Add new parameter to disable rounding of buffer/period sizes to
multiples of 128 bytes. This is more efficient in terms of memory
access but isn't required by the HDA spec and prevents users from
specifying exact period/buffer sizes. For example for 44.1kHz, a
period size set to 20ms will be rounded to 19.59ms.

Tested and enabled on Intel HDA controllers. Option is disabled by
default for other controllers.
Tested-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NPierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add a few tracepoints to HD-audio driver.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

It works fine with auto-parser and now the digital mic workaround was
implemented in auto-parser fixup, let's drop the static model quirks for
these models.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The digital-mic unit on ASUS Eee PC gives PDM signals instead of the
normal stereo PCM, thus you can't record a mono stream from the stereo
stream as is; the summed stereo signal results in almost zero level, and
you'll hear only soft noise.

As a workaround, use ALC269-specific COEF to manipulate the dmic route
for mono, like used for ALC271x.  This is implemented as a fix-up, thus
it works only with model=auto or without REALTEK_QUIRKS Kconfig.
Reported-and-tested-by: NPavel Roskin <proski@gnu.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Copying hp_pins and speaker_pins from line_out_pins may confuse the
parser, and it can lead to duplicated initializations for the same pin
with a wrong DAC assignment.  The problem appears in 3.0 kernel code.

Cc: <stable@kernel.org> (for 3.0)
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Somce quirk models don't set adc_nids but let the parser filling it.
But the recent code has unnecessary NULL-checks of spec->input_mux,
and it resulted in NULL dereferences.
This patch fixes that regression.
Reported-and-tested-by: NOliver Neukum <oneukum@suse.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch also registers all necessary callbacks to support mute LED
only when such control is enabled. And it keeps codec AFG in D0 or D1
state all the time when aggressive power managemnt is enabled for vref-out
control (and mute LED) work correctly.
Signed-off-by: NVitaliy Kulikov <Vitaliy.Kulikov@idt.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This update includes the changes necessary for supporting the
CS421x family of codecs.  Previously this file only supported
the CS420x family of codecs.

This file also contains init verbs to correct several issues in
the CS421x hardware.

Behavior between the CS421x and CS420x codec families is similar,
so several functions have been reused with "if" statements to
determine which codec family (CS421x or CS420x) is present.

Also, this file will be updated sometime in the near future in
order to add support for a system using CS421x that requires
mono mix on the speaker output only.

[Fix const usages and adaption for new APIs by tiwai]
Signed-off-by: NTim Howe <tim.howe@cirrus.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The procedure for codec D-state change may have exceptional cases
depending on the codec chip, such as a longer delay or suppressing D3.

This patch adds a new codec ops, set_power_state() to override the system
default function.  For ease of porting, snd_hda_codec_set_power_to_all()
helper function is extracted from the default set_power_state() function.

As an example, the Conexant codec-specific delay is removed from the
default routine but moved to patch_conexant.c.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add a new ops, post_suspend(), which is called after suspend() ops is
performed.  This is called only in the case of the real PM suspend, and
the codec driver can use this for further changing of D-state or
clearing the LED, etc.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

It makes little sense to enable power-saving without PM.
This removes SND_HDA_NEEDS_RESUME define so that we can use CONFIG_PM
in all places.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch adds checking of mute state on all outputs besides just
speakers to calculate the master mute state for mute led support.
It also renames and splits the function that does it for better code
clarity.
Signed-off-by: NVitaliy Kulikov <Vitaliy.Kulikov@idt.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Codec state is not restored immediately on resume but on the first
access when power-save is enabled.  That leads to an invalid mute led
state after resume until either sound is played or some control is
changed.  This patch adds a possibility for a vendor specific patch to
restore codec state immediately after resume if required.  And it adds
code to restore IDT codecs state immediately on resume on HP systems
with mute led support.
Signed-off-by: NVitaliy Kulikov <Vitaliy.Kulikov@idt.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NVitaliy Kulikov <Vitaliy.Kulikov@idt.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fix a regression in the DAC filling code in patch_realtek.c.  The already
filled DACs in multiout.dac_nids[] were ignored because of num_dacs=0,
thus always pointed to the first DAC.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

"force" argument is always true, so let's strip it off.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch fixes non-working indep-HP control on VT1708* codecs.
The problems are that via_independent_hp_put() wasn't fixed to follow
the recent change of three HP paths, and hp_indep_path didn't contain
the amp nids of mixer elements.

