Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Tom Tromey <tom@tromey.com>
To: Matthieu Longo <matthieu.longo@arm.com>
Cc: Tom Tromey <tom@tromey.com>,  gdb-patches@sourceware.org
Subject: Re: [PATCH v2 4/9] gdb: add new helpers for retrieving a type's fully qualified name
Date: Fri, 06 Mar 2026 12:45:40 -0700	[thread overview]
Message-ID: <878qc4yc63.fsf@tromey.com> (raw)
In-Reply-To: <40f9c944-66e7-4a6d-8b86-80c7796feb6b@arm.com> (Matthieu Longo's message of "Fri, 6 Mar 2026 17:49:09 +0000")

>> I think this case can't actually be hit... can it?

> It currently can when you build GDB with python 3.11 for instance with
> --enable-py-limited-api=yes.  Unfortunately, I haven't added the logic
> to error when someone requests to build GDB against the limited API,
> but does not have a suitable version of Python.  This #error were a
> workaround to provide a meaningful message to the developer in this
> case.

Ah ok, I see.

I guess I was assuming that we'd get an error from "somewhere" (Python I
guess) since gdb would be requesting a limited API with a version newer
than what's available.

> Here is the fix to remove #error macros, and detect the issue at configure time.
> I will split the configure fix from the removal of the #error macros in the next revision.

> +  # The choice of the minimal version for the Python limited API won't be
> +  # frozen until the end of the migration.
> +  AM_PYTHON_CHECK_VERSION([$python_prog], [3.14], [
> +    AC_DEFINE(Py_LIMITED_API, 0x030e0000,
> +      [Define if GDB should be built against the Python limited C API.])
> +    ],[
> +      AC_MSG_ERROR([Python limited API support requires at least Python version 3.14])
> +    ])
>  fi

Seems reasonable, thanks.

FWIW for a speciality configure option like this, I think it's also fine
to bail out at build time, like just a check and a #error in python-internal.h.
But configure failing is definitely a bit friendlier.

thanks,
Tom

  reply	other threads:[~2026-03-06 19:46 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-03 16:16 [PATCH v2 0/9] gdb: more fixes for Python limited C API support Matthieu Longo
2026-03-03 16:16 ` [PATCH v2 1/9] gdb: switch tuple object helpers to Python limited API equivalents Matthieu Longo
2026-03-03 18:09   ` Tom Tromey
2026-03-03 16:16 ` [PATCH v2 2/9] gdb: introduce rgb_color type to simplify existing code Matthieu Longo
2026-03-03 18:16   ` Tom Tromey
2026-03-04 16:30     ` Matthieu Longo
2026-03-03 16:16 ` [PATCH v2 3/9] gdb: switch bytes object helpers to Python limited API equivalents Matthieu Longo
2026-03-03 18:03   ` Tom Tromey
2026-03-03 16:16 ` [PATCH v2 4/9] gdb: add new helpers for retrieving a type's fully qualified name Matthieu Longo
2026-03-03 18:59   ` Tom Tromey
2026-03-06 17:49     ` Matthieu Longo
2026-03-06 19:45       ` Tom Tromey [this message]
2026-03-03 16:16 ` [PATCH v2 5/9] gdb/python: allow ref_ptr<T, Policy>::new_reference to accept subclasses of T Matthieu Longo
2026-03-03 18:18   ` Tom Tromey
2026-03-04 16:56     ` Matthieu Longo
2026-03-04 18:55       ` Tom Tromey
2026-03-06 11:37     ` Matthieu Longo
2026-03-06 11:43       ` Matthieu Longo
2026-03-06 16:47       ` Tom Tromey
2026-03-09 11:38         ` Matthieu Longo
2026-03-03 16:16 ` [PATCH v2 6/9] gdb/python: flatten functions calling PyObject_New and use gdbpy_ref Matthieu Longo
2026-03-03 18:22   ` Tom Tromey
2026-03-09 11:41     ` Matthieu Longo
2026-03-03 18:22   ` Tom Tromey
2026-03-03 16:16 ` [PATCH v2 7/9] gdb/python: accept gdbpy_ref in init helpers and return bool Matthieu Longo
2026-03-03 18:24   ` Tom Tromey
2026-03-09 13:25     ` Matthieu Longo
2026-03-03 16:16 ` [PATCH v2 8/9] gdb/python: add gdbpy_dict_wrapper:allocate_dict helper Matthieu Longo
2026-03-03 18:30   ` Tom Tromey
2026-03-06 12:03     ` Matthieu Longo
2026-03-03 16:16 ` [PATCH v2 9/9] gdb/python: add accessor helpers for __dict__ in Python extension objects Matthieu Longo
2026-03-03 19:02   ` Tom Tromey
2026-03-06 14:33     ` Matthieu Longo
2026-03-06 16:04       ` Tom Tromey

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=878qc4yc63.fsf@tromey.com \
    --to=tom@tromey.com \
    --cc=gdb-patches@sourceware.org \
    --cc=matthieu.longo@arm.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