The reference NID for the analog outputs of STAC/IDT codecs is set
to a fixed number 0x02.  But this isn't always correct and in many
codecs it points to a non-existing NID.

This patch fixes the initialization of the PCM reference NID taken
from the actually probed DAC list.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Some revisions of the 92hd8xxx codec's not supporting port power
downs in which the using of it causes capture and also randomly
playback streams to not function at all. Thus by disabling it by
default and adding a option to enable it manually will fix all issue
on current and future revisions.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Port 0xe power mapping was incorrect set to 0x80 changed to the correct
value 0x40.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added model=hp-dv5 for HP dv4 (103c:30f7).

Reference: kernel bug #12440
	http://bugzilla.kernel.org/show_bug.cgi?id=12440Signed-off-by: NTakashi Iwai <tiwai@suse.de>

On the 92hd8xxx codecs port 0xe needs the connection selected to be the
last DAC in the list.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fixed invalid power mappings for ports 0xd and 0xe on 93hd83xxx codecs.
They were shifted right one too many bits.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The codec-parsing of STAC925x was utterly broken due to its unique
design unlike other STAC codecs.  It has a volume control only in NID
0x0e (similar as STAC9200), but the parser assumes that the amp is
available on each DAC widget.

The patch fixes the whole wrong stories: fix the initial volume,
assign the fixed "Master" volume, and avoid to create wrong volume
controls.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The value set in the commit 2465fb66
is actually wrong.  The value range is from 0 to 0x1f while the patch
sets to 0x7f.  Let's fix it.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fix HP dv5 (103c:3603) built-in mic input.

Reference: kernel bug 12440
	http://bugzilla.kernel.org/show_bug.cgi?id=12440Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Cc: stable@kernel.org

The selector widget 0x0e isn't initialized properly in the whole probe
process, thus it can be a wrong value depending on the BIOS setup.

This patch adds the init verb to set it to the max & unmuted.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fix the bug introduced in commit c15c5060:
  sound/pci/hda/patch_sigmatel.c: In function ‘patch_stac92hd83xxx’:
  sound/pci/hda/patch_sigmatel.c:4765: warning: assignment from incompatible pointer type
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Resetting HP pinctl at the unplugged state may cause a sort of regression
on some devices because of their wrong pin configuration.

A simple workaround is to disable the pin reset.  This is ugly and may be
not good from the power-saving POV (if any), but damn simple.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Cc: stable@kernel.org

Added amp nid for stac92hd8xxx families of codecs so the input amp
mixer is created.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

CF-74 does the headphone/speaker switching on hardware, thus the driver
shouldn't do any software-toggling of pins.  Otherwise it results in a
silent headphone output.

This patch simply resets the hp_detect flag to fix the problem.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add the model=hp-m4 quirk for another HP dv5 (103c:3603)
Reference: kernel bug#12440
	http://bugzilla.kernel.org/show_bug.cgi?id=12440Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Cc: stable@kernel.org

Fix a typo in stac92hd83xxx_cfg_tbl[].  The actual number is identical
thus there is no behavior change.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NHerton Ronaldo Krzesinski <herton@mandriva.com.br>
Cc: stable@kernel.org
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fixed incorrect mixer index values for 92hd83xx codec's audio
input mixer.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add check to determine if dinput_mux is set by any of patch_stac*() functions,
otherwise a invalid pointer my be referenced causing gibberish to mixer values.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the model=hp-m4 quirk for another HP dv7 (103c:30fc) with IDT
92HD71b* codec.

Reference: Novell bnc#461108
	https://bugzilla.novell.com/show_bug.cgi?id=461108

Cc: stable@kernel.org
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When no jack detection is available, the pins should be always
turned on since it can't be turned on/off dynamically via unsol
events.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fixed "unused varible" warnings in patch_sigmatel.c that have been
introduced by the last changes.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When the line_out has only one DAC and it's unique (i.e. not shared
by other outputs), assign a more reasonable and distinct mixer name
such as "Headphone" or "Speaker".
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The current auto-configuration code has several problems especially
for the new IDT codecs, e.g. wrong assignment of pins and DACs or
coupled volume for speaker and headphone.

