Comment # 2 on bug 1017824 from Takashi Iwai

Hrm, it's hard to guess what went wrong.  Basically there haven't been so much
changes.  PA didn't get updates recently.  ALSA got an update, and this might
be the case.  Or, it might be kernel.  But I can't imagine how the kernel
changes result in this behavior change...  If it's a kernel regression, it'd be
great if you can know which kernel started regression.

Also as an easy test: could you try to downgrade libasound2.rpm to the version
in openSUSE:Leap:42.2 (not *:Update) to see whether you still see the issue?