[alsa-devel] "No codecs found" on Intel HDA + AD1988A

[Takashi Iwai]

At Tue, 16 Dec 2008 01:28:04 +0300,
Alexander Chemeris wrote:
> 
> Sorry for mail burst...
> 
> On Sun, Dec 14, 2008 at 12:17 PM, Takashi Iwai <tiwai at suse.de> wrote:
> > At Sat, 13 Dec 2008 21:57:04 +0300,
> > Alexander Chemeris wrote:
> > Anyway, try to build with --debug=verbose configure option.  Usually
> > you'll have a bit more messages.
> > At least, you'll get a message like "codec_mask = 0xxxx" at probing.
> > If the value is zero, it means that your hardware doesn't set the
> > codec slot bit, usually a BIOS issue.
> 
> If I manually set codec_mask to 0xf I get following output:
> 
> [  634.273106] HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level,
> low) -> IRQ 22
> [  634.273133] HDA Intel 0000:00:1b.0: setting latency timer to 64
> [  634.273138] ALSA alsa-kernel/pci/hda/hda_intel.c:2216: chipset
> global capabilities = 0x4401
> [  634.304568] ALSA alsa-kernel/pci/hda/hda_intel.c:786: Overriding
> codecs probe mask
> [  634.304574] ALSA alsa-kernel/pci/hda/hda_intel.c:793: codec_mask = 0xf
> [  635.308511] ALSA alsa-kernel/pci/hda/hda_intel.c:633: hda_intel:
> azx_get_response timeout, switching to polling mode: last
> cmd=0x000f0000
> [  636.313617] ALSA alsa-kernel/pci/hda/hda_intel.c:1277: hda_intel:
> Codec #0 probe error; disabling it...
> [  637.348012] ALSA alsa-kernel/pci/hda/hda_intel.c:1277: hda_intel:
> Codec #1 probe error; disabling it...
> [  638.384508] ALSA alsa-kernel/pci/hda/hda_intel.c:1277: hda_intel:
> Codec #2 probe error; disabling it...
> [  639.420508] ALSA alsa-kernel/pci/hda/hda_intel.c:1277: hda_intel:
> Codec #3 probe error; disabling it...
> [  639.452023] ALSA alsa-kernel/pci/hda/hda_intel.c:786: Overriding
> codecs probe mask
> [  639.452029] ALSA alsa-kernel/pci/hda/hda_intel.c:793: codec_mask = 0xf
> [  640.456014] ALSA alsa-kernel/pci/hda/hda_intel.c:648: hda_intel:
> azx_get_response timeout, switching to single_cmd mode: last
> cmd=0x000f0000
> [  640.457380] ALSA alsa-kernel/pci/hda/hda_intel.c:706: hda-intel:
> get_response timeout: IRS=0x1
> [  640.458587] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0xf0001
> [  640.459641] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0xf0002
> [  640.460709] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0xf0004
> [  640.460716] ALSA
> /home/ipse/packages/alsa-git/alsa-driver/pci/hda/hda_codec.c:783:
> hda_codec: no AFG or MFG node found
> [  640.462509] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0x100f0000
> [  640.463624] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0x100f0000
> [  640.464731] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0x100f0001
> [  640.465795] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0x100f0002
> [  640.466858] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0x100f0004
> [  640.466864] ALSA
> /home/ipse/packages/alsa-git/alsa-driver/pci/hda/hda_codec.c:783:
> hda_codec: no AFG or MFG node found
> [  640.468730] ALSA alsa-kernel/pci/hda/hda_intel.c:688: hda-intel:
> send_cmd timeout: IRS=0x1, val=0x200f0000
> [  640.470038] ALSA
> /home/ipse/packages/alsa-git/alsa-driver/pci/hda/hda_codec.c:783:
> hda_codec: no AFG or MFG node found
> [  640.471210] ALSA
> /home/ipse/packages/alsa-git/alsa-driver/pci/hda/hda_codec.c:783:
> hda_codec: no AFG or MFG node found
> [  640.471217] ALSA alsa-kernel/pci/hda/hda_intel.c:1303: hda-intel:
> no codecs initialized
> [  640.472915] HDA Intel 0000:00:1b.0: PCI INT A disabled
> 
> So it looks  like codec is not present or is not recognized...
> Is it possible, or my hardware looks like broken?

The probing control might not be perfect.  At least you can try to
enable only one slot at each probe, e.g. probe_mask=1, probe_mask=2,
probe_mask=4, and probe_mask=8.

If all of these don't work, it should be really a deeper issue.


Takashi