On Tue, May 28, 2024 at 12:38 PM David Anderson via Dwarf-discuss < dwarf-discuss@lists.dwarfstd.org> wrote:
> Issue: Clarification > > I cannot find any mention of these issues as open > in the DWARF6 issues list. > > Dwarf5, Appendix F, table F.1: > > DW_AT_loclists_base has a checkmark for conventional > but no checkmark for Skeleton and Split. > It should have a checkmark for Skeleton. > I don't think it should - the rnglists_base in the skeleton only applies to ranges in the skeleton. (ranges in the split unit use the ranges in debug_rnglists.dwo via sort of an implicit rnglists_base). There aren't any CU level attributes that'd use a loclist, are there? So I wouldn't expect there to be a loclists_base on a skeleton unit. Perhaps it should have a checkmark for loclists_base+split, since I guess we changed DWARFv6 to have an explicit rnglists_base in the split CU to refer to rnglists.dwo? > > DW_AT_base_types has a checkmark for conventional > but no checkmark for Skeleton and Split. > It should have a checkmark for Split Full and Split Type. > Sec 3.1.2 page 47 line 35 says it is not defined > for a skeleton compilation unit. > This probably falls under the eternal issue for myself and Cary to look into - Split DWARF doesn't offer any way to refer to another unit via a cross-CU reference, so I don't think DW_AT_base_types can be used with Split DWARF as-is. > > DW_AT_rnglists_base is corrected in current draft > of DWARF6, has a checkmark for Skeleton in Skeleton and Split > > David Anderson > 28 May 2024 > -- > Dwarf-discuss mailing list > Dwarf-discuss@lists.dwarfstd.org > https://lists.dwarfstd.org/mailman/listinfo/dwarf-discuss >
-- Dwarf-discuss mailing list Dwarf-discuss@lists.dwarfstd.org https://lists.dwarfstd.org/mailman/listinfo/dwarf-discuss
