[cfe-dev] clang -g -fsanitize=address produces broken debug info?

[Eric Christopher]

Exactly, we don't produce it at the moment. It's on the list mind you, but ...

-eric

On Mon, Aug 12, 2013 at 11:02 AM, Robinson, Paul
<Paul_Robinson at playstation.sony.com> wrote:
> Clang can produce .debug_aranges?  Somebody asked me about this recently,
> and I couldn’t get clang to produce a non-empty section, unless I was
> compiling an assembler source.
>
> --paulr
>
>
>
> From: cfe-dev-bounces at cs.uiuc.edu [mailto:cfe-dev-bounces at cs.uiuc.edu] On
> Behalf Of David Blaikie
> Sent: Monday, August 12, 2013 10:13 AM
> To: Kal Conley
> Cc: cfe-dev Developers
> Subject: Re: [cfe-dev] clang -g -fsanitize=address produces broken debug
> info?
>
>
>
>
> On Aug 12, 2013 9:55 AM, "Kal Conley" <kcconley at gmail.com> wrote:
>>
>> Hi,
>> DWARF info produced with "clang -g -fsanitize=address" doesn't have a
>> complete .debug_aranges section. LLDB relies on .debug_aranges if it is
>> present, thus LLDB fails to load any debug info for clang binaries built
>> with -fsanitize=address. The LLDB folks believe this is a clang bug. Can
>> someone from the clang side look into this?
>
> Can you describe in what way it is broken? Do you have a reduced test case?
>
>> Thanks,
>> -Kal
>> _______________________________________________
>> cfe-dev mailing list
>> cfe-dev at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>
>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>