[PATCH] D107533: Handle encoding personalities of same names but different kinds.

[Vincent Lee via Phabricator via llvm-commits]

thevinster added a comment.

I only reviewed the mechanical changes as I'm not too familiar with the functional changes. I'll let someone more experienced comment on that.



================
Comment at: lld/MachO/UnwindInfoSection.cpp:238
+            oldToNew[iSym->gotIndex + 1] = ds->gotIndex + 1;
+            personality = const_cast<Symbol *>(s);
+          }
----------------
Probably it's just my lack of knowledge here, but is this line useful? It eventually gets overridden in the start of the loop on L228 right?


================
Comment at: lld/MachO/UnwindInfoSection.cpp:402
     auto it = find(personalities, cu->personality);
-    uint32_t personalityIndex; // 1-based index
+    uint32_t personalityIndex = 0; // 1-based index
+
----------------
Why does this need to be initialized to 0? Isn't it always going to be eventually defined below?


================
Comment at: lld/test/MachO/compact-unwind-both-local-and-dylib-personality.s:25
+## postlink check: FIXME: maybe need a better verification here.
+# RUN: llvm-nm %t/a.out | FileCheck %s --check-prefix=POSTCHECK
+# POSTCHECK: {{.*}} U ___gxx_personality_v0
----------------
Shouldn't there also be a check on the symbol table for `b.out` and `c.out`?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107533/new/

https://reviews.llvm.org/D107533