On Wed, 2018-08-15 at 19:00 -0700, Linus Torvalds wrote: > On Wed, Aug 15, 2018 at 6:27 PM Linus Torvalds > <torva...@linux-foundation.org> wrote: > > > > I have no idea _why_, but it's consistent, and it bisects down to > > > > c647f806b8c2 "ALSA: hda - Allow multiple ADCs for mic mute LED controls" > > > > where the previous commit works fine, but that commit definitely > > causes the problem. > > Hmm. > > That commit changes the return value of the > snd_hda_gen_add_micmute_led(), but doesn't change any of the callers. > > Now, almost no caller actually checks the return value. > > Almost. > > There is *one* caller that does, though: > > sound/pci/hda/dell_wmi_helper.c: > alc_fixup_dell_wmi() > > And yes, the laptop that breaks is a Dell. An XPS13. > > I bet that instead of reverting, I can just change that > > (snd_hda_gen_add_micmute_led(codec, > dell_micmute_update) <= > 0); > > to test for "< 0" instead of "<= 0".
Why does the commit change snd_hda_gen_add_micmute_led to return 0; instead of return 1; That bit looks wrong.
