Comment 2 for bug 1890526

Currently audio/alsa_record_playback_automated and suspend/record_playback_after_suspend_auto almost always fail during SRU testing. I believe this bug is the cause of these failures. I'm not sure if there are other audio tests, or whether we are effectively unable to catch regressions related to audio playback and recording now. In case the latter is true, I would argue that a 0.55.1 plainbox-provider-checkbox point release with the fix for this bug is warranted.