<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - It's not safe to share wl_display fd with other threads." href="https://bugs.freedesktop.org/show_bug.cgi?id=91273#c6">Comment # 6</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - It's not safe to share wl_display fd with other threads." href="https://bugs.freedesktop.org/show_bug.cgi?id=91273">bug 91273</a> from <a class="email" href="mailto:ppaalanen@gmail.com" title="Pekka Paalanen <ppaalanen@gmail.com>"> Pekka Paalanen</a> <pre>All libraries, including Mesa, need to be fixed to use the prepare_read/read_events API, because that is the only API we even think as thread-safe. I suppose it could be nice to file bugs per library, and add them here as blockers (or just references in case the library does not use fd.o bugzilla). Lazily written application can use the non-safe API just fine, when they are not threaded and assuming the libraries they use do not use the same wl_display object from other threads. The non-safe API was there first, so apps that don't require thread-safety have not been ported. Mixing non-safe and safe APIs is only dangerous if their use may overlap. If you only ever have a single thread, I believe these can be mixed safely if paying attention to the details. As for sharing the wl_display *file descriptor*, that is never intended to work. I mean, you cannot create multiple wl_displays from the same fd and expect that to work. I don't think anyone even tries to, just clarifying. Jonas' "proxy wrapper" is an attempt to solve one more threading issue. I still don't understand what you mean by "the kernel may not wake up the thread" if it refers to a different issue than the two mentioned above.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> </ul> </body> </html>