From: Andrew Pinski via Gdb <gdb@sourceware.org>
To: Pedro Alves <pedro@palves.net>
Cc: gdb@sourceware.org
Subject: Re: [RFC] GDB BoF at the Cauldron 2025 next Saturday
Date: Sun, 21 Sep 2025 16:48:21 -0700 [thread overview]
Message-ID: <CA+=Sn1ne6AgzJMYreoNR8fX8zGiesgVL_KjGh-KtpjX554x_iA@mail.gmail.com> (raw)
In-Reply-To: <c60cd66c-47df-4b69-9029-f27f1ecfa526@palves.net>
On Sun, Sep 21, 2025 at 4:30 PM Pedro Alves <pedro@palves.net> wrote:
>
> Hi all!
>
> Request for comments below.
>
> Cauldron 2025 is just around the corner, starting this coming Friday.
>
> As in previous years, I'll be hosting a GDB BoF session there. It is
> scheduled for Saturday from 14:30 to 15:30:
> https://conf.gnu-tools-cauldron.org/opo25/talk/8CAXUU/
>
> There will be several GDB maintainers, contributors and users in
> attendance, and the goal of the BoF is to take the opportunity for the
> community to meet to discuss all things related to the GNU Debugger
> project.
>
> And as in previous years, I plan to take notes live on the projected
> screen while we discuss, and then share the result here on the list.
>
> I thought I'd bring along a preliminary list of agenda topics to
> discuss, just to break the ice, and send this email to see if there
> are things people want to add to the list.
>
> The list of topics is of course not closed, and you are very welcome
> to bring your own to the session.
>
> Here's my potential list discussion topics. Please feel free to share
> your ideas. They're more than welcome.
>
> - The recurring topics :-)
>
> - Can we / should we require C++20?
>
> The move to C++17 happened around the 2023 BoF. :-)
I think the bigger issue here is GCC still considers C++20 support as
experimental even on the GCC trunk. So there are no ABI guarantees at
this stage for GCC.
Yes, that should be brought up to the GCC about that. Maybe the best
time to start talking about that requirement in GDB would be after a
GCC release where C++20 is not considered experimental.
(and yes there is a C++20 ABI break between GCC 15 and 16 recently
[one was this week and another one was the week before] in a few areas
including in the library support for views).
Thanks,
Andrew
>
> - Patch review/approval mechanisms
>
> Feedback on usage of Approved-by etc. tags?
>
> - Licensing, DCO, etc. There's a Licensing BoF on Sunday. We can
> collect thoughts and take them there.
>
> - Incoming GDB 17.1 release. Anything people want to discuss about
> this? Blockers, etc.?
>
> - DWARF 6 is progressing, with some important changes that allow GPU
> debugging very likely to be accepted there soon. How does the
> community feel about adding experimental DWARF 6 support to
> GDB master, especially the locations on the stack changes?
>
> Issue 230524.1: Location Descriptions on the DWARF Stack
> https://dwarfstd.org/issues/230524.1.html
>
> - Where we do see GDB in 2026 and beyond? Aka, big ticket items we'd
> like see happening in the coming year and after. We can discuss
> potential blockers and if different people and teams can coordinate.
>
> I'll share my own very biased list, please share yours!
>
> - (Another) year of the GPU. :-)
>
> GDB gains support for lane debugging, as per the results of the
> collaboration between AMD and Intel: See "Lane support in GDB for
> debugging GPUs" BoF session on Friday:
> https://conf.gnu-tools-cauldron.org/opo25/talk/FN7FKL/
>
> Intel GPU support is merged to master. GDB ends up with support
> for at least two GPUs (AMD, Intel), maybe more!
>
> We come up with an address spaces user interface that works
> for everyone.
>
> - Windows non-stop support.
>
> - Native Windows testsuite in good shape.
>
> - The libpython ABI issue goes away. (One can dream.) The Python
> linking model situation is a pain point for everyone who distributes
> GDB for multiple distros. Matthieu Longo, are you attending? We can
> discuss the allowing a more recent version of Python at runtime
> than GDB was compiled with project, aka, the Python limited C API
> effort. And we can all share our experiences with the different
> approaches to handle the situation.
>
> - Others? What are people working on? I'd love to hear about it.
>
>
> Looking forward to meeting with all at the Cauldron soon!
> Pedro Alves
prev parent reply other threads:[~2025-09-21 23:49 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-21 23:29 Pedro Alves
2025-09-21 23:48 ` Andrew Pinski via Gdb [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CA+=Sn1ne6AgzJMYreoNR8fX8zGiesgVL_KjGh-KtpjX554x_iA@mail.gmail.com' \
--to=gdb@sourceware.org \
--cc=pedro@palves.net \
--cc=pinskia@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox