https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=287813
--- Comment #13 from J.R. Oldroyd <f...@opal.com> --- (In reply to Christos Margiolis from comment #10) Christos, I think you've misunderstood what I said and have it backwards. YOU said that the close code was broken before the commit. I am agreeing - and saying that with the broken close code, it possibly isn't closing properly so therefore settings are being retained and the device works. With the commit, the close code is now fixed so now FreeBSD works like Linux and the settings are lost and the device no longer works. Tomasz, I have played with lots of sysctls and even wrote a C program to do the same sequence of ioctls to the device that we can see in the "pactl exit" truss trace and also attempted some of the device writes that we see in the usbdump. But I have not been able to duplicate the effect of running "pactl exit". I will try the firmware update soon and report back. If my assumption about the settings being retained with the pre-D50488 close code is wrong, what would be the explanation for why running and exiting the pulseaudio daemon makes it work? -- You are receiving this mail because: You are the assignee for the bug.
