Summary: | Transport's reference count is not thread-safe in live_messages_notify() | ||
---|---|---|---|
Product: | dbus | Reporter: | Adrian Szyndela <adrian.s> |
Component: | core | Assignee: | D-Bus Maintainers <dbus> |
Status: | RESOLVED FIXED | QA Contact: | D-Bus Maintainers <dbus> |
Severity: | normal | ||
Priority: | medium | CC: | ralf.habacker |
Version: | git master | ||
Hardware: | ARM | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Attachments: | A patch that extends lock's range in live_messages_notify(). |
Description
Adrian Szyndela
2015-05-05 10:06:55 UTC
Comment on attachment 115540 [details] [review] A patch that extends lock's range in live_messages_notify(). Review of attachment 115540 [details] [review]: ----------------------------------------------------------------- This makes sense. Not your fault, but the circular dependency between the transport and the connection seems rather crazy: I don't know why we have these as separate layers if we aren't going to layer them properly. But the people who could answer that are no longer involved with D-Bus :-( (In reply to Simon McVittie from comment #1) > Review of attachment 115540 [details] [review] [review]: > ----------------------------------------------------------------- > > This makes sense. Is this patch intended for master only or also for dbus-1.8 ? Fixed in git for 1.8.18 and 1.9.16. (In reply to Ralf Habacker from comment #2) > Is this patch intended for master only or also for dbus-1.8 ? The surrounding code is the same either way, and it fixes a real bug (lack of thread-safety) in production code, so it seems worthwhile for 1.8 too. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.