[PATCH] D71437: [PDB] Print the most redundant type record indices with /summary

[Fangrui Song via Phabricator via llvm-commits]

MaskRay added inline comments.


================
Comment at: lld/COFF/PDB.cpp:431
+    // collecting statistics.
+    tpiCounts.resize(tMerger.getTypeTable().size());
+    ipiCounts.resize(tMerger.getIDTable().size());
----------------
Can `tpiCounts` be a non-empty vector before `resize()`?

If yes, `assign(tMerger.getTypeTable().size(), 0);`


================
Comment at: lld/COFF/PDB.cpp:437
+      if (dstIdx.isSimple())
+        continue; // Untranslated or other.
+      SmallVectorImpl<uint32_t> &counts =
----------------
Excuse my ignorance. What does "Untranslated or other" mean? I cannot find "untranslated" at other call sites of `isSimple`.


================
Comment at: lld/COFF/PDB.cpp:517
+    ipiCounts.resize(tMerger.getIDTable().size());
+    for (auto ti : indexMap.tpiMap)
+      ++tpiCounts[ti.toArrayIndex()];
----------------
auto -> TypeIndex


================
Comment at: lld/COFF/PDB.cpp:1390
+    };
+    SmallVector<TypeSizeInfo, 0> tsis;
+    for (auto e : enumerate(recCounts)) {
----------------
I've seen `SmallVector<..., 0>` in some code. Does that has any advantage over `std::vector<...>`?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71437/new/

https://reviews.llvm.org/D71437