This patch is a fairly large rewrite of the auto-configuration code.
Some remaks

- mic_switch and line_switch contain NIDs instead of bool
- dac_list isn't fixed for IDT 92HD* codecs now, they are all probed
- extra HP and speakers are stored in extra_dacs[].
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The previous commit re-enabled hp_nid setup for IDT92HD73*, but
it's unneeded indeed for Dell laptops that have multiple headphones.
Setting the extra hp_nid results in a non-working "Headpohne" mixer
control.  Thus hp_nid should be 0 for these dell models.

Also, the automatic addition of hp_nid should check whether it's
a dual-HP model or not.  For dual-HPs, the pins are already checked
by the early workaround.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fixed the call of snd_ctl_add() by replacing with snd_hda_ctl_add()
so that this mixer element can be tracked for re-configuration.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the model without the jack-detection for some desktops that
have really no jack-detection.  The recent driver caused regressions
regarding the sound output on such machines.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This reverts commit 07f455f7.
    ALSA: hda: removed unneeded hp_nid references

    Removed unneeded hp_nid references for 92hd73xx codec family.

This caused the silent output on some Intel desktops due to missing
routing of widget 0x0a and 0x0d.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the matching model=dell-m6 for Dell Studio 17 laptop.
Signed-off-by: NJoerg Schirottke <master@kanotix.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Some desktops seems to have no HP/mic jack detection on the front panel,
which results in the silent output in the recent driver, because the
driver mutes the output (to save power) when no plug is detected.

This patch adds a new model that disables the jack-detection.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The current code overrides the event type on input pins always to
PWR_EVENT.  Although this still works (PWR_EVENT and INSERT_EVENT
are handled samely), it'd be better to avoid such overrides.

Also, currently the unsol events are registered even for fixed pins
which will never raise the pin-detection event.

This patch fixes both issues.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The pin-detection function used in patch_sigmatel.c shouldn't be specific
to HP pin because it's used for input pins in general, too.

This patch fixes the detection function, removes the HP check from it
and moves to stac92xx_hp_detect().
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The power-state changes in patch_sigmatel.c are accessed via *_cached()
but they shouldn't be really cached.  Fixed to the normal write.

Also, stac92hd71xx_suspend and resume are no longer necessary as the
power-state changes are handled properly in the common routine.
Removed these hacks now.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The amp switch of SPDIF outputs have to be cached in the amp cache
instead of codec cache.  Otherwise it conflicts with the IEC958
playback switch control in hda_codec.c.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Gateway autodetection and entries are incomplete.

This patch adds the entries found at the .INI file for their driver version
5.10.5082.0.

It also uses the proper code to seek for notebook ID, since this is based on
codec subsystem ID on those devices.

This should provide a proper pinup for several gateways notebooks:
	Gateway M465-E Notebook [Part #1008637]
	Gateway M465-G Notebook [Part #1008667]
	Gateway NX260X Notebook [Part #1008794]
	Gateway NX100X Notebook [Part #1008798]
	Gateway E-100M Notebook [Part #1008799]
	Gateway E-100M G Notebook [Part #1008800]
	Gateway M255-E Notebook [Part #1008801]
	Gateway M255-G Notebook [Part #1008803]
	Gateway M285-E Convertible Notebook [Part #1008804]
	Gateway M285-G Convertible Notebook [Part #1008805]
	Gateway CX210S Convertible Notebook [Part #1008807]
	Gateway CX210X Convertible Notebook [Part #1008808]
	Gateway E-100M SB Notebook [Part #1008973]
	Gateway M255-E SB Notebook [Part #1008989]
	Gateway M285-E SB Convertible Notebook [Part #1008990]
	Gateway M465-E Notebook [Part #1009022]
	Gateway CX2724 Convertible Notebook [Part #1009036]
	Gateway MX1025 Notebook [Part #1009046]
	Gateway CX2720 Convertible Notebook [Part #1009063]
	Gateway CX2724h Convertible Notebook [Part #1009089]
	Gateway MX1023 Notebook [Part #1009097]
	Gateway MX1023h Notebook [Part #1009098]
	Gateway NX260X Notebook [Part #1009112]
	Gateway E-100M Notebook [Part #1009126]
	Gateway MX7533 Notebook [Part #1009146] [Part #1009163]
	Gateway CX210X Convertible Notebook [Part #1009346]
	Gateway NX570X Notebook [Part #1009442]
	Gateway NX570X Notebook [Part #1009448]
	Gateway NX270S Notebook [Part #1009550]
	Gateway MX6448 Notebook [Part #1013912R]
	Gateway MX6453 Notebook [Part #1013913R]
	Gateway MX6216 Notebook [Part #1013916R]
	Gateway MX6931 Notebook [Part #1013918R]
	Gateway CX2726 Convertible Notebook [Part #1013921R]
	Gateway MP8708 Notebook [Part #1013924R]
	Gateway MX6446 Notebook [Part #1013927R]
	Gateway MX6930 Notebook [Part #1013928R]
	Gateway MX6447 Notebook [Part #1013932R]
	Gateway MX6454 Notebook [Part #1013943R]
	Gateway MX6439 Notebook [Part #1013947R] [Part #1013955R] [Part #1013971R]
	Gateway MX6930h Notebook [Part #1013973R] [Part #1013974R] [Part #1013975R]
	Gateway MX6955 Notebook [Part #1014028R]
	Gateway MX6956 Notebook [Part #1014033R]
	Gateway MX6959 Notebook [Part #1014061R]
	Gateway MX6957 Notebook [Part #1014065R]
	Gateway MX6960 Notebook [Part #1014068R]
	Gateway MX6958 Notebook [Part #1014072R]
	Gateway NX570X Notebook [Part #1014077R]
	Gateway NX570XL Notebook [Part #1014078R]
	Gateway NX570QS Notebook [Part #1014079R]
	Gateway MX6961 Notebook [Part #1014080R] [Part #1014106R]
	Gateway MX6961h Notebook [Part #1014112R]
	Gateway NX270S Notebook [Part #1014120R]
	Gateway MX6431 Notebook [Part #1014121R]
	Gateway MX8710 Notebook [Part #2905895R]
	Gateway MX3702 Notebook [Part #2905898R]
	Blade-K8F GW UMA Single Core Motherboard w/RS485M and 1394 - Quanta (FRU) [Part #4006133R]

Since some entries conflict with existing pinups, I'm providing a separate
patch to fix those entries.

Tested only with Gateway MX6453.
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added power-map and analog-loopback information to proc output for
IDT/STAC codecs.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Split the monolithc HD-audio driver into several pieces:
 - snd-hda-intel   HD-audio PCI controller driver; loaded via udev
 - snd-hda-codec   HD-audio codec bus driver
 - snd-hda-codec-* Specific HD-audio codec drivers

When built as modules, snd-hda-codec (that is invoked by snd-hda-intel)
looks up the codec vendor ID and loads the corresponding codec module
automatically via request_module().

When built in a kernel, each codec drivers are statically hooked up
before probing the PCI.

This patch adds appropriate EXPORT_SYMBOL_GPL()'s and the module
information for each driver, and driver-linking codes between
codec-bus and codec drivers.

TODO:
  - Avoid EXPORT_SYMBOL*() when built-in kernel
  - Restore __devinit appropriately depending on the condition
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Check the model type instead of PCI SSID for detection of the mic types
on Dell laptops with IDT 92HD73xx codecs.  In this way, a new laptop
can be tested via model module option.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fixed the quirk string for Dell studio 1535 (the product name wasn't
published at the time the patch was made).
Signed-off-by: NTakashi Iwai <tiwai@suse.de>