* [PATCH] [gdb/python] Document quit_force uses
@ 2026-08-11 8:00 Tom de Vries
2026-08-27 7:50 ` [PING][PATCH] " Tom de Vries
2026-08-27 15:56 ` [PATCH] " Tom Tromey
0 siblings, 2 replies; 3+ messages in thread
From: Tom de Vries @ 2026-08-11 8:00 UTC (permalink / raw)
To: gdb-patches
The Python extension has two functions that convert exceptions from the GDB
domain to the Python domain and vice versa: gdbpy_convert_exception and
gdbpy_handle_exception.
Each function has an exception though, where it calls quit_force:
- for gdbpy_convert_exception, when handling gdb_exception_forced_quit, and
- for gdbpy_handle_exception, when handling SystemExit.
I think it's a good idea to document these exceptions in more detail, and
since AFAICT the reasoning is the same, I've added the same comment in both
places.
---
gdb/python/py-utils.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/gdb/python/py-utils.c b/gdb/python/py-utils.c
index a831b96eb38..1c68b206601 100644
--- a/gdb/python/py-utils.c
+++ b/gdb/python/py-utils.c
@@ -221,7 +221,14 @@ gdbpy_convert_exception (const struct gdb_exception &exception)
if (exception.reason == RETURN_QUIT)
exc_class = PyExc_KeyboardInterrupt;
else if (exception.reason == RETURN_FORCED_QUIT)
- quit_force (NULL, 0);
+ {
+ /* Ideally we'd like to propagate the exception to top-level, where
+ we'll call quit_force. Propagating the exception would ensure that
+ appropriate cleanups are run.
+ OTOH, there is the risk that the exception will be swallowed.
+ For now, don't take this risk and immediately call quit_force. */
+ quit_force (NULL, 0);
+ }
else if (exception.error == MEMORY_ERROR)
exc_class = gdbpy_gdb_memory_error;
else
@@ -485,6 +492,11 @@ gdbpy_handle_exception ()
exit_arg = 1;
}
+ /* Ideally we'd like to propagate the exception to top-level, where
+ we'll call quit_force. Propagating the exception would ensure that
+ appropriate cleanups are run.
+ OTOH, there is the risk that the exception will be swallowed.
+ For now, don't take this risk and immediately call quit_force. */
quit_force (&exit_arg, 0);
}
else if (! fetched_error.type_matches (gdbpy_gdberror_exc)
base-commit: 8e6c4f526a68b796b7bd71d53e4dc459ac56c3ef
--
2.51.0
^ permalink raw reply [flat|nested] 3+ messages in thread* [PING][PATCH] [gdb/python] Document quit_force uses
2026-08-11 8:00 [PATCH] [gdb/python] Document quit_force uses Tom de Vries
@ 2026-08-27 7:50 ` Tom de Vries
2026-08-27 15:56 ` [PATCH] " Tom Tromey
1 sibling, 0 replies; 3+ messages in thread
From: Tom de Vries @ 2026-08-27 7:50 UTC (permalink / raw)
To: gdb-patches
On 8/11/26 10:00 AM, Tom de Vries wrote:
> The Python extension has two functions that convert exceptions from the GDB
> domain to the Python domain and vice versa: gdbpy_convert_exception and
> gdbpy_handle_exception.
>
> Each function has an exception though, where it calls quit_force:
> - for gdbpy_convert_exception, when handling gdb_exception_forced_quit, and
> - for gdbpy_handle_exception, when handling SystemExit.
>
> I think it's a good idea to document these exceptions in more detail, and
> since AFAICT the reasoning is the same, I've added the same comment in both
> places.
Ping.
Thanks,
- Tom
> ---
> gdb/python/py-utils.c | 14 +++++++++++++-
> 1 file changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/gdb/python/py-utils.c b/gdb/python/py-utils.c
> index a831b96eb38..1c68b206601 100644
> --- a/gdb/python/py-utils.c
> +++ b/gdb/python/py-utils.c
> @@ -221,7 +221,14 @@ gdbpy_convert_exception (const struct gdb_exception &exception)
> if (exception.reason == RETURN_QUIT)
> exc_class = PyExc_KeyboardInterrupt;
> else if (exception.reason == RETURN_FORCED_QUIT)
> - quit_force (NULL, 0);
> + {
> + /* Ideally we'd like to propagate the exception to top-level, where
> + we'll call quit_force. Propagating the exception would ensure that
> + appropriate cleanups are run.
> + OTOH, there is the risk that the exception will be swallowed.
> + For now, don't take this risk and immediately call quit_force. */
> + quit_force (NULL, 0);
> + }
> else if (exception.error == MEMORY_ERROR)
> exc_class = gdbpy_gdb_memory_error;
> else
> @@ -485,6 +492,11 @@ gdbpy_handle_exception ()
> exit_arg = 1;
> }
>
> + /* Ideally we'd like to propagate the exception to top-level, where
> + we'll call quit_force. Propagating the exception would ensure that
> + appropriate cleanups are run.
> + OTOH, there is the risk that the exception will be swallowed.
> + For now, don't take this risk and immediately call quit_force. */
> quit_force (&exit_arg, 0);
> }
> else if (! fetched_error.type_matches (gdbpy_gdberror_exc)
>
> base-commit: 8e6c4f526a68b796b7bd71d53e4dc459ac56c3ef
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH] [gdb/python] Document quit_force uses
2026-08-11 8:00 [PATCH] [gdb/python] Document quit_force uses Tom de Vries
2026-08-27 7:50 ` [PING][PATCH] " Tom de Vries
@ 2026-08-27 15:56 ` Tom Tromey
1 sibling, 0 replies; 3+ messages in thread
From: Tom Tromey @ 2026-08-27 15:56 UTC (permalink / raw)
To: Tom de Vries; +Cc: gdb-patches
>>>>> "Tom" == Tom de Vries <tdevries@suse.de> writes:
Tom> The Python extension has two functions that convert exceptions from the GDB
Tom> domain to the Python domain and vice versa: gdbpy_convert_exception and
Tom> gdbpy_handle_exception.
Tom> Each function has an exception though, where it calls quit_force:
Tom> - for gdbpy_convert_exception, when handling gdb_exception_forced_quit, and
Tom> - for gdbpy_handle_exception, when handling SystemExit.
Tom> I think it's a good idea to document these exceptions in more detail, and
Tom> since AFAICT the reasoning is the same, I've added the same comment in both
Tom> places.
Ok.
Approved-By: Tom Tromey <tom@tromey.com>
Tom
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-08-27 15:57 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-08-11 8:00 [PATCH] [gdb/python] Document quit_force uses Tom de Vries
2026-08-27 7:50 ` [PING][PATCH] " Tom de Vries
2026-08-27 15:56 ` [PATCH] " Tom Tromey
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox