From: Tom de Vries <tdevries@suse.de>
To: Lancelot SIX <lancelot.six@amd.com>, gdb-patches@sourceware.org
Cc: tom@tromey.com
Subject: Re: [PATCH] gdb/python/python-internal.h: avoid uninitialized constexpr
Date: Tue, 18 Jun 2024 15:00:26 +0200 [thread overview]
Message-ID: <0e8d0e8b-d03c-43fe-a2ac-b26743b331b9@suse.de> (raw)
In-Reply-To: <20240618090952.537478-1-lancelot.six@amd.com>
On 6/18/24 11:09, Lancelot SIX wrote:
> The following recent change introduced a regression when building using
> clang++:
>
> commit 764af878259768bb70c65bdf3f3285c2d6409bbd
> Date: Wed Jun 12 18:58:49 2024 +0200
>
> [gdb/python] Add typesafe wrapper around PyObject_CallMethod
>
> The error message is:
>
> ../../gdb/python/python-internal.h:151:16: error: default initialization of an object of const type 'const char'
> constexpr char gdbpy_method_format;
> ^
> = '\0'
> CXX python/py-block.o
> 1 error generated.
> make[2]: *** [Makefile:1959: python/py-arch.o] Error 1
> make[2]: *** Waiting for unfinished jobs....
> In file included from ../../gdb/python/py-auto-load.c:25:
> ../../gdb/python/python-internal.h:151:16: error: default initialization of an object of const type 'const char'
> constexpr char gdbpy_method_format;
> ^
> = '\0'
> 1 error generated.
> make[2]: *** [Makefile:1959: python/py-auto-load.o] Error 1
> In file included from ../../gdb/python/py-block.c:23:
> ../../gdb/python/python-internal.h:151:16: error: default initialization of an object of const type 'const char'
> constexpr char gdbpy_method_format;
> ^
> = '\0'
> 1 error generated.
>
> This patch fixes this by changing gdbpy_method_format to be a templated
> struct, and only have its specializations define the static constexpr
> member "format". This way, we avoid having an uninitialized constexpr
> expression, regardless of it being instantiated or not.
>
Hi Lancelot,
thanks for looking into this.
This seems to have been my doing, since I dropped the " = '\0'" bit.
I can reproduce the problem.
I also tried adding back the dropped bit, and saw that it fixes the
compilation error.
However, that also brings back the problem I was trying to solve: doing
this:
...
- gdbpy_ref<> result = gdbpy_call_method (m_window, "close");
+ gdbpy_ref<> result = gdbpy_call_method (m_window, "close", 1.0);
...
compiles, using '\0' as format specificier for double, and we only run
into trouble a runtime.
The patch you propose both:
- fixes the build with clang, and
- gives a compile time error for the unsupported format specifier case.
So, LGTM.
Reviewed-By: Tom de Vries <tdevries@suse.de>
Thanks,
- Tom
> Change-Id: I5bec241144f13500ef78daea30f00d01e373692d
> ---
> gdb/python/python-internal.h | 24 ++++++++++++++++++------
> 1 file changed, 18 insertions(+), 6 deletions(-)
>
> diff --git a/gdb/python/python-internal.h b/gdb/python/python-internal.h
> index f4c35babe46..fec0010a444 100644
> --- a/gdb/python/python-internal.h
> +++ b/gdb/python/python-internal.h
> @@ -148,19 +148,31 @@ typedef long Py_hash_t;
> /* A template variable holding the format character (as for
> Py_BuildValue) for a given type. */
> template<typename T>
> -constexpr char gdbpy_method_format;
> +struct gdbpy_method_format {};
>
> template<>
> -constexpr char gdbpy_method_format<gdb_py_longest> = GDB_PY_LL_ARG[0];
> +struct gdbpy_method_format<gdb_py_longest>
> +{
> + static constexpr char format = GDB_PY_LL_ARG[0];
> +};
>
> template<>
> -constexpr char gdbpy_method_format<gdb_py_ulongest> = GDB_PY_LLU_ARG[0];
> +struct gdbpy_method_format<gdb_py_ulongest>
> +{
> + static constexpr char format = GDB_PY_LLU_ARG[0];
> +};
>
> template<>
> -constexpr char gdbpy_method_format<int> = 'i';
> +struct gdbpy_method_format<int>
> +{
> + static constexpr char format = 'i';
> +};
>
> template<>
> -constexpr char gdbpy_method_format<unsigned> = 'I';
> +struct gdbpy_method_format<unsigned>
> +{
> + static constexpr char format = 'I';
> +};
>
> /* A helper function to compute the PyObject_CallMethod /
> Py_BuildValue format given the argument types. */
> @@ -169,7 +181,7 @@ template<typename... Args>
> constexpr std::array<char, sizeof... (Args) + 1>
> gdbpy_make_fmt ()
> {
> - return { gdbpy_method_format<Args>..., '\0' };
> + return { gdbpy_method_format<Args>::format..., '\0' };
> }
>
> /* Typesafe wrapper around PyObject_CallMethod.
>
> base-commit: 0915235d341841ac7f13bd3136991c19b4a6746b
next prev parent reply other threads:[~2024-06-18 12:59 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-18 9:09 Lancelot SIX
2024-06-18 13:00 ` Tom de Vries [this message]
2024-06-18 15:25 ` Tom de Vries
2024-06-18 16:17 ` Tom Tromey
2024-06-18 18:27 ` Lancelot SIX
2024-06-18 19:48 ` Tom Tromey
2024-06-18 21:01 ` Lancelot SIX
2024-06-18 21:03 ` Tom Tromey
2024-06-18 21:26 ` Lancelot SIX
2024-06-19 0:11 ` Tom Tromey
2024-06-19 9:16 ` Lancelot SIX
2024-06-19 13:41 ` Tom Tromey
2024-06-19 14:29 ` Lancelot SIX
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=0e8d0e8b-d03c-43fe-a2ac-b26743b331b9@suse.de \
--to=tdevries@suse.de \
--cc=gdb-patches@sourceware.org \
--cc=lancelot.six@amd.com \
--cc=tom@tromey.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