This reverts commit 785f857d.

The commit causes a problem with the wrong D3 state after suspend
because the call of hda_set_power_state() involves with the power-up
sequence, which changes the power_count, and this confuses the resume
sequence that checks the power_count as well.

Originally, this go-to-D3 sequence should be a simple task without the
power-up sequence.  But, it'd need some proper sanity checks in the
case of power-saved state, so it's not too easy to write now in the
3.4-rc cycle.

In short, the safest option now is to revert this affecting commit.

Of course, we need to clean up and robustify the power-saving code
better for 3.5 kernel.
Reported-by: NKonstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Since it's not always safe to assume that the vmaster hook is purely the
mute-LED control, add the flag indicating whether to expose the mute-LED
enum control or not.  Currently, conexant codec sets this off for non-HP
laptops where EAPD may be used really as EAPD.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Create snd_hda_add_vmaster_hook() and snd_hda_sync_vmaster_hook()
helper functions to handle the mute-LED in vmaster hook more
commonly.  In the former function, a new enum control "Mute-LED Mode"
is added.  This provides user to choose whether the mute-LED should be
turned on/off explicitly or to follow the master-mute status.
Reviewed-by: NDavid Henningsson <david.henningsson@canonical.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

It'll be used for adding hooks in later patches.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When the driver is changed to use vmaster or a new slave element is
added by the improvement of the parser code, user may face often the
silent output because of the muted slave mixer although Master volume
is properly set.  And they complain.  And I get upset.

Although such a mixer element should be initialized via "alsactl init",
it'd be more user-friendly if the known output slaves are unmuted and
set to 0dB so that user can control the volume only with Master as
default.  Since Master is still set muted as default even with this
change, no risk of the speaker blow up, too.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

We've seen a problem with a pop-noise at suspend/resume on a HP
machine with ALC269, and it turned out to be an issue that the
controller going to D3 while the codec is unused.

When the device is once suspended and resumed and kept unused, the
driver doesn't initialize the codecs.  Instead, the codec chips are
set up dynamically at the first usage.  Now, suppose the device going
to suspend again before the codec is set up.  The controller is turned
off to D3 while the codec chips are untouched.  This caused a pop
noise because the codec chip might have been turned on implicitly by
the hardware.

As a workaround, the codec chip needs to be set to D3 when going to
suspend no matter whether it was used or not.  Also, for making it
happening, the controller has to be always set up in the resume path.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Kill hyphens from "Line-Out" name strings, as suggested by Mark Brown.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Some codecs don't supply the mute amp-capabilities although the lowest
volume gives the mute.  It'd be handy if the parser provides the mute
mixers in such a case.

This patch adds an extension amp-cap bit (which is used only in the
driver) to represent the min volume = mute state.  Also modified the
amp cache code to support the fake mute feature when this bit is set
but the real mute bit is unset.

In addition, conexant cx5051 parser uses this new feature to implement
the missing mute controls.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=42825

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

Wrongly slipped in from the commit 9322ca54.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

In most cases, the slave strings for vmaster are identical between
volumes and switches except for "xxx Volume" and "xxx Switch" suffix.
Now snd_hda_add_vmaster() takes the optional suffix argument so that
each string can be composed with the given suffix, and we can share the
slave name strings in both volume and switch calls nicely.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This typo caused the wrong codec's nid to be checked for wcaps type.
As a result, sometimes speakers would duplicate the output sent to
HDMI output.

Cc: stable@kernel.org
BugLink: https://bugs.launchpad.net/bugs/924320Signed-off-by: NDavid Henningsson <david.henningsson@canonical.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

So far, the driver supports up to 10 streams.  This is a restriction in
hda_intel.c and hda_codec.c: in the former, the fixed array size limits
the amount, and in the latter, the fixed device-number assignment table
(in get_empty_pcm_device()) limits the possibility.

