canonicalized the
diff --git a/llvm/test/Analysis/ScalarEvolution/trip-count.ll b/llvm/test/Analysis/ScalarEvolution/trip-count.ll
index cbe07effdeb265..8fc5b9b4096127 100644
--- a/llvm/test/Analysis/ScalarEvolution/trip-count.ll
+++ b/llvm/test/Analysis/ScalarEvolution/trip-count.ll
@@ -145,7 +145,7 @@ define void @dual_sext_ne_with_slt_guard(i8 %s, i8 %n) {
 ; CHECK-LABEL: 'dual_sext_ne_with_slt_guard'
 ; CHECK-NEXT:  Determining loop execution counts for: @dual_sext_ne_with_slt_guard
 ; CHECK-NEXT:  Loop %for.body: backedge-taken count is (-1 + (sext i8 %n to i64) + (-1 * (sext i8 %s to i64))<nsw>)
-; CHECK-NEXT:  Loop %for.body: constant max backedge-taken count is i64 -1
+; CHECK-NEXT:  Loop %for.body: constant max backedge-taken count is i64 -2
 ; CHECK-NEXT:  Loop %for.body: symbolic max backedge-taken count is (-1 + (sext i8 %n to i64) + (-1 * (sext i8 %s to i64))<nsw>)
 ; CHECK-NEXT:  Loop %for.body: Trip multiple is 1
 ;

``````````

</details>


https://github.com/llvm/llvm-project/pull/84247