https://bugs.freedesktop.org/show_bug.cgi?id=75303
--- Comment #7 from Pekka Paalanen <[email protected]> --- (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. Case A: client messagebuffer server surface1.attach(B) -> surface1.commit -> -> surface1.attach(B) -> surface1.commit surface2.attach(B) -> surface2.commit -> -> surface2.attach(B) -> surface2.commit repaint, GL texture damage flush <- B.release B.release <- B is free. Case B: surface1.attach(B) -> surface1.commit -> -> surface1.attach(B) -> surface1.commit repaint, GL texture damage flush <- B.release surface2.attach(B) -> surface2.commit -> -> surface2.attach(B) -> surface2.commit B.release <- B is busy on surface2. >From the client point of view, cases A and B are equivalent, yet the buffer is free in one and busy in the other. -- 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
