From: Markus Armbruster <armbru@redhat.com>
To: qemu-devel@nongnu.org
Cc: jmarcin@redhat.com, peterx@redhat.com,
marcandre.lureau@redhat.com, farosas@suse.de,
berrange@redhat.com, vsementsov@yandex-team.ru,
mail@maciej.szmigiero.name, clg@redhat.com,
peter.maydell@linaro.org
Subject: [PATCH 2.5/6] error: Poison g_autoptr(Error) to prevent its use
Date: Thu, 27 Nov 2025 08:10:28 +0100 [thread overview]
Message-ID: <20251127071028.2745789-1-armbru@redhat.com> (raw)
In-Reply-To: <20251125204648.857018-3-peterx@redhat.com>
The previous commit reverted support for g_autoptr(Error). This one
should stop it from coming back.
Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
---
include/qapi/error.h | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
diff --git a/include/qapi/error.h b/include/qapi/error.h
index f3ce4a4a2d..2356b84bb3 100644
--- a/include/qapi/error.h
+++ b/include/qapi/error.h
@@ -437,6 +437,26 @@ Error *error_copy(const Error *err);
*/
void error_free(Error *err);
+/*
+ * Poison g_autoptr(Error) to prevent its use.
+ *
+ * Functions that report or propagate an error take ownership of the
+ * Error object. Explicit error_free() is needed when you handle an
+ * error in some other way. This is rare.
+ *
+ * g_autoptr(Error) would call error_free() automatically on return.
+ * To avoid a double-free, we'd have to manually clear the pointer
+ * every time we propagate or report.
+ *
+ * Thus, g_autoptr(Error) would make the rare case easier to get right
+ * (less prone to leaks), and the common case easier to get wrong
+ * (more prone to double-free).
+ */
+extern void
+__attribute__((error("Do not use g_autoptr() to declare Error * variables")))
+error_free_poisoned(Error *err);
+G_DEFINE_AUTOPTR_CLEANUP_FUNC(Error, error_free_poisoned)
+
/*
* Convenience function to assert that *@errp is set, then silently free it.
*/
--
2.49.0
next prev parent reply other threads:[~2025-11-27 7:17 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-25 20:46 [PATCH for-11.0 0/6] migration: Error reporting cleanups Peter Xu
2025-11-25 20:46 ` [PATCH for-11.0 1/6] migration: Use explicit error_free() instead of g_autoptr Peter Xu
2025-11-26 6:43 ` Markus Armbruster
2025-11-25 20:46 ` [PATCH for-11.0 2/6] Revert "error: define g_autoptr() cleanup function for the Error type" Peter Xu
2025-11-26 6:45 ` Markus Armbruster
2025-11-26 7:43 ` Cédric Le Goater
2025-11-26 8:08 ` Markus Armbruster
2025-11-26 14:34 ` [PATCH 2.5/6] error: Explain why we don't g_autoptr(Error) Markus Armbruster
2025-11-26 15:14 ` Cédric Le Goater
2025-11-26 16:18 ` Peter Xu
2025-11-27 7:01 ` Markus Armbruster
2025-11-26 20:21 ` [PATCH for-11.0 2/6] Revert "error: define g_autoptr() cleanup function for the Error type" Maciej S. Szmigiero
2025-11-27 7:10 ` Markus Armbruster [this message]
2025-11-27 15:34 ` [PATCH 2.5/6] error: Poison g_autoptr(Error) to prevent its use Cédric Le Goater
2025-11-25 20:46 ` [PATCH for-11.0 3/6] migration: Make migration_connect_set_error() own the error Peter Xu
2025-11-26 7:27 ` Markus Armbruster
2025-11-28 16:58 ` Peter Xu
2025-11-25 20:46 ` [PATCH for-11.0 4/6] migration: Make multifd_send_set_error() " Peter Xu
2025-11-26 7:30 ` Markus Armbruster
2025-11-26 7:32 ` Markus Armbruster
2025-11-25 20:46 ` [PATCH for-11.0 5/6] migration: Make multifd_recv_terminate_threads() " Peter Xu
2025-11-25 20:46 ` [PATCH for-11.0 6/6] migration: Replace migrate_set_error() with migrate_error_propagate() Peter Xu
2025-11-26 7:57 ` Markus Armbruster
2025-12-01 17:32 ` Peter Xu
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=20251127071028.2745789-1-armbru@redhat.com \
--to=armbru@redhat.com \
--cc=berrange@redhat.com \
--cc=clg@redhat.com \
--cc=farosas@suse.de \
--cc=jmarcin@redhat.com \
--cc=mail@maciej.szmigiero.name \
--cc=marcandre.lureau@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=peterx@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=vsementsov@yandex-team.ru \
/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;
as well as URLs for NNTP newsgroup(s).