https://bugs.freedesktop.org/show_bug.cgi?id=77595
--- Comment #4 from Arun Raghavan <[email protected]> --- David, if I understand correctly, your concern is that someone not using libpulse could emulate our protocol and cause a server crash? There are probably a few more instances where this can happen, not the least of which is writing past the SHM boundary which I believe can cause a SIGBUS. One thing I can think of is that if we find such a situation, the assert could be converted into an "exception" that causes the sink input to be killed. I would say that this would need to be part of a larger effort to eliminate ways in which malicious clients could kill the daemon (which would need to include a definition of our trust model, etc.). In the mean time, I think we should unblock the release by patching both sides as you suggest. Do you want to write up patches (if you agree)? -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the bug.
_______________________________________________ pulseaudio-bugs mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/pulseaudio-bugs
