[llvm] [InstCombine] Extend Phi-Icmp use to include or (PR #67682)

[Nikita Popov via llvm-commits]

================
@@ -1441,18 +1441,38 @@ Instruction *InstCombinerImpl::visitPHINode(PHINode &PN) {
         PHIUser->user_back() == &PN) {
       return replaceInstUsesWith(PN, PoisonValue::get(PN.getType()));
     }
-    // When a PHI is used only to be compared with zero, it is safe to replace
-    // an incoming value proved as known nonzero with any non-zero constant.
-    // For example, in the code below, the incoming value %v can be replaced
-    // with any non-zero constant based on the fact that the PHI is only used to
-    // be compared with zero and %v is a known non-zero value:
-    // %v = select %cond, 1, 2
-    // %p = phi [%v, BB] ...
-    //      icmp eq, %p, 0
-    auto *CmpInst = dyn_cast<ICmpInst>(PHIUser);
-    // FIXME: To be simple, handle only integer type for now.
-    if (CmpInst && isa<IntegerType>(PN.getType()) && CmpInst->isEquality() &&
-        match(CmpInst->getOperand(1), m_Zero())) {
+  }
+
+  // When a PHI is used only to be compared with zero, it is safe to replace
+  // an incoming value proved as known nonzero with any non-zero constant.
+  // For example, in the code below, the incoming value %v can be replaced
+  // with any non-zero constant based on the fact that the PHI is only used to
+  // be compared with zero and %v is a known non-zero value:
+  // %v = select %cond, 1, 2
+  // %p = phi [%v, BB] ...
+  //      icmp eq, %p, 0
+  // FIXME: To be simple, handle only integer type for now.
+  // This handles a small number of uses to keep the complexity down, and an
+  // icmp(or(phi)) can equally be replaced with any non-zero constant as the
+  // "or" will only add bits.
+  if (!PN.hasNUsesOrMore(3)) {
+    bool AllUsesOfPhiEndsInCmp = true;
+    for (const auto *U : PN.users()) {
+      auto *CmpInst = dyn_cast<ICmpInst>(U);
+      if (!CmpInst) {
+        // This is always correct as OR only add bits and we are checking
+        // against 0.
+        if (U->hasOneUse() && match(U, m_Or(m_Specific(&PN), m_Value())))
----------------
nikic wrote:

m_c_Or and add commuted test? I don't think anything guarantees the phi node is on the left operand, right?

https://github.com/llvm/llvm-project/pull/67682