[llvm] [IR] Reject unhandled assume bundles and seperate them from normal attributes (PR #197007)

[Nikolas Klauser via llvm-commits]

================
@@ -3647,66 +3648,45 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst &CI) {
       return eraseInstFromFunction(*II);
     }
 
-    for (unsigned Idx = 0; Idx < II->getNumOperandBundles(); Idx++) {
-      OperandBundleUse OBU = II->getOperandBundleAt(Idx);
+    for (auto [Idx, OBU] : llvm::enumerate(II->operand_bundles())) {
+      switch (getBundleAttrFromOBU(OBU)) {
+      case BundleAttr::None:
+        llvm_unreachable("Unexpected Attribute");
+      case BundleAttr::Align: {
+        // Try to remove redundant alignment assumptions.
+        auto [Ptr, _, Alignment, Offset] = getAssumeAlignInfo(OBU);
 
-      // Separate storage assumptions apply to the underlying allocations, not
-      // any particular pointer within them. When evaluating the hints for AA
-      // purposes we getUnderlyingObject them; by precomputing the answers here
-      // we can avoid having to do so repeatedly there.
-      if (OBU.getTagName() == "separate_storage") {
-        assert(OBU.Inputs.size() == 2);
-        auto MaybeSimplifyHint = [&](const Use &U) {
-          Value *Hint = U.get();
-          // Not having a limit is safe because InstCombine removes unreachable
-          // code.
-          Value *UnderlyingObject = getUnderlyingObject(Hint, /*MaxLookup*/ 0);
-          if (Hint != UnderlyingObject)
-            replaceUse(const_cast<Use &>(U), UnderlyingObject);
-        };
-        MaybeSimplifyHint(OBU.Inputs[0]);
-        MaybeSimplifyHint(OBU.Inputs[1]);
-      }
-
-      // Try to remove redundant alignment assumptions.
-      if (OBU.getTagName() == "align" && OBU.Inputs.size() == 2) {
-        RetainedKnowledge RK = getKnowledgeFromOperandInAssume(
-            *cast<AssumeInst>(II), II->arg_size() + Idx);
-        if (!RK || RK.AttrKind != Attribute::Alignment ||
-            !isPowerOf2_64(RK.ArgValue) || !isa<ConstantInt>(RK.IRArgValue))
-          continue;
+        if (Offset || !Alignment || !isPowerOf2_64(*Alignment))
----------------
philnik777 wrote:

I don't think it's a correctness issue here, though definitely unintentional. We lose some information about `Offset`, but all the simplifications below still hold. Anyways, I've added a test to show that we're not simplifying it for now and we can look at it in a separate PR. (Though it looks like the knowledge retention framework has another bug...)

https://github.com/llvm/llvm-project/pull/197007