DavidSpickett wrote: >Thats what I need suggestions and inputs what changes we can accommodate to >better support the DWARF64.
Start by looking at the main differences DWARF64 has. Write DWARF64 specific tests that cover those corner cases e.g. that some symbol can now be described as a greater size or have a greater offset, whatever it is. I'm assuming the answer isn't "everything changes", because the changes in this PR are very minimal. Run `check-lldb` with DWARF64 enabled everywhere you can. So that means adding it to CMAKE_C/CXX_FLAGS for the build, to `lldb/test/Shell/helper/build.py` for shell tests and somewhere to make the API tests use it. Istr there is a makefile for that. Anything that fails there, reduce the failure down to a targeted test case that can be run anywhere without having to run the whole test suite with DWARF64 enabled. So if you fail to set a breakpoint because the symbol lookup didn't work, reduce that to a test that constructs the DWARF64 data manually and does a symbol lookup. Then we end up with DWARF64 specific tests that can run on all the existing buildbots. https://github.com/llvm/llvm-project/pull/145645 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
