https://bugs.freedesktop.org/show_bug.cgi?id=75303
--- Comment #8 from Jonas Ådahl <[email protected]> --- Ok, yes I see it now.(In reply to Pekka Paalanen from comment #7) > (In reply to Jonas Ådahl from comment #6) > > Why can't we just let wl_buffer.release simply be a notification about the > > buffer being reusable > > Because it has exactly the race explained in the description of this bug. > This example applies particular to Weston with gl-renderer and a client > using wl_shm. Yes, I see it now. Thanks for the more detailed explanation. (In reply to Pekka Paalanen from comment #5) > Client side reference counting, that is, for every committed attach there > will be a release, seems like the preferred solution to me at the moment. How do you expect the client to get the new semantics (one release per commit) or do you mean to change it for everyone? > > I've forgot the details related to the possibility of implementing > wl_proxy_get_version(), but from what I recall, it should be perfectly > doable since we already have runtime object version inheritance in > libwayland-client (don't we?). I don't think we do. The client needs to track this information itself. The client just passes the version to the wl_registry.bind request and the server uses that to adapt what it sends etc. We can't add it to wl_proxy because sometimes (wl_buffer, wl_sync), the wl_proxy that is created does not inherit the version of the parent proxy. Digging some, I found one field name "version" (in wl_interface), but that is just the version that was in the XML file, hard coded in the generated -protocol.c file. -- You are receiving this mail because: You are the assignee for the bug.
_______________________________________________ wayland-bugs mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/wayland-bugs
