The reconfiguration of PCM affected all PCM streams on the bus, but
this this should be done rather only for the target codec.

This patch does the following:
- introduce bitmap indicating the PCM device usages on a hda_bus
- refactor the PCM build functions
- fix __devinit prefix in some fucntions
- add a proper ifdef around HDA-reconfig-specific functions
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Move power_save option into hda_intel.c, and make a field in hda_bus,
instead of keeping module parameters in separate files.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Minor clean ups: move snd_hda_codecs_inuse() into hda_intel.c and
make static.  Also, make snd_hda_query_supported_pcm() static
as it's used only in hda_codec.c.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

There is no argument named @state in snd_hda_resume,
remove its' comment.
Signed-off-by: NQinghuang Feng <qhfeng.kernel@gmail.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

SPDIF status bits controls are written via snd_hda_codec_write()
without caching.  This causes a regression at resume that the bits
are lost.

Simply replacing it with the cached version fixes the problem.

Reference:
	http://lkml.org/lkml/2008/11/24/324Signed-off-by: NTakashi Iwai <tiwai@suse.de>

snd_print_pcm_rates() and snd_print_pcm_bits() are used by both
hda_proc.c and hda_eld.c, thus they have to be defined in the common
place.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The jack instances created in patch_sigmatel.c may be double-freed.
The device management code checks the invalid element, and thus there
is no real breakage, but it spews annoying warning messages.
But, we can't simply remove the release calls of these jack instances
because they have to be freed when the codec is re-configured.

Now, a new flag, bus->shutdown is introduced to indicate that the bus
is really being unloaded, i.e. the objects managed by the device
manager will be automatically deleted.  We release these objects only
when this flag isn't set.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fix another silly bug in the amp cache list management.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The chaining for amp hash got broken due to the rewrite with
snd_array.  Fixed now.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added a check for max allocation size in snd_array_new() for a
debugging purpose.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the missing NULL check from allocator in get_alloc_hash().
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The hda_bus ops callback take struct hda_bus pointer.
Also, the command callback takes the composed command word, instead of
each small bits in arguments.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add support for Intel G45 integrated HDMI audio codecs.

This initial release supports:
	- 2 channel stereo sound output
	- report monitor's ELD information
Signed-off-by: NWu Fengguang <wfg@linux.intel.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Due to the hda-reconfiguration patches, the check of empty stream
is gone, and this results in an error with the codec setup with empty
streams.

This patch adds the check again to avoid the error at probing.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add NULL-check of the return value of snd_kctl_new1() before
accessing it.  Also, make a sanity NULL check to snd_BUG_ON()
for debugging only.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Add Wolfson Microelectronics to the HDA vendor ID table.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch adds snd_hda_get_jack* functions for reporting jack location,
device, and connectivity type.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This patch enables the additional init verbs for each codec.  The verbs
can be entered via hwdep sysfs file.  These verbs are executed at
reconfiguring the codec for non-standard setups like overriding
the pin-defcfg.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the sysfs entries to hwdep devices so that the new features
like reconfiguration can be done via sysfs.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the reconfiguration feature of any individual codec.
Via the reconfiguration, the old resources are released and
the patch is called again to recreate the PCM and mixers in
addition to the re-initialization.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Manage all kcontrol elements created in the hda-intel driver.
This makes it possible to remove and reconfigure the controls
of each codec.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Allocate dynamically the name string of each codec instead of
pointing to a static string.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Use generic array helpers to simplify array handling in snd-hda-intel.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added helper functions to handle generic arrays.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added the infrastructure for dynamic stream allocation on HD-audio.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

fix a typo in comment for is_in_nid_list().
Signed-off-by: NWu Fengguang <wfg@linux.intel.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

fix a typo in comment for process_unsol_events().

Signed-off-by: Wu Fengguang<wfg@linux.intel.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Add NVIDIA HDMI HD-audio codec support in snd-hda-intel driver,
include NVIDIA MCP78/7A HDMI.
Signed-off-by: NWei Ni <wni@nvidia.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

The recent slave_dig_out addition has some rooms to clean up.
Also it doesn't call snd_hda_cleanup_stream() properly for slaves
at closing.  The patch fixes both issues.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Removed invalid references to slave_dig_outs inside the S/PDIF IN capture switch
control. Beforehand this was basically a mute switch for the S/PDIF outs as well.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Added support for controlling hardware gain amps on input ports
using a volume control mixer with a mux selecting the port being controlled.
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Many places fail to check if codec has slave_dig_outs entries (the most common
case is not having any entry), leading to various possible oopses in hda_codec
code.
Signed-off-by: NHerton Ronaldo Krzesinski <herton@mandriva.com.br>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Added support for playing a stream on multiple digital outs. This is done
by defining codec->slave_dig_outs as array of hda_nid_t with a null-terminated entry to set the
slave SPDIF outs, in which the slave outs have cloned settings of the master out (e.g. dig_out_nid).
Signed-off-by: NMatthew Ranostay <mranostay@embeddedalley.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Export HDA codec subvendor ID and revision ID to user space via the
components variable. Our alsactl utility requires these values for
the perfect hardware identification.
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