Together with the fixes, a few code clean-ups are done.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch changes the behavior of independent-HP enum switch.  Now
instead of returning a busy error, the driver switches dynamically the
stream of the HP (and shared) DACs according to the current mode.
The logic is similar like the dual-mic ADC switch, but a bit more
complicated because of the presence of shared DAC.

Together with the change, a mutex is introduced to protect against the
possible races for the indep-HP mode setting.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch adds the dynamic control of analog-loopback for VIA codecs.

When the loopback is enabled, the inputs from line-ins and mics are
mixed with the front DAC, and sent to the front outputs.  The very same
input is routed to the headhpones and speakers in loopback mode.
However, since the loopback mix can't take other than the front DAC,
there is no longer individual volume controls for headphones and
speakers.  Once when the loopback control is off, these volumes take
effect.

Since the individual volumes are more desired in general use caess, the
loopback mode is set to off as default for now.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

It was obviously wrong, grr....
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

During the rewrite, the check of spec->need_dac_fix and the corresponding
num_dacs change was dropped from the channel-mode control.

This patch re-adds it, and also enables need_dac_fix for ALC880 as default,
as this feature was originally introduced to fix h/w bugs of this chip.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This commit is a fix up for commit acfa634f.

	commit acfa634f
	Author: Takashi Iwai <tiwai@suse.de>
	Date:   Tue Jul 12 17:27:46 2011 +0200

		  ALSA: hda - Add Kconfig for the default buffer size
Signed-off-by: NPaul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add a Kconfig entry to specify the default buffer size.
Distros using PulseAudio can choose a larger value here.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NTakashi Iwai <tiwai@suse.de>

VT1718S and co have a secret connection from DAC to AA-mix, which
doesn't appear in the connection list obtained from the h/w.
Currently the driver fixes the connection index locally at init, but
now we can expose it statically via snd_hda_override_connections()
so that this conection can be checked better by the parser in future.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

In the codec proc outputs, read the raw connections instead of the
cached connection list, i.e. proc files contain only raw values.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add a function to add/modify the connection-list cache entry.
It'll be useful to fix a buggy hardware result.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Some machines seem to use EAPD control of the unused pin for controlling
the overall EAPD.  Since the driver currently doesn't check the EAPD of
unused pins, the EAPD isn't enabled.  For avoiding such a problem, turn
all extra EAPDs on as default.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

For smart51 pins, we need to preserve the input pin-control bits at
auto-mute controls instead of overwriting zero or pin-out-only.
Otherwise the VREF won't be set properly when smart51 is disabled
again.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When Independent-HP mode is changed for VIA, the driver needs to
re-issue the auto-mute check so that the line-out pins are set properly
without influence of HP pin state.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When the line-jack is plugged/unplugged, the driver must check also
the headphone jack state in addition to the line-out jack.  Currently
it checks only the line-out state and ignores the headphone.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Instead of checking the model quirk, use a fixup table for workaround
of 44kHz-fixed PCM for Lenovo IdeaPad with ALC269.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

It's harmless but annyoing.
  sound/pci/hda/patch_realtek.c: In function ‘alc_cap_getput_caller’:
  sound/pci/hda/patch_realtek.c:2722:9: warning: ‘err’ may be used uninitialized in this function
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Now all alc*_parse_auto_config() do almost same thing except for the
NID list to ignore and the PINs for SSID-check, we can merge all these
to a single function.  A good amount of code reduction.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Finally the last one.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

One more code reduction.  This codec has less DACs, thus the wiring
to DAC can't be filled uniquely for all output pins, i.e. some outputs
share the same volume control.
Except for that, all seems working fine.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Now coming to ALC268/269 parser codes.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Merge more auto-parser code in patch_realtek.c, now for ALC861.
The topology of this codec is pretty simple, and can be parsed well
by the current starndard parser.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

query_amp_caps() may return non-zero if the amp cap isn't supported
by the codec.  Thus one needs to check widget-caps first, then check
the corresponding amp-caps.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The existing standard auto-parser can work well with this codec, too.
Let's merge.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>