[Dwarf-discuss] Proposal: rnglists_base missing

[Cary Coutant]

Added as Issue 240618.2 <https://dwarfstd.org/issues/240618.2.html>.

-cary

On Tue, Jun 18, 2024 at 3:40 PM David Anderson via Dwarf-discuss <
dwarf-discuss@lists.dwarfstd.org> wrote:

> BACKGROUND:
> References are to DWARF5 unless otherwise indicated.
>
>
> If a split-full CU uses DW_FORM_rnglistx
> is a DW_AT_rnglists_base required?
> Apparently not, though
> in cases I have seen in object files are instances
> of a single rnglist in .debug_rnglists.dwo .
> The operative assumption is consumers will
> simply assume zero as the (missing)
> DW_AT_rnglists_base.
>
> Seen in llvm and gcc.
>
> The intent of this proposal is to get clarity.
>
> An alternative version could state
> If a Split Full Compilation Unit
> refers to .debug_rnglists.dwo
> with  DW_FORM_rnglistx
> the CU DIE must have a DW_AT_rnglists_base
> attribute.
>
>
> PROPOSAL:
>
> At the end Sec F.1 Overview just before
> Table F.1:
>
> If a Split Full Compilation Unit
> refers to .debug_rnglists.dwo
> with  DW_FORM_rnglistx
> and the correct DW_AT_rnglists_base
> would be zero, the DW_AT_rnglists_base
> may be omitted.
>
> David Anderson
> --
> Dwarf-discuss mailing list
> Dwarf-discuss@lists.dwarfstd.org
> https://lists.dwarfstd.org/mailman/listinfo/dwarf-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.dwarfstd.org/pipermail/dwarf-discuss/attachments/20240701/5006c635/attachment.htm>