================ @@ -0,0 +1,91 @@ +// REQUIRES: lld + +// This test will make a type that will be compiled differently into two +// different .dwo files in a type unit with the same type hash, but with +// differing contents. I have discovered that the hash for the type unit is +// simply based off of the typename and doesn't seem to differ when the contents +// differ, so that will help us test foreign type units in the .debug_names ---------------- dwblaikie wrote:
Rather than the personal statement, could maybe just be explicit: "Clang's type unit signature is based only on the mangled name of the type, regardless of the contents of the type" (I can tell you that's how it works - that's how I implemented it - it is a violation of the DWARF spec (the spec says to use a content hash - though that content hash is of the semantic contents, not the syntactic content (eg: if the TU contained a type with a single int member, the spec-compliant hash would be the same whether it was the type DIE followed by the int DIE, or the other way around) - so it would still be the same despite some variations in layout, so the index entries still wouldn't be portable to all matched-hash definitions)) https://github.com/llvm/llvm-project/pull/87740 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
