GitHub/MotorolaMobilityLLC/kernel-slsi.git
13 years agoALSA: hda - More volume-init fixes for ALC267 codec
Takashi Iwai [Mon, 27 Jun 2011 14:17:07 +0000 (16:17 +0200)]
ALSA: hda - More volume-init fixes for ALC267 codec

More similar fixes like previous commits: handle the exceptional case
like ALC267 where no volume amp is found in ADC widget but in the
capsrc widget instead.

Also minor checks for avoiding possible erros: no connection-select
when the pin has a single selection, and add beep verbs only when the
0x1d is used for beep.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix volume-init for ALC259 with invalid widget caps
Takashi Iwai [Mon, 27 Jun 2011 13:53:38 +0000 (15:53 +0200)]
ALSA: hda - Fix volume-init for ALC259 with invalid widget caps

ALC259 seems to provide an invalid widget capability for the input-src
selector widget.  The widget shows the input-amp while it's a selector,
and this confuses the current ALC882 initialization code that is used
for ALC259, too.  For fixing this, check the amp capability and handle
the connection selection individually.

Also, ALC259 has no mute bit in DAC volume, so we need to initialize
it as ZERO instead of MUTE.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix volume-init of ALC299 & co
Takashi Iwai [Mon, 27 Jun 2011 13:48:17 +0000 (15:48 +0200)]
ALSA: hda - Fix volume-init of ALC299 & co

ALC269 and compatible codecs have the output volume in DACs, thus we
can't use the ALC880's code as is.  Fixed by checking the amp caps and
picking up the right widget for initialization.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Simplify EAPD control in patch_realtek.c
Takashi Iwai [Mon, 27 Jun 2011 13:28:57 +0000 (15:28 +0200)]
ALSA: hda - Simplify EAPD control in patch_realtek.c

Look through the known NIDs that may have EAPD capabilities and turn
on/off them appropriately instead of checking the individual vendor ids.

This will also avoid the forgotten entries of newly added codec ids
in future.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix auto-init of output volumes of Realtek codecs
Takashi Iwai [Mon, 27 Jun 2011 13:00:48 +0000 (15:00 +0200)]
ALSA: hda - Fix auto-init of output volumes of Realtek codecs

Fix the regression introduced by the commit
1f0f4b8036b1fe1347cb4f1f199601b87de9be46
  ALSA: hda - Reduce static init verbs for Realtek auto-parsers

The input amps of mixer widgets should be unmuted as default (as
usually they have no assigned mixer switches).

More fixes in this commit are, however, for ALC260: ALC260 codec can
have multiple output mixers connnected to a single DAC althouh the
driver didn't pick up them properly.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoMerge branch 'topic/via-cleanup' into topic/hda
Takashi Iwai [Mon, 27 Jun 2011 12:32:50 +0000 (14:32 +0200)]
Merge branch 'topic/via-cleanup' into topic/hda

13 years agoALSA: hda - Check hard-wired DACs at first for ALC662 & co
Takashi Iwai [Mon, 27 Jun 2011 10:34:01 +0000 (12:34 +0200)]
ALSA: hda - Check hard-wired DACs at first for ALC662 & co

Some Realtek codecs have the output pins hardwired with certain DACs.
These DACs have to be assigned at first and assign the rest for
multi-DAC pins so that all DACs can be assigned properly.

Without such an optimization, speaker outputs may be assigned to the
same DAC as the headphone or others.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Call proper DAC-filler function for Realtek auto-parser
Takashi Iwai [Mon, 27 Jun 2011 09:32:07 +0000 (11:32 +0200)]
ALSA: hda - Call proper DAC-filler function for Realtek auto-parser

In alc_auto_add_multi_channel_mode(), when the primary HP workaround
is enabled, it re-initializes the DAC list but calls alc662's function
in a fixed way.  This isn't pretty suitable for other codecs, of course.

Now we call it with fill_dac function pointer so that the proper
function can be called at that point.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Reduce static init verbs for Realtek auto-parsers
Takashi Iwai [Mon, 27 Jun 2011 08:52:59 +0000 (10:52 +0200)]
ALSA: hda - Reduce static init verbs for Realtek auto-parsers

Instead of using fixed init verbs, initialize DACs, ADCs and mixers
more dynamically for Realtek auto-parsers.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Add snd_hda_get_conn_list() helper function
Takashi Iwai [Fri, 24 Jun 2011 12:10:28 +0000 (14:10 +0200)]
ALSA: hda - Add snd_hda_get_conn_list() helper function

Add a new helper function snd_hda_get_conn_list().
Unlike snd_hda_get_connections(), this function doesn't copy the
connection-list but gives the raw pointer for the cached list.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Use alc_get_pfx_name() for all Realtek codecs
Takashi Iwai [Fri, 24 Jun 2011 09:27:22 +0000 (11:27 +0200)]
ALSA: hda - Use alc_get_pfx_name() for all Realtek codecs

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Clean up multi-channel mixer name assignment in patch_realtek.c
Takashi Iwai [Fri, 24 Jun 2011 09:03:58 +0000 (11:03 +0200)]
ALSA: hda - Clean up multi-channel mixer name assignment in patch_realtek.c

Change alc_get_line_out_pfx() in patch_realtek.c to provide the channel
specific name and assign the index so that each caller doesn't have to
set the channel name by itself.

Also, check the multi-io case with the primary hp-out; for the multi-io
channels, assign the channel name instead of "Headphone" with indices.
This makes the mixer names more intuitive and reduces confusion.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Add a workaround for invalid line-out setups
Takashi Iwai [Fri, 24 Jun 2011 08:43:03 +0000 (10:43 +0200)]
ALSA: hda - Add a workaround for invalid line-out setups

Some BIOS set up the pin config wrongly as line-out although it's
supposed to be a speaker out.  In most cases, though, we can judge
the validity by checking the connection type -- when it's FIXED,
mostly it's an invalid line-out but a speaker.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoMerge branch 'fix/hda' into topic/hda
Takashi Iwai [Fri, 24 Jun 2011 08:36:25 +0000 (10:36 +0200)]
Merge branch 'fix/hda' into topic/hda

13 years agoALSA: hda - Allow multi-io with HP output for ALC662 & co
Takashi Iwai [Fri, 24 Jun 2011 08:35:05 +0000 (10:35 +0200)]
ALSA: hda - Allow multi-io with HP output for ALC662 & co

Even if the machine has no line-out but only HP-out, try to detect the
multi-io.  It'll allow more possibilities for 5.1 outputs on laptops.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: HDA: Pinfix quirk for HP Z200 Workstation
David Henningsson [Wed, 22 Jun 2011 07:58:37 +0000 (09:58 +0200)]
ALSA: HDA: Pinfix quirk for HP Z200 Workstation

BIOS lists the internal speaker as an internal line-out. Change to
internal speaker + model=auto for better auto-mute capabilities.

BugLink: http://bugs.launchpad.net/bugs/754964
Reported-by: Marc Legris <marc.legris@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Implement dynamic-ADC switching for VIA codecs
Takashi Iwai [Wed, 22 Jun 2011 13:23:25 +0000 (15:23 +0200)]
ALSA: hda - Implement dynamic-ADC switching for VIA codecs

Some VIA codecs like VT1702 provide the input-route only to specific
ADCs such as digital-mic inputs.  These routes aren't covered by the
normal primary ADC, and for now, user had to open the capture stream
assigned to that special ADC manually for using such inputs.

This patch implements a way to switch the current ADC dynamically per
the input-source selection in such a case.  When this workaround is
activated, the driver provides only one capture stream and one input-
source control but with the full possible inputs.  The driver switches
the ADC to be used (or being used) according to the input-source on the
fly.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Auto-mute smart51 surround pins for VIA codecs
Takashi Iwai [Tue, 21 Jun 2011 14:52:39 +0000 (16:52 +0200)]
ALSA: hda - Auto-mute smart51 surround pins for VIA codecs

When smart51 mode is enabled, auto-mute these surround outputs
as well as the primary line-out.  Also this patch includes minor
clean-ups.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Remove redundant VT1709 and VT1708B codes
Takashi Iwai [Tue, 21 Jun 2011 14:33:55 +0000 (16:33 +0200)]
ALSA: hda - Remove redundant VT1709 and VT1708B codes

Unify the VT1709 10ch and 6ch parsers, as well as VT1708B 8ch and 4ch
parsers.  They have no difference now.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - VT1708 independent HP routing fix
Takashi Iwai [Tue, 21 Jun 2011 13:57:44 +0000 (15:57 +0200)]
ALSA: hda - VT1708 independent HP routing fix

The codecs like VT1708 needs more complicated routing using the mixer
widget rather than the simple selector widgets.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix surround-volume parsing for VT1708B codecs
Takashi Iwai [Tue, 21 Jun 2011 12:22:14 +0000 (14:22 +0200)]
ALSA: hda - Fix surround-volume parsing for VT1708B codecs

The surround/CLFE/side DACs on VT1708B and co have no amp but the
connected selector widgets have the amp instead.  Fix the parser to
check these selector widgets for the possible mixer controls as well.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix the check of loopback-mixer element index in patch_via.c
Takashi Iwai [Tue, 21 Jun 2011 10:57:22 +0000 (12:57 +0200)]
ALSA: hda - Fix the check of loopback-mixer element index in patch_via.c

Fix the check of the multiple loopback-mixer, which gave sometimes
a wrong index assigned to an element even for different names, e.g.
Mic and Front Mic.  Now check the label properly for avoid duplication.

Reported-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Assign smart51 only in the same stack for VIA codecs
Takashi Iwai [Tue, 21 Jun 2011 10:51:33 +0000 (12:51 +0200)]
ALSA: hda - Assign smart51 only in the same stack for VIA codecs

The input jacks assigned as the smart51 outputs must be in the same
stack, either rear, front or other.  Also, prefer line-in as the surround
to mic-in.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix re-routing of HP-independent mode in patch_via.c
Takashi Iwai [Tue, 21 Jun 2011 09:48:29 +0000 (11:48 +0200)]
ALSA: hda - Fix re-routing of HP-independent mode in patch_via.c

Re-route the whole output path when HP-independent mode is changed.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix creations of playback volume controls in patch_via.c
Lydia Wang [Tue, 21 Jun 2011 08:11:11 +0000 (16:11 +0800)]
ALSA: hda - Fix creations of playback volume controls in patch_via.c

Fix a issue to create playback volume control if pin has amplifier capability
but not DAC.

Signed-off-by: Lydia Wang <lydiawang@viatech.com.cn>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Revisit output_path parsing in patch_via.c
Takashi Iwai [Tue, 21 Jun 2011 07:01:36 +0000 (09:01 +0200)]
ALSA: hda - Revisit output_path parsing in patch_via.c

Change the order of the output-path list in a way from the DAC to the
target pin.  Also now the list include the target pin, too.

Together with this format change, simplify the arguments of
parse_output_path() function, and fix the initialization in
via_auto_init_output().

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Use xxx Boost Volume for VIA
Takashi Iwai [Tue, 21 Jun 2011 06:37:41 +0000 (08:37 +0200)]
ALSA: hda - Use xxx Boost Volume for VIA

Drop "Capture" prefix from the mic-boost names.
Otherwise some control names can overflow the max name length.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix a compile error in patch_ca0132.c for the recent SPDIF change
Takashi Iwai [Tue, 21 Jun 2011 05:44:51 +0000 (07:44 +0200)]
ALSA: hda - Fix a compile error in patch_ca0132.c for the recent SPDIF change

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoMerge branch 'fix/hda' into topic/via-cleanup
Takashi Iwai [Tue, 21 Jun 2011 05:34:00 +0000 (07:34 +0200)]
Merge branch 'fix/hda' into topic/via-cleanup

Conflicts:
sound/pci/hda/patch_via.c

13 years agoALSA: hda - Add Creative CA0132 HDA codec support
Ian Minett [Wed, 15 Jun 2011 22:35:17 +0000 (15:35 -0700)]
ALSA: hda - Add Creative CA0132 HDA codec support

Create patch_ca0132.c, to add support for devices featuring the
Creative CA0132 HD-audio codec.

This driver implements :-
* 1 playback subdevice to headphone and speaker
* 2 capture subdevices:
   i - Mic-in
   ii- Line-in
* mixer device

Advanced DSP features are not yet included.
Developed and maintained by Creative Labs, Inc.

Signed-off-by: Ian Minett <ian_minett@creativelabs.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: VIA HDA: Create a master amplifier control for VT1718S.
Lydia Wang [Mon, 20 Jun 2011 06:17:56 +0000 (14:17 +0800)]
ALSA: VIA HDA: Create a master amplifier control for VT1718S.

Create a master volume and mute control of playback for VT1718S.

Signed-off-by: Lydia Wang <lydiawang@viatech.com.cn>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: VIA HDA: Mute/unmute mixer conncted to Headphone for VT1718S.
Lydia Wang [Mon, 20 Jun 2011 06:16:33 +0000 (14:16 +0800)]
ALSA: VIA HDA: Mute/unmute mixer conncted to Headphone for VT1718S.

When switch HP independent mode, mute/unmute connctions of mixer  which is
connected to headphone for VT1718S.

Signed-off-by: Lydia Wang <lydiawang@viatech.com.cn>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: VIA HDA: Modify initial verbs list for VT1718S.
Lydia Wang [Mon, 20 Jun 2011 06:14:37 +0000 (14:14 +0800)]
ALSA: VIA HDA: Modify initial verbs list for VT1718S.

Remove some invalid initial verbs and correct some wrong initial verbs
for VT1718S codec.

Signed-off-by: Lydia Wang <lydiawang@viatech.com.cn>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Remove ALC268 model override for CPR2000
Tony Vroon [Mon, 20 Jun 2011 21:11:11 +0000 (22:11 +0100)]
ALSA: hda - Remove ALC268 model override for CPR2000

The "diverse" Quanta ID 0x0763 is overridden to ALC268_ACER.
This keeps headphone automute and microphone input from operating
on at least one laptop from Opti Systems.
Without the override, the BIOS parser does a fine job setting the
card up and everything works.

Tested-By: Peter Schneider <e.at.chi.kaen@googlemail.com>
Signed-off-by: Tony Vroon <tony@linx.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: HDA: Remove quirk for an HP device
David Henningsson [Mon, 20 Jun 2011 08:27:07 +0000 (10:27 +0200)]
ALSA: HDA: Remove quirk for an HP device

The reporter, who is running kernel 2.6.38, reports that
he needs to set model=auto for the headphone output to work
correctly.

BugLink: http://bugs.launchpad.net/bugs/761022
Cc: stable@kernel.org (v2.6.38+)
Reported-by: Jo
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Simplify analog-low-current mode check for VIA codecs
Takashi Iwai [Mon, 20 Jun 2011 13:40:19 +0000 (15:40 +0200)]
ALSA: hda - Simplify analog-low-current mode check for VIA codecs

Use the existing aa-loop list for simplifying the check for analog
low-current mode.  Also fix the stream count test for playback streams.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Remove NID_MAPPING hacks in patch_via.c
Takashi Iwai [Mon, 20 Jun 2011 13:10:28 +0000 (15:10 +0200)]
ALSA: hda - Remove NID_MAPPING hacks in patch_via.c

There is no longer virtual kmixer element for NID mapping.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Remove unused defines and struct fields in patch_via.c
Takashi Iwai [Mon, 20 Jun 2011 12:11:59 +0000 (14:11 +0200)]
ALSA: hda - Remove unused defines and struct fields in patch_via.c

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Name the primary out as Speaker when needed for VIA codecs
Takashi Iwai [Mon, 20 Jun 2011 12:09:02 +0000 (14:09 +0200)]
ALSA: hda - Name the primary out as Speaker when needed for VIA codecs

When the primary output is the speaker output, rather name it as
"Speaker".  This will be more intuitive.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Create loopback-list dynamically in patch_via.c
Takashi Iwai [Mon, 20 Jun 2011 12:05:46 +0000 (14:05 +0200)]
ALSA: hda - Create loopback-list dynamically in patch_via.c

Create loopback list dynamically from the parsed input pins for VIA
codecs instead of the fixed arrays.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix smart51 handling again
Takashi Iwai [Mon, 20 Jun 2011 11:52:33 +0000 (13:52 +0200)]
ALSA: hda - Fix smart51 handling again

Fix the broken detection of smart51 and its handling.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Create virtual-master control for VIA codecs
Takashi Iwai [Mon, 20 Jun 2011 10:47:45 +0000 (12:47 +0200)]
ALSA: hda - Create virtual-master control for VIA codecs

Now let's add the missing Master control to VIA codecs.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Initialize unsol events dynamically in patch_via.c
Takashi Iwai [Mon, 20 Jun 2011 10:39:26 +0000 (12:39 +0200)]
ALSA: hda - Initialize unsol events dynamically in patch_via.c

Issue the init verbs of unsolicited events dynamically from the parsed
results for VIA codecs.  Also, consolidate the unsol handlers for HP
and line-out mutes.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Initialize input-path dynamically in patch_via.c
Takashi Iwai [Mon, 20 Jun 2011 10:09:02 +0000 (12:09 +0200)]
ALSA: hda - Initialize input-path dynamically in patch_via.c

Similarly like the previous commit, initialize the input-paths dynamically
from the parsed results instead of the fixed array for VIA codecs.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Initialize output path dynamically in patch_via.c
Takashi Iwai [Mon, 20 Jun 2011 09:32:27 +0000 (11:32 +0200)]
ALSA: hda - Initialize output path dynamically in patch_via.c

Instead of fixed array for each codec type, initialize the output path
dynamically from the parsed results.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Replace with standard consts in patch_via.c
Takashi Iwai [Sun, 19 Jun 2011 14:27:53 +0000 (16:27 +0200)]
ALSA: hda - Replace with standard consts in patch_via.c

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix independent-HP handling in patch_via.c
Takashi Iwai [Sun, 19 Jun 2011 14:24:21 +0000 (16:24 +0200)]
ALSA: hda - Fix independent-HP handling in patch_via.c

Fix races in handling of HP DAC and independent streams for VIA codecs.
Also, allow the HP output path without front-DAC, and removed
unnecessary activation of HP mixer elements.

This also removes the handling of shared side/HP stream; it's anyway
implemented in a broken way, so we need to re-implement the feature
later...

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Unify auto-parser in patch_via.c
Takashi Iwai [Sat, 18 Jun 2011 15:45:49 +0000 (17:45 +0200)]
ALSA: hda - Unify auto-parser in patch_via.c

Now all codecs use the same parser-path, so we can reduce into a single
auto-parser function.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Return error for invalid setup for VIA
Takashi Iwai [Sat, 18 Jun 2011 15:33:34 +0000 (17:33 +0200)]
ALSA: hda - Return error for invalid setup for VIA

Instead of ignoring the invalid pin configuration, return the error.
This will avoid unexpected crash, anyway.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Unify capture-mixer creations in patch_via.c
Takashi Iwai [Sat, 18 Jun 2011 15:24:46 +0000 (17:24 +0200)]
ALSA: hda - Unify capture-mixer creations in patch_via.c

Create capture-related mixer elements dynamically from the parsed
ADCs and input-pins instead of fixed values for each codec.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Assign HP-independent PCM to individual stream
Takashi Iwai [Sat, 18 Jun 2011 14:40:14 +0000 (16:40 +0200)]
ALSA: hda - Assign HP-independent PCM to individual stream

Instead of using the secondary substream, create an individual PCM
stream for HP-independent PCM.  Otherwise it's difficult to handle
different channel numbers with multi-channel stream in the sam PCM
stream structure.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Unify PCM assignments in patch_via.c
Takashi Iwai [Sat, 18 Jun 2011 14:17:45 +0000 (16:17 +0200)]
ALSA: hda - Unify PCM assignments in patch_via.c

Unify PCM streams for all codecs by assigning the NID dynamically.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Don't create secondary substream when no independent-hp is used
Takashi Iwai [Sat, 18 Jun 2011 08:58:49 +0000 (10:58 +0200)]
ALSA: hda - Don't create secondary substream when no independent-hp is used

For VIA codecs, we shouldn't create a substream for independent HP mode,
when no individual HP DAC is found.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Re-implement smart51 detection for VIA codecs
Takashi Iwai [Fri, 17 Jun 2011 16:46:48 +0000 (18:46 +0200)]
ALSA: hda - Re-implement smart51 detection for VIA codecs

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Unify output-control parsing in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 15:53:38 +0000 (17:53 +0200)]
ALSA: hda - Unify output-control parsing in patch_via.c

Parse the output-paths more dynamically, i.e. traverse the paths
from each output pin instead of fixed assignment for each codec.
Now all codecs are using the same output parser code.

The smart51 setup doesn't work with this change, and will be fixed
in the next commits.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Unify input-volume creations in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 15:19:19 +0000 (17:19 +0200)]
ALSA: hda - Unify input-volume creations in patch_via.c

Now storing the analog-mixer widget in spec, we can simplify the rest
parts.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Auto-mute all LO and speakers in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 14:51:39 +0000 (16:51 +0200)]
ALSA: hda - Auto-mute all LO and speakers in patch_via.c

Muting all line-outs and/or speakers is more common in other drivers,
so we should follow it, too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Change pin-ctl for auto-muting in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 14:37:45 +0000 (16:37 +0200)]
ALSA: hda - Change pin-ctl for auto-muting in patch_via.c

Mute the outputs via pin-controls instead of amps for the auto-mute
handling.  This makes our life easier as it avoids conflict of the states
between the mixer elements and the auto-mute toggles.

With this change, we can use vmaster for the master control easily now.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Generate PCM names dynamically in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 14:24:21 +0000 (16:24 +0200)]
ALSA: hda - Generate PCM names dynamically in patch_via.c

This reduces lots of static strings.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Refactor ctl array handling in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 14:15:26 +0000 (16:15 +0200)]
ALSA: hda - Refactor ctl array handling in patch_via.c

No functional change.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Defer mixer element creation to the right time in patch_via.c
Takashi Iwai [Fri, 17 Jun 2011 13:46:13 +0000 (15:46 +0200)]
ALSA: hda - Defer mixer element creation to the right time in patch_via.c

The jack-detect control should be created at the time of build_controls
callback instead of calling snd_hda_add_ctls() at the tree-parsing time.
For that, copy the control to the temporary array like other cases.

Also, fixed typos of vt1708_jack_detect in all places.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fill ADCs dynamically for VIA codecs
Takashi Iwai [Fri, 17 Jun 2011 07:01:29 +0000 (09:01 +0200)]
ALSA: hda - Fill ADCs dynamically for VIA codecs

Instead of giving the fixed ADC list, parse the widgets and fill in
ADCs dynamically.

Also, probe the stereo-mixer input more dynamically, too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Add control to suppress the dynamic pin-power for VIA
Takashi Iwai [Fri, 17 Jun 2011 14:59:21 +0000 (16:59 +0200)]
ALSA: hda - Add control to suppress the dynamic pin-power for VIA

Currently VIA driver controls the power-state of each pin per jack
detection.  But, it means that the power-state mismatch may occur when
the machine doesn't give the proper jack-detection.

For avoiding this problem, a new control element "Dynamic Power-Control"
is provided so that user can turn on/off the pin-power control.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Remove superfluous NID_MAPPING use for smart51 mixer
Takashi Iwai [Fri, 17 Jun 2011 12:55:02 +0000 (14:55 +0200)]
ALSA: hda - Remove superfluous NID_MAPPING use for smart51 mixer

Just a minor clean up; nid-mapping can be set directly to the smart51
mixer element.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: HDA: Remove redundant LPIB quirks for ATI chipset
David Henningsson [Fri, 17 Jun 2011 14:27:01 +0000 (16:27 +0200)]
ALSA: HDA: Remove redundant LPIB quirks for ATI chipset

Now that we have changed the position_fix default for ATI and AMD
to be LPIB (see commit 50e3bbf989), we can remove the quirks that
were added for ATI chipsets.

Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix no NID error with VIA codecs
Takashi Iwai [Fri, 17 Jun 2011 12:23:46 +0000 (14:23 +0200)]
ALSA: hda - Fix no NID error with VIA codecs

The via driver spews warnigs like
hda-codec: no NID for mapping control Independent HP:0:0
with some codecs because snd_hda_add_nid() is called with nid=0.
This patch fixes it by skipping the call when no corresponding widget
is found.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: isight: adjust for new queueing API
Clemens Ladisch [Fri, 17 Jun 2011 06:18:35 +0000 (08:18 +0200)]
ALSA: isight: adjust for new queueing API

Since commit 13882a82ee16 (optimize iso queueing by setting
wake only after the last packet), drivers are required to call
fw_iso_context_queue_flush() after queueing a batch of packets.
The missing call would have an effect only if the controller
queue underruns, but then the DMA would stop completely.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: 6fire - Fix signedness bug
Torsten Schenk [Thu, 16 Jun 2011 19:06:27 +0000 (21:06 +0200)]
ALSA: 6fire - Fix signedness bug

Fixed remaining issues of the signedness bug discovered by Dan Carpenter.
A check was remaining that tests if unsigned rt->rate is >= 0.
Changed that so that rt->rate now consistently uses ARRAY_SIZE(rates)
as invalid rate value and not -1.

Signed-off-by: Torsten Schenk <torsten.schenk@zoho.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: lola - Fix section mismatch
Takashi Iwai [Wed, 15 Jun 2011 13:14:49 +0000 (15:14 +0200)]
ALSA: lola - Fix section mismatch

Add missing __devinit.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix missing static inline to beep dummy function
Takashi Iwai [Tue, 14 Jun 2011 06:14:32 +0000 (08:14 +0200)]
ALSA: hda - Fix missing static inline to beep dummy function

The commit 2308f4add3de9f6c9c9f02e49461e94d84bb200a missed static inline
thus it resulted in multiple-definitions error at linking.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: 6fire: Fix double-free bug in usb6fire_fw_ezusb_upload()
Jesper Juhl [Mon, 13 Jun 2011 21:52:02 +0000 (23:52 +0200)]
ALSA: 6fire: Fix double-free bug in usb6fire_fw_ezusb_upload()

We have a double-free bug in
sound/usb/6fire/firmware.c::usb6fire_fw_ezusb_upload().
We already call release_firmware(fw) on line 258, so when we then do it
again after usb6fire_fw_ezusb_write() returns <0, we have a double-free.
Easily fixed by just removing the last call to release_firmware().

Signed-off-by: Jesper Juhl <jj@chaosbits.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda: check make_exec_verb() return value
Greg Thelen [Mon, 13 Jun 2011 14:45:45 +0000 (07:45 -0700)]
ALSA: hda: check make_exec_verb() return value

If given a -1 cmd parameter then make_exec_verb() returns -1 without
setting the res output value.

Prior to this change snd_hda_codec_read() assumed that make_exec_verb()
unconditionally set res regardless of the cmd value.

This change explicitly checks the make_exec_verb() return value before
consuming the potentially unset res value.

Signed-off-by: Greg Thelen <gthelen@google.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoMerge branch 'fix/hda' into topic/hda
Takashi Iwai [Mon, 13 Jun 2011 06:37:53 +0000 (08:37 +0200)]
Merge branch 'fix/hda' into topic/hda

13 years agoALSA: hda - Fix beep_device compilation warnings
Joe Perches [Sun, 12 Jun 2011 20:02:43 +0000 (13:02 -0700)]
ALSA: hda - Fix beep_device compilation warnings

Using static inline functions can reduce compilation messages
and macro misuse.

 sound/pci/hda/patch_conexant.c: In function ‘patch_cxt5045’:
 sound/pci/hda/patch_conexant.c:1232:3: warning: statement with no effect

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix wrong auto-mute type for Acer Aspire-one
Takashi Iwai [Mon, 13 Jun 2011 06:32:06 +0000 (08:32 +0200)]
ALSA: hda - Fix wrong auto-mute type for Acer Aspire-one

The auto-mute setup for Acer Aspire-one with ALC268 was set wrongly
during the clean-up of auto-mute function.  Fixed now.

Tested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: emu10k1: Add details for E-mu 0404 PCIe version
Florian Zeitz [Sat, 11 Jun 2011 23:15:42 +0000 (01:15 +0200)]
ALSA: emu10k1: Add details for E-mu 0404 PCIe version

This patch adds the necessary details to support the PCIe version of
E-MU's 0404 card.
From comparing the PCBs it seems the PCIe version just added a PCIe
chipset and left all other components pretty much in place.
For anyone intrigued to take a look at the PCB there are pictures I took
at <http://babelmonkeys.de/~florob/E-MU%200404/>.

Signed-off-by: Florian Zeitz <florob@babelmonkeys.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hdspm - Add firmware revision ID for RME MADI PCI version
Adrian Knoth [Sun, 12 Jun 2011 15:26:19 +0000 (17:26 +0200)]
ALSA: hdspm - Add firmware revision ID for RME MADI PCI version

The PCI version of the RME HDSP MADI card uses 0xcf as revision ID. Just
add this to the list of supported cards.

Signed-off-by: Adrian Knoth <adi@drcomp.erfurt.thur.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hdspm - Fix jumping external wordclock frequency in AutoSync mode
Adrian Knoth [Sun, 12 Jun 2011 15:26:18 +0000 (17:26 +0200)]
ALSA: hdspm - Fix jumping external wordclock frequency in AutoSync mode

When using Word Clock on RME MADI cards, AutoSync mode was alternating
betweeen MADI and WC due to a typo: AutoSync is indicated in the second
status register (status2), not the first one (status).

While the proc output was always correct, the reported WC frequency to
ALSA was unstable as mentioned in

http://mailman.alsa-project.org/pipermail/alsa-devel/2008-March/006723.html

Signed-off-by: Adrian Knoth <adi@drcomp.erfurt.thur.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hdspm - Fix locking in snd_hdspm_midi_input_read
Adrian Knoth [Sun, 12 Jun 2011 15:26:17 +0000 (17:26 +0200)]
ALSA: hdspm - Fix locking in snd_hdspm_midi_input_read

For the MIDI part, we need to acquire (and release) the hmidi->lock,
access to the global hdspm structure is serialized through
hmidi->hdspm->lock instead.

Signed-off-by: Adrian Knoth <adi@drcomp.erfurt.thur.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoMerge branch 'fix/asoc' into for-linus
Takashi Iwai [Fri, 10 Jun 2011 15:49:34 +0000 (17:49 +0200)]
Merge branch 'fix/asoc' into for-linus

13 years agoALSA: hda: Fix inaudible internal speakers on CyberpowerPC Gamer Xplorer N57001 laptop
Daniel T Chen [Fri, 10 Jun 2011 14:14:01 +0000 (10:14 -0400)]
ALSA: hda: Fix inaudible internal speakers on CyberpowerPC Gamer Xplorer N57001 laptop

BugLink: https://launchpad.net/bugs/761171
The original reporter needs the model=auto quirk for his internal
speakers to be audible in the latest daily snapshot, so add an entry in
the quirk table for his PCI SSID.

A trivially different version of this patch using the model=asus quirk
should be applied to the 2.6.38 and 2.6.39 stable kernels. We don't use
the asus quirk in 3.0-rc2, because 3.0-rc2's autoparser is much
improved.

Reported-and-tested-by: tomdeering7
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: Use %pV for snd_printk()
Takashi Iwai [Fri, 10 Jun 2011 13:32:31 +0000 (15:32 +0200)]
ALSA: Use %pV for snd_printk()

Clean up snd_printk() helper using the %pV prefix for recursive printks.
This also automagically fixes an Oops with RO/NX-enabled modules.

Tested-by: Maarten Lankhorst <m.b.lankhorst@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix initialization of hp pins with master_mute in Realtek
Takashi Iwai [Fri, 10 Jun 2011 13:28:15 +0000 (15:28 +0200)]
ALSA: hda - Fix initialization of hp pins with master_mute in Realtek

Some Reatlek model quirks use master_mute bool switch for controlling
the master-mute of outputs.  For these cases, the initialization of HP
pins/amps were forgotten during the transition to the common automute
helper function in 3.0 development time, and resulted in the muted HP
output as default.

This patch fixes the issue by adjusting the HP output explicitly with
master_mute switch.

Tested-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix invalid unsol tag for some alc262 model quirks
Takashi Iwai [Fri, 10 Jun 2011 07:31:54 +0000 (09:31 +0200)]
ALSA: hda - Fix invalid unsol tag for some alc262 model quirks

The tag number was forgotten to be fixed after cleaning up the model
quirks for ALC262 fujitsu and lenovo-3000 models.

Tested-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Fix SSYNC register value for non-Intel controllers
Takashi Iwai [Fri, 10 Jun 2011 12:56:26 +0000 (14:56 +0200)]
ALSA: hda - Fix SSYNC register value for non-Intel controllers

SSYNC register was once defined as 0x34-37 in the old Intel datasheet,
but corrected later to 0x38-3b.  For fixing the register usage, a new
bit-flag is introduced for indicating the old ICH SSYNC register, and
ICH* PCI entries are added explicitly to enable this quirk.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Disable SPDIF only when no pin config set for HP with AD1981
Takashi Iwai [Fri, 10 Jun 2011 12:37:04 +0000 (14:37 +0200)]
ALSA: hda - Disable SPDIF only when no pin config set for HP with AD1981

Some HP laptops with AD1981 have SPDIF connections, but currently the
driver disables it statically.  Better to check the pin default config
to judge whether to enable or disable the SPDIF.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoASoC: SAMSUNG: Fix the incorrect referencing of I2SCON register
Sangbeom Kim [Fri, 10 Jun 2011 01:36:54 +0000 (10:36 +0900)]
ASoC: SAMSUNG: Fix the incorrect referencing of I2SCON register

If DMA active status should be checked, I2SCON register should be referenced.
In this patch, Fix the incorrect referencing of I2SCON register.

Reported-by : Lakkyung Jung <lakkyung.jung@samsung.com>
Signed-off-by: Sangbeom Kim <sbkim73@samsung.com>
Acked-by: Jassi Brar <jassisinghbrar@gmail.com>
Acked-by: Liam Girdwood <lrg@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org
13 years agoASoC: snd_soc_new_{mixer,mux,pga} make sure to use right DAPM context
Lars-Peter Clausen [Thu, 9 Jun 2011 11:22:36 +0000 (13:22 +0200)]
ASoC: snd_soc_new_{mixer,mux,pga} make sure to use right DAPM context

Currently it is possible that snd_soc_new_{mixer,mux,pga} is called with a
DAPM context not matching the widgets context. This can lead to a wrong
prefix_len calculation, which will result in undefined behaviour. To avoid
this always use the DAPM context from the widget itself.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Liam Girdwood <lrg@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org
13 years agoASoC: fsl: fix initialization of DMA buffers
Timur Tabi [Wed, 8 Jun 2011 20:02:56 +0000 (15:02 -0500)]
ASoC: fsl: fix initialization of DMA buffers

The DMA (PCM) driver used by some Freescale PowerPC supports separate DAIs
for playback and capture, so DMA buffers should be allocated only for the
initialized streams.  Instead of checking for the number of active channels,
which apparently is not reliable, check to see if the actual stream object
exists.

Also provide a better name for the DMA interrupt.

Signed-off-by: Timur Tabi <timur@freescale.com>
Acked-by: Liam Girdwood <lrg@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
13 years agoASoC: WM8804 does not support sample rates below 32kHz
Mark Brown [Wed, 8 Jun 2011 17:07:49 +0000 (18:07 +0100)]
ASoC: WM8804 does not support sample rates below 32kHz

Reported-by: Kieran O'Leary <Kieran.O'Leary@wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
Cc: stable@kernel.org
13 years agoASoC: Fix WM8962 headphone volume update for use of advanced caches
Mark Brown [Tue, 7 Jun 2011 22:42:04 +0000 (23:42 +0100)]
ASoC: Fix WM8962 headphone volume update for use of advanced caches

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
Cc: stable@kernel.org
13 years agoALSA: hda - Judge playback stream from stream id in azx_via_get_position()
Takashi Iwai [Tue, 7 Jun 2011 10:26:56 +0000 (12:26 +0200)]
ALSA: hda - Judge playback stream from stream id in azx_via_get_position()

Instead of checking the azx_dev index with a fixed number (4), check
the stream direction of the assigned substream.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoALSA: hda - Handle -1 as invalid position, too
Takashi Iwai [Tue, 7 Jun 2011 10:23:23 +0000 (12:23 +0200)]
ALSA: hda - Handle -1 as invalid position, too

When reading from the position-buffer results in -1, handle as it's
invalid and falls back to LPIB mode as well as 0.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoASoC: Blackfin: bf5xx-ad1836: Fix codec device name
Lars-Peter Clausen [Tue, 7 Jun 2011 08:24:46 +0000 (10:24 +0200)]
ASoC: Blackfin: bf5xx-ad1836: Fix codec device name

Fix the codec_name field of the dai_link to match the actual device name
of the codec. Otherwise the card won't be instantiated.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Liam Girdwood <lrg@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org
13 years agoALSA: hda: Fix quirk for Dell Inspiron 910
Daniel T Chen [Mon, 6 Jun 2011 22:55:34 +0000 (18:55 -0400)]
ALSA: hda: Fix quirk for Dell Inspiron 910

BugLink: https://launchpad.net/bugs/792712
The original reporter states that sound from the internal speakers is
inaudible until using the model=auto quirk. This symptom is due to an
existing quirk mask for 0x102802b* that uses the model=dell quirk. To
limit the possible regressions, leave the existing quirk mask but add
a higher priority specific mask for the reporter's PCI SSID.