This patch reduces the restriction by
- using linked list for managing PCM instances in hda_intel.c, and
- assigning non-fixed device numbers for the extra devices
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

In snd_hda_check_board_codec_sid_config(), not only comparing with the
exact value but allow the bit-mask comparison for vendor-only, etc.
Tested-by: NLinus Torvalds <torvalds@linux-foundation.org>
Tested-by: NDirk Hohndel <hohndel@infradead.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The multiple headphone pins are usually handled as copied from the same
source, not as individual channels like front and surround.  Thus it'd
be more correct to avoid the channel suffix for "Headphone" pin labels
in snd_hda_get_pin_label() but give an index number instead.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Reported-by: NDavid Henningsson <david.henningsson@canonical.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The function now gives more unique names for the output pins by adding
some prefix and suffix for the location and the channels.  Otherwise, it
can pass the index number.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

We can use the very same table in hda_jack.c for managing the list for
input-jack elements, too.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Introduce a table containing the pins and their jack-detection states
for avoiding the unnecessary verbs to check the pin status at each time.

When the unsol event is enabled via snd_hda_jack_detect_enable(), it
automatically adds the given NID to the table.  Then the driver supposes
that the codec driver will set the dirty flag appropariately when an
unsolicited event is invoked for that pin.

The behavior for reading other pins that aren't registered in the table
doesn't change.  Only the pins assigned to the table are cached, so far.

In near futre, this table can be extended to use the central place for
the unsolicited events of all pins, etc, and eventually include the
jack-detect kcontrols that replace the current input-jack stuff.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Create a new helper function snd_hda_get_pin_label() for getting a label
string for both input and output pins.  hda_get_input_pin_label() is
obsoleted by this function, and the callers are replaced appropriately
now by this patch.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

We disabled the check of NO_PRESENCE bit of the default pin-config
in commit f4419172 temporarily.  One problem was that the first
implementation was wrong -- the bit after the shift must be checked.
However, this would still give many regressions on machines with broken
BIOS.  They set this bit wrongly even on active pins.

A workaround is to check whether all pins contain this bit.  As far as
I've checked, broken BIOSen set this bit on all pins, no matter whether
active or not.  In such a case, the driver should ignore this bit check.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When a virtual mater control is created, the driver looks for slave
elements from the assigned card instance.  But this may include the
elements of other codecs when multiple codecs are on the same HD-audio
bus.  This works at the first time, but it'll give Oops when it's once
freed and re-created via reconfig sysfs.

This patch changes the element-look-up strategy to limit only to the
mixer elements of the same codec.
Reported-by: NDavid Henningsson <david.henningsson@canonical.com>
Cc: <stable@kernel.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Lots of sound drivers were getting module.h via the implicit presence
of it in <linux/device.h> but we are going to clean that up.  So
fix up those users now.
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>

When 5.1 or more headphone or speaker pins are provided, the parser still
takes as is without fixing the order of channel mapping, which leads in
the unexpected strange channel order by surround outputs.

This patch fixes the issue by applying the same fix-up not only to
line_out_pins[] but also hp_pins[] and speaker_pins[].
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Since hda_proc.c is now the only user of snd_print_pcm_rates(), better to
put it back locally to hda_proc.c and revert to the old style.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

SAD sampling rate information reported in
/proc/asound/cardX/eldX is incorrect due to a mismatch
between HDA and HDMI frequencies. Add new routine to provide
relevant values.
Signed-off-by: NPierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Similar to Line Out, these constants form the base for future
patches enabling input jack reporting for Line in jacks.
Signed-off-by: NDavid Henningsson <david.henningsson@canonical.com>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Since really_cleanup_stream() is called from both purity_inactive_streams()
and hda_cleanup_all_streams(), the verbs to clear the PCM channel and
format may be called multiple times unnecessarily.

This patch adds checks to skip these unneeded verbs.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The recursive search of widget connections in snd_hda_get_conn_index()
must be terminated at the pin and the audio-out widgets.  Otherwise
you'll get "too deep connection" warnings unnecessarily.
Reported-by: NFrancis Moreau <francis.moro@gmail.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

For supporting both the multiple headphones and the multiple speakers,
add the new field in struct hda_multi_out, and evaluate in the standard
setup functions.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

... and add a new bit-flags argument to specify the behavior of the
function.  The older function is kept as is (as a wrapper).
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

In snd_hda_parse_pin_def_config(), we checked the associated number
of speaker pins and accepts only one number exclusively.  But many BIOS
seem to give different assoc number for surround speakers, thus we'd
better to accept all speaker pins no matter which assoc number, and sort
like done for the headphone pins.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

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

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

Add a few tracepoints to HD-audio driver.
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>

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>

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

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