On Wed, Jul 07, 2021 at 08:25:16AM -0600, Theo de Raadt wrote: > > This diff is an attempt to make identifying audio devices easier by > > giving each device a human-readable description that can be read from > > userspace (and without scraping dmesg). > > But why does anyone want that? > > Isn't everyone served best when there is a public portable interface, > and all the back-ends must try to work the same?
Exactly. The long-term aim is to expose this information to programs through sndiod with a stable libsndio interface, which in turn would ease switching between audio devices. This diff is the first step; IMO, it's not necessary until we get the whole chain, but it doesn't hurt either; it improves audioctl(8) output, which is a dignostics-only tool.