Reported-and-tested-by: rodni hipp
Cc: <stable@kernel.org> [2.6.38+]
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
13 years agoASoC: AD1836: Fix setting the PCM format
Lars-Peter Clausen [Mon, 6 Jun 2011 11:38:35 +0000 (13:38 +0200)]
ASoC: AD1836: Fix setting the PCM format

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Liam Girdwood <lrg@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org
13 years agoASoC: Check for NULL register bank in snd_soc_get_cache_val()
Mark Brown [Mon, 6 Jun 2011 10:26:15 +0000 (11:26 +0100)]
ASoC: Check for NULL register bank in snd_soc_get_cache_val()

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
13 years agoASoC: Add missing break in WM8915 FLL source selection
Mark Brown [Fri, 3 Jun 2011 17:20:50 +0000 (18:20 +0100)]
ASoC: Add missing break in WM8915 FLL source selection

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
13 years agoASoC: Only update SYSCLK_ENA when pausing WM8915 SYSCLK
Mark Brown [Fri, 3 Jun 2011 16:13:57 +0000 (17:13 +0100)]
ASoC: Only update SYSCLK_ENA when pausing WM8915 SYSCLK

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
13 years agoALSA: hda: HDMI: Support codecs with fewer cvts than pins
Stephen Warren [Wed, 1 Jun 2011 17:14:21 +0000 (11:14 -0600)]
ALSA: hda: HDMI: Support codecs with fewer cvts than pins

The general concept of this change is to create a PCM device for each
pin widget instead of each converter widget. Whenever a PCM is opened,
a converter is dynamically selected to drive that pin based on those
available for muxing into the pin.

The one thing this model doesn't support is a single PCM/converter
sending audio to multiple pin widgets at once.

Note that this means that a struct hda_pcm_stream's nid variable is
set to 0 except between a stream's open and cleanup calls. The dynamic
de-assignment of converters to PCMs occurs within cleanup, not close,
in order for it to co-incide with when controller stream IDs are
cleaned up from converters.

While the PCM for a pin is not open, the pin is disabled (its widget
control's PIN_OUT bit is cleared) so that if the currently routed
converter is used to drive a different PCM/pin, that audio does not
leak out over a disabled pin.

We use the recently added SPDIF virtualization feature in order to
create SPDIF controls for each pin widget instead of each converter
widget, so that state is specific to a PCM.

In order to support this, a number of more mechanical changes are made:

* s/nid/pin_nid/ or s/nid/cvt_nid/ in many places in order to make it
  clear exactly what the code is dealing with.

* We now have per_pin and per_cvt arrays in hdmi_spec to store relevant
  data. In particular, we store a converter's capabilities in the per_cvt
  entry, rather than relying on a combination of codec_pcm_pars and
  the struct hda_pcm_stream.

* ELD-related workarounds were removed from hdmi_channel_allocation
  into hdmi_instrinsic in order to simplifiy infoframe calculations and
  remove HW dependencies.

* Various functions only apply to a single pin, since there is now
  only 1 pin per PCM. For example, hdmi_setup_infoframe,
  hdmi_setup_stream.

* hdmi_add_pin and hdmi_add_cvt are more oriented at pure codec parsing
  and data retrieval, rather than determining which pins/converters
  are to be used for creating PCMs.

This is quite a large change; it may be appropriate to simply read the
result of the patch rather than the diffs. Some small parts of the change
might be separable into different patches, but I think the bulk of the
change will probably always be one large patch. Hopefully the change
isn't too opaque!

This has been tested on:

* NVIDIA GeForce 400 series discrete graphics card. This model has the
  classical 1:1:1 codec:converter:pcm widget model. Tested stereo PCM
  audio to a PC monitor that supports audio.

* NVIDIA GeForce 520 discrete graphics card. This model is the new
  1 codec n converters m pins m>n model. Tested stereo PCM audio to a
  PC monitor that supports audio.

* NVIDIA GeForce 400 series laptop graphics chip. This model has the
  classical 1:1:1 codec:converter:pcm widget model. Tested stereo PCM,
  multi-channel PCM, and AC3 pass-through to an AV receiver.

* Intel Ibex Peak laptop. This model is the new 1 codec n converters m
  pins m>n model. Tested stereo PCM, multi-channel PCM, and AC3 pass-
  through to an AV receiver.

Note that I'm not familiar at all with AC3 pass-through. Hence, I may
not have covered all possible mechanisms that are applicable here. I do
know that my receiver definitely received AC3, not decoded PCM. I tested
with mplayer's "-afm hwac3" and/or "-af lavcac3enc" options, and alsa a
WAV file that I believe has AC3 content rather than PCM.

I also tested:
* Play a stream
* Mute while playing
* Stop stream
* Play some other streams to re-assign the converter to a different
  pin, PCM, set of SPDIF controls, ... hence hopefully triggering
  cleanup for the original PCM.
* Unmute original stream while not playing
* Play a stream on the original pin/PCM.

This was to test SPDIF control virtualization.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>