Messages in this thread | | | Date | Thu, 04 Aug 2011 07:17:18 +0200 | From | Takashi Iwai <> | Subject | Re: Regression in 3.0-git after commit c82693db52beced0419cecf09a3c81adfe95a544 - sound does not work |
| |
At Wed, 03 Aug 2011 16:41:08 -0500, Larry Finger wrote: > > In the latest git pull of 3.0-git from Linus's tree, the sound on my HP laptop > does not work. This problem was bisected to the commit in the subject. > > The lspci output for my sound device is: > > 00:07.0 Audio device [0403]: nVidia Corporation MCP67 High Definition Audio > [10de:055c] (rev a1) > Subsystem: Hewlett-Packard Company Device [103c:30d6] > Kernel driver in use: snd_hda_intel > > I further isolated the problem to the following hunk: > > diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c > index 6e90b6b..4ca880b 100644 > --- a/sound/pci/hda/patch_conexant.c > +++ b/sound/pci/hda/patch_conexant.c > @@ -1994,10 +1990,8 @@ static int patch_cxt5051(struct hda_codec *codec) > board_config = snd_hda_check_board_config(codec, CXT5051_MODELS, > cxt5051_models, > cxt5051_cfg_tbl); > -#if 0 /* use the old method just for safety */ > if (board_config < 0) > - board_config = CXT5051_AUTO; > -#endif > + board_config = CXT5051_AUTO; /* model=auto as default */ > if (board_config == CXT5051_AUTO) > return patch_conexant_auto(codec); > > In my case, snd_hda_check_board_config() returns -1. Thus, the new code returns > the value of patch_conexant_auto(), whereas the original version fell through to > the setup code below. Clearly, this is needed for my device.
And reverting the above fixes your problem, right? Then please give alsa-info.sh outputs on both working and non-working kernels (run with --no-upload option).
thanks,
Takashi
| |