clayborg added a comment.

So I am worried by seeing any mention of "load address" and "load bias" in this 
patch. All information that is extracted from the DWARF should be solely in 
terms if "file addresses" IMHO, unless I am not understanding something from 
the new location lists. For mach-o DWARF in .o files, the "file address" will 
be a file address from the .o file (which is fine), but it will need to be 
linked into an executable file address before being handed out. We fix up 
addresses like this for any DW_OP_addr opcodes in location expressions. So 
typically we would have a virtual SymbolFileDWARF method that 
SymbolFileDWARFDebugMap would override and this would use the default 
SymbolFileDWARF::GetLocationList(...) function, and then link up the location 
list or DWARFExpression prior to handing it out. Is there something I am not 
understanding here?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D70532/new/

https://reviews.llvm.org/D70532



_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

Reply via email to