[llvm] [llvm-symbolizer] Recognize AIX big archive (PR #150401)

[James Henderson via llvm-commits]

================
@@ -557,57 +558,149 @@ LLVMSymbolizer::getOrCreateObjectPair(const std::string &Path,
   if (!DbgObj)
     DbgObj = Obj;
   ObjectPair Res = std::make_pair(Obj, DbgObj);
-  std::string DbgObjPath = DbgObj->getFileName().str();
   auto Pair =
       ObjectPairForPathArch.emplace(std::make_pair(Path, ArchName), Res);
-  BinaryForPath.find(DbgObjPath)->second.pushEvictor([this, I = Pair.first]() {
-    ObjectPairForPathArch.erase(I);
-  });
+  std::string DbgObjPath = DbgObj->getFileName().str();
+  auto BinIter = BinaryForPath.find(DbgObjPath);
+  if (BinIter != BinaryForPath.end()) {
+    BinIter->second.pushEvictor(
+        [this, I = Pair.first]() { ObjectPairForPathArch.erase(I); });
+  }
   return Res;
 }
 
-Expected<ObjectFile *>
-LLVMSymbolizer::getOrCreateObject(const std::string &Path,
-                                  const std::string &ArchName) {
-  Binary *Bin;
+Expected<object::Binary *>
+LLVMSymbolizer::loadOrGetBinary(const std::string &Path) {
   auto Pair = BinaryForPath.emplace(Path, OwningBinary<Binary>());
   if (!Pair.second) {
-    Bin = Pair.first->second->getBinary();
     recordAccess(Pair.first->second);
-  } else {
-    Expected<OwningBinary<Binary>> BinOrErr = createBinary(Path);
-    if (!BinOrErr)
-      return BinOrErr.takeError();
+    return Pair.first->second->getBinary();
+  }
 
-    CachedBinary &CachedBin = Pair.first->second;
-    CachedBin = std::move(BinOrErr.get());
-    CachedBin.pushEvictor([this, I = Pair.first]() { BinaryForPath.erase(I); });
-    LRUBinaries.push_back(CachedBin);
-    CacheSize += CachedBin.size();
-    Bin = CachedBin->getBinary();
+  Expected<OwningBinary<Binary>> BinOrErr = createBinary(Path);
+  if (!BinOrErr) {
+    BinaryForPath.erase(Pair.first);
----------------
jh7370 wrote:

1) I'm not convinced the old behaviour was wrong. Indeed, it looks deliberate to me. If lookup failed before, it's reasonable to expect that trying to do it again will yield the same result, so we don't want to waste time looking up again and instead can return the dead entry.

2) Please don't try to fix generic bugs in the middle of a PR for adding big archive support. They should be in separate PRs, so that they can be reviewed on their own merits.

https://github.com/llvm/llvm-project/pull/150401