I filed this bug against the debian kernel package a while ago: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=531090 Things have changed a bit since then. It seems to depend on the kind of suspend resume mechanism I use, if I can use the soundcard after resume or not. With s2disk/uswsusp it won't work at all, with BIOS suspend/resume (hibernate), I can unload/reload the snd modules to get sound. In any case, directly after resume any application accessing the sound devices will hang, but is killable with SIGKILL (so it's probably not IO-wait, unless things have changed since I last had an application in uninterruptible sleep). There's more, hopefully relevant, information in the debian bug. I'd be willing, if necessary, to try with a vanilla kernel.