Bug#660111: [3.0 -> 3.1.8 regression] intel ibexpeak hdmi no sound out

[Jonathan Nieder <jrnieder@gmail.com>]

Andres Cimmarusti wrote:

> I installed two kernels from snapshots. Using date 2012-02-01 I was
> able to pull in a 3.1.8 kernel that exhibited the same problem as
> kernels 3.2.x.
> I also used data 20111201 and pulled in a debian 3.0.x kernel. This
> one didn't have the problem (as in the case of my own 3.0.20).

Thanks, that's very useful.

Perhaps v3.1-rc2~4^2~47^2~89 (ALSA: hda: HDMI: Support codecs with
fewer cvts than pins, 2011-07-01) is the problematic patch.  Can you
check this guess, like so?

0. Prerequisites:
	apt-get install git build-essential

1. Get a copy of the kernel:
	git clone \
	  git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
	cd linux

   Or, if you already have a copy, update it:
	cd linux
	git fetch origin

2. Try the suspect commit:
	git checkout v3.1-rc2~4^2~47^2~89
	cp /boot/config-$(uname -r) .config; # current configuration
	make localmodconfig; # optional: minimize configuration
	make deb-pkg; # optionally with -j<num> for parallel build
	dpkg -i ../<name of package>
	reboot
	aplay -l

3. Try without it:
	cd linux
	git reset --keep HEAD^; # checking out the immediately previous version
	make deb-pkg; # maybe with -j4
	dpkg -i ../<name of package>
	reboot
	aplay -l

If the suspect commit reproduces the problem and its parent does not,
please contact Stephen Warren <swarren@nvidia.com> for advice, cc-ing
Takashi Iwai <tiwai@suse.de>, alsa-devel@alsa-project.org, and either
me or this bug log so we can track it.  Be sure to mention:

 - steps to reproduce the problem, expected result, actual result, and
   how the difference indicates a bug
 - which kernel versions you have tested and what happened with each
 - where one can find your alsa-info.sh output with a problematic and
   unproblematic kernel (e.g., attaching them should be fine)
 - what kinds of tests you would be able to carry out (e.g., are you
   able to test patches for debugging if provided?)
 - any other weird observations or workarounds
 - a link to http://bugs.debian.org/660111 for context

If my guess is wrong, we can try using "git bisect" for a more
systematic search (see the git-bisect(1) manpage for details).

Hope that helps,
Jonathan