The SPDIF output is toggled at each time any SPDIF status bits are changed
because of the known problems on some codecs.  But, this also results in
loosing the sync, and the problem is more obvious on HDMI output over
SPDIF.  Since the toggle is necessary only for some codecs, we should
check whether this workaround is needed and skip if unnecessary.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Kill snd_assert() in sound/pci/*, either removed or replaced with
if () with snd_BUG_ON().
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NJaroslav Kysela <perex@perex.cz>

Replace CONFIG_SND_DEBUG_DETECT with CONFIG_SND_DEBUG_VERBOSE to
represent its meaning more better.  This config isn't provided only
for the detection but for more verbose debug prints in general.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

On Wed, Apr 02, 2008 at 08:19:29AM -0400, Miles Lane wrote:
> [   48.765906] [ BUG: bad unlock balance detected! ]
> [   48.765912] -------------------------------------
> [   48.765918] pulseaudio/4277 is trying to release lock
> (&codec->spdif_mutex) at:
> [   48.765930] [<c03031b7>] mutex_unlock+0x8/0xa
> [   48.765945] but there are no more locks to release!
> [   48.765950]
> [   48.765952] other info that might help us debug this:
> [   48.765959] 2 locks held by pulseaudio/4277:
> [   48.765965]  #0:  (&pcm->open_mutex){--..}, at: [<f89f134b>]
> snd_pcm_open+0xc1/0x1ba [snd_pcm]
> [   48.766003]  #1:  (&chip->open_mutex){--..}, at: [<f8b4f13d>]
> azx_pcm_open+0x36/0x184 [snd_hda_intel]
> [   48.766057]
> [   48.766059] stack backtrace:
> [   48.766066] Pid: 4277, comm: pulseaudio Not tainted 2.6.25-rc8-mm1 #12
> [   48.766086]  [<c013afc6>] print_unlock_inbalance_bug+0xce/0xd8
> [   48.766107]  [<c0109e1c>] ? save_stack_trace+0x1d/0x3b
> [   48.766130]  [<c012f54e>] ? __kernel_text_address+0x1b/0x27
> [   48.766146]  [<c0104533>] ? dump_trace+0xcd/0xd9
> [   48.766160]  [<c0109d9e>] ? save_stack_address+0x0/0x2c
> [   48.766176]  [<c013b80a>] ? find_usage_backwards+0xa4/0xc3
> [   48.766193]  [<c013cfb5>] lock_release_non_nested+0x84/0x120
> [   48.766209]  [<c03031b7>] ? mutex_unlock+0x8/0xa
> [   48.766222]  [<c013d1bb>] lock_release+0x16a/0x199
> [   48.766238]  [<c0303137>] __mutex_unlock_slowpath+0xa9/0x121
> [   48.766252]  [<c03031b7>] mutex_unlock+0x8/0xa
> [   48.766263]  [<f8b4ffd8>] snd_hda_multi_out_analog_open+0xd3/0xef
> [snd_hda_intel]

The following patch should fix it.

Cc: "Miles Lane" <miles.lane@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Keep the format verb at closing PCM streams.
Introduced snd_hda_codec_cleanup_stream() for the parcicular purpose.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Move the array declaration to hda_codec.c where it is used and add includes
where the individual presets are declared.

Fixes the following sparse warnings:
sound/pci/hda/patch_realtek.c:13744:25: warning: symbol 'snd_hda_preset_realtek' was not declared. Should it be static?
sound/pci/hda/patch_cmedia.c:729:25: warning: symbol 'snd_hda_preset_cmedia' was not declared. Should it be static?
sound/pci/hda/patch_analog.c:3656:25: warning: symbol 'snd_hda_preset_analog' was not declared. Should it be static?
sound/pci/hda/patch_sigmatel.c:3995:25: warning: symbol 'snd_hda_preset_sigmatel' was not declared. Should it be static?
sound/pci/hda/patch_si3054.c:286:25: warning: symbol 'snd_hda_preset_si3054' was not declared. Should it be static?
sound/pci/hda/patch_atihdmi.c:156:25: warning: symbol 'snd_hda_preset_atihdmi' was not declared. Should it be static?
sound/pci/hda/patch_conexant.c:1721:25: warning: symbol 'snd_hda_preset_conexant' was not declared. Should it be static?
sound/pci/hda/patch_via.c:1962:25: warning: symbol 'snd_hda_preset_via' was not declared. Should it be static?
Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>