This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH v2 4/8] DWARF-5 basic functionality
On Wed, Feb 22 2017, Jan Kratochvil wrote:
> What compiler/version/options do you use? I have tried now
> gcc-4.8.5-11.el7.x86_64
> CFLAGS=-O3 CXXFLAGS=-O3 ./configure;make
I saw the warning with an upstream GCC I recently built myself for
s390x. Today I retried with a fresh version from GCC git, with the same
result: gcc (GCC) 7.0.1 20170222 (experimental)
> While I get some warnings/errors none of them is in dwarf2read.c.
>
> I do not want to patch it without reproducibility of the warnings.
Sure. Unfortunately the maybe-uninitialized warnings highly depend on
compiler version, platform, optimization level, etc. Since this is a
false positive, I don't necessarily propose a change in this case.
[...]
>> But then I wonder
>> about the effect of the call to create_debug_type_hash_table in
>> create_all_type_units:
>>
>> create_debug_type_hash_table (NULL, &dwarf2_per_objfile->info, types_htab,
>> rcuh_kind::COMPILE);
>>
>> Is that needed?
>
> Yes. Otherwise GDB would not read definitions of classes during
> -gdwarf-5 -fdebug-types-section
> as then DW_UT_type are located inside .debug_info.
Ah, I see; thanks for explanation. Seems I was misled by the function
description that mentions "all entries in the .debug_types section", but
fails to mention the .debug_info section.
I also stumbled upon the fact that read_comp_unit_head performs the
conditional initialization under "if (section_kind == rcuh_kind::TYPE)"
instead of checking for the appropriate unit type. This works because
section_kind is adjusted accordingly above, but it might be clearer
otherwise.
--
Andreas