[llvm] [LLVM-Reduce] - Null pointer handling during distinct metadata reduction (PR #117570)

[Robert Barinov via llvm-commits]

https://github.com/rbintel updated https://github.com/llvm/llvm-project/pull/117570

>From 6b4102ebcf53a3c768aa437914c9987b9b4341fa Mon Sep 17 00:00:00 2001
From: "Barinov, Robert" <robert.barinov at intel.com>
Date: Mon, 25 Nov 2024 16:18:09 +0100
Subject: [PATCH] Some distinct metadata nodes, e.g DICompileUnit, have
 implicit nullptrs inside them. Iterating over them with dyn_cast leads to a
 crash, change the behavior so that the nullptr operands are skipped.

Add the test distinct-metadata-nullptr.ll which will crash if null
pointers are not handled correctly.
---
 .../llvm-reduce/distinct-dimetadata-nullptr.ll  | 17 +++++++++++++++++
 .../deltas/ReduceDistinctMetadata.cpp           |  9 +++++----
 2 files changed, 22 insertions(+), 4 deletions(-)
 create mode 100644 llvm/test/tools/llvm-reduce/distinct-dimetadata-nullptr.ll

diff --git a/llvm/test/tools/llvm-reduce/distinct-dimetadata-nullptr.ll b/llvm/test/tools/llvm-reduce/distinct-dimetadata-nullptr.ll
new file mode 100644
index 00000000000000..5861adc65b83c5
--- /dev/null
+++ b/llvm/test/tools/llvm-reduce/distinct-dimetadata-nullptr.ll
@@ -0,0 +1,17 @@
+; Test checking that distinct metadata reduction pass handles null pointers properly.
+; This test will lead to a crash if nullptrs inside distinct metadata are not handled correctly, in this case inside DICompileUnit
+
+; RUN: llvm-reduce --delta-passes=distinct-metadata --aggressive-named-md-reduction --test FileCheck --test-arg %s --test-arg --input-file %s -o %t
+; CHECK: {{.*}}distinct !DICompileUnit{{.*}}
+
+
+!llvm.module.flags = !{!0, !1, !6}
+!llvm.dbg.cu = !{!4}
+
+!0 = !{i32 7, !"Dwarf Version", i32 4}
+!1 = !{i32 2, !"Source Lang Literal", !2}
+!2 = !{!3}
+!3 = !{!4, i32 33}
+!4 = distinct !DICompileUnit(language: DW_LANG_OpenCL, file: !5, producer: "foobar", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug)
+!5 = !DIFile(filename: "main.cpp", directory: "foodir")
+!6 = !{i32 2, !"Debug Info Version", i32 3}
diff --git a/llvm/tools/llvm-reduce/deltas/ReduceDistinctMetadata.cpp b/llvm/tools/llvm-reduce/deltas/ReduceDistinctMetadata.cpp
index 02129263f6af44..0f46409977a336 100644
--- a/llvm/tools/llvm-reduce/deltas/ReduceDistinctMetadata.cpp
+++ b/llvm/tools/llvm-reduce/deltas/ReduceDistinctMetadata.cpp
@@ -39,7 +39,7 @@ reduceNodes(MDNode *Root,
     // Mark the nodes for removal
     for (unsigned int I = 0; I < CurrentNode->getNumOperands(); ++I) {
       if (MDNode *Operand =
-              dyn_cast<MDNode>(CurrentNode->getOperand(I).get())) {
+              dyn_cast_or_null<MDNode>(CurrentNode->getOperand(I).get())) {
         // Check whether node has been visited
         if (VisitedNodes.insert(Operand))
           NodesToTraverse.push(Operand);
@@ -71,7 +71,7 @@ static void cleanUpTemporaries(NamedMDNode &NamedNode, MDTuple *TemporaryTuple,
   for (auto I = NamedNode.op_begin(); I != NamedNode.op_end(); ++I) {
     // If the node hasn't been traversed yet, add it to the queue of nodes to
     // traverse.
-    if (MDTuple *TupleI = dyn_cast<MDTuple>((*I))) {
+    if (MDTuple *TupleI = dyn_cast_or_null<MDTuple>((*I))) {
       if (VisitedNodes.insert(TupleI))
         NodesToTraverse.push(TupleI);
     }
@@ -108,7 +108,8 @@ static void cleanUpTemporaries(NamedMDNode &NamedNode, MDTuple *TemporaryTuple,
 
     // Push the remaining nodes into the queue
     for (unsigned int I = 0; I < CurrentTuple->getNumOperands(); ++I) {
-      MDTuple *Operand = dyn_cast<MDTuple>(CurrentTuple->getOperand(I).get());
+      MDTuple *Operand =
+          dyn_cast_or_null<MDTuple>(CurrentTuple->getOperand(I).get());
       if (Operand && VisitedNodes.insert(Operand))
         // If the node hasn't been traversed yet, add it to the queue of nodes
         // to traverse.
@@ -127,7 +128,7 @@ static void extractDistinctMetadataFromModule(Oracle &O,
        Program.named_metadata()) { // Iterate over the named nodes
     for (unsigned int I = 0; I < NamedNode.getNumOperands();
          ++I) { // Iterate over first level unnamed nodes..
-      if (MDTuple *Operand = dyn_cast<MDTuple>(NamedNode.getOperand(I)))
+      if (MDTuple *Operand = dyn_cast_or_null<MDTuple>(NamedNode.getOperand(I)))
         reduceNodes(Operand, NodesToDelete, TemporaryTuple, O, Program);
     }
   }