On 8/12/22 17:48, Tom de Vries via Gdb-patches wrote:
> Hi,
> 
> [ Requires "[gdb/symtab] Fix assert in set_length" submitted here (
> https://sourceware.org/pipermail/gdb-patches/2022-August/191313.html ). ]
> 
> When running test-case gdb.cp/cpexprs-debug-types.exp on target board
> cc-with-debug-names/gdb:debug_flags=-gdwarf-5, we get an executable with
> a .debug_names section, but no .debug_types section.  For dwarf-5, the TUs
> are no longer put in a separate unit, but instead they're put in the
> .debug_info section.
> 
> When loading the executable, the .debug_names section is silently ignored
> because of this check in dwarf2_read_debug_names:
> ...
>    if (map->tu_count != 0)
>      {
>        /* We can only handle a single .debug_types when we have an
>           index.  */
>        if (per_bfd->types.size () != 1)
>          return false;
> ...
> which triggers because per_bfd->types.size () == 0.
> 
> The intention of the check is to make sure we don't have both .debug_types and
> .zdebug_types sections.
> 

Well, that's what I thought when I read the code, but I actually found 
back the corresponding PR: PR12984, and it turns out it's about object 
files, which can contain:
...
$ grep "\.debug_types" 11.s
         .section 
.debug_types,"G",@progbits,wt.75c042c23a9a07ee,comdat
         .section 
.debug_types,"G",@progbits,wt.c59c413bf50a4607,comdat
...

I've updated the rationale to reflect this.

Also, I've added a test-case.

Committed as attached.

Thanks,
- Tom