1=============================
2More Notes on HD-Audio Driver
3=============================
4
5Takashi Iwai <tiwai@suse.de>
6
7
8General
9=======
10
11HD-audio is the new standard on-board audio component on modern PCs
12after AC97.  Although Linux has been supporting HD-audio since long
13time ago, there are often problems with new machines.  A part of the
14problem is broken BIOS, and the rest is the driver implementation.
15This document explains the brief trouble-shooting and debugging
16methods for the	HD-audio hardware.
17
18The HD-audio component consists of two parts: the controller chip and
19the codec chips on the HD-audio bus.  Linux provides a single driver
20for all controllers, snd-hda-intel.  Although the driver name contains
21a word of a well-known hardware vendor, it's not specific to it but for
22all controller chips by other companies.  Since the HD-audio
23controllers are supposed to be compatible, the single snd-hda-driver
24should work in most cases.  But, not surprisingly, there are known
25bugs and issues specific to each controller type.  The snd-hda-intel
26driver has a bunch of workarounds for these as described below.
27
28A controller may have multiple codecs.  Usually you have one audio
29codec and optionally one modem codec.  In theory, there might be
30multiple audio codecs, e.g. for analog and digital outputs, and the
31driver might not work properly because of conflict of mixer elements.
32This should be fixed in future if such hardware really exists.
33
34The snd-hda-intel driver has several different codec parsers depending
35on the codec.  It has a generic parser as a fallback, but this
36functionality is fairly limited until now.  Instead of the generic
37parser, usually the codec-specific parser (coded in patch_*.c) is used
38for the codec-specific implementations.  The details about the
39codec-specific problems are explained in the later sections.
40
41If you are interested in the deep debugging of HD-audio, read the
42HD-audio specification at first.  The specification is found on
43Intel's web page, for example:
44
45* https://www.intel.com/standards/hdaudio/
46
47
48HD-Audio Controller
49===================
50
51DMA-Position Problem
52--------------------
53The most common problem of the controller is the inaccurate DMA
54pointer reporting.  The DMA pointer for playback and capture can be
55read in two ways, either via a LPIB register or via a position-buffer
56map.  As default the driver tries to read from the io-mapped
57position-buffer, and falls back to LPIB if the position-buffer appears
58dead.  However, this detection isn't perfect on some devices.  In such
59a case, you can change the default method via ``position_fix`` option.
60
61``position_fix=1`` means to use LPIB method explicitly.
62``position_fix=2`` means to use the position-buffer.
63``position_fix=3`` means to use a combination of both methods, needed
64for some VIA controllers.  The capture stream position is corrected
65by comparing both LPIB and position-buffer values.
66``position_fix=4`` is another combination available for all controllers,
67and uses LPIB for the playback and the position-buffer for the capture
68streams.
69``position_fix=5`` is specific to Intel platforms, so far, for Skylake
70and onward.  It applies the delay calculation for the precise position
71reporting.
72``position_fix=6`` is to correct the position with the fixed FIFO
73size, mainly targeted for the recent AMD controllers.
740 is the default value for all other
75controllers, the automatic check and fallback to LPIB as described in
76the above.  If you get a problem of repeated sounds, this option might
77help.
78
79In addition to that, every controller is known to be broken regarding
80the wake-up timing.  It wakes up a few samples before actually
81processing the data on the buffer.  This caused a lot of problems, for
82example, with ALSA dmix or JACK.  Since 2.6.27 kernel, the driver puts
83an artificial delay to the wake up timing.  This delay is controlled
84via ``bdl_pos_adj`` option.
85
86When ``bdl_pos_adj`` is a negative value (as default), it's assigned to
87an appropriate value depending on the controller chip.  For Intel
88chips, it'd be 1 while it'd be 32 for others.  Usually this works.
89Only in case it doesn't work and you get warning messages, you should
90change this parameter to other values.
91
92
93Codec-Probing Problem
94---------------------
95A less often but a more severe problem is the codec probing.  When
96BIOS reports the available codec slots wrongly, the driver gets
97confused and tries to access the non-existing codec slot.  This often
98results in the total screw-up, and destructs the further communication
99with the codec chips.  The symptom appears usually as error messages
100like:
101::
102
103    hda_intel: azx_get_response timeout, switching to polling mode:
104          last cmd=0x12345678
105    hda_intel: azx_get_response timeout, switching to single_cmd mode:
106          last cmd=0x12345678
107
108The first line is a warning, and this is usually relatively harmless.
109It means that the codec response isn't notified via an IRQ.  The
110driver uses explicit polling method to read the response.  It gives
111very slight CPU overhead, but you'd unlikely notice it.
112
113The second line is, however, a fatal error.  If this happens, usually
114it means that something is really wrong.  Most likely you are
115accessing a non-existing codec slot.
116
117Thus, if the second error message appears, try to narrow the probed
118codec slots via ``probe_mask`` option.  It's a bitmask, and each bit
119corresponds to the codec slot.  For example, to probe only the first
120slot, pass ``probe_mask=1``.  For the first and the third slots, pass
121``probe_mask=5`` (where 5 = 1 | 4), and so on.
122
123Since 2.6.29 kernel, the driver has a more robust probing method, so
124this error might happen rarely, though.
125
126On a machine with a broken BIOS, sometimes you need to force the
127driver to probe the codec slots the hardware doesn't report for use.
128In such a case, turn the bit 8 (0x100) of ``probe_mask`` option on.
129Then the rest 8 bits are passed as the codec slots to probe
130unconditionally.  For example, ``probe_mask=0x103`` will force to probe
131the codec slots 0 and 1 no matter what the hardware reports.
132
133
134Interrupt Handling
135------------------
136HD-audio driver uses MSI as default (if available) since 2.6.33
137kernel as MSI works better on some machines, and in general, it's
138better for performance.  However, Nvidia controllers showed bad
139regressions with MSI (especially in a combination with AMD chipset),
140thus we disabled MSI for them.
141
142There seem also still other devices that don't work with MSI.  If you
143see a regression wrt the sound quality (stuttering, etc) or a lock-up
144in the recent kernel, try to pass ``enable_msi=0`` option to disable
145MSI.  If it works, you can add the known bad device to the blacklist
146defined in hda_intel.c.  In such a case, please report and give the
147patch back to the upstream developer.
148
149
150HD-Audio Codec
151==============
152
153Model Option
154------------
155The most common problem regarding the HD-audio driver is the
156unsupported codec features or the mismatched device configuration.
157Most of codec-specific code has several preset models, either to
158override the BIOS setup or to provide more comprehensive features.
159
160The driver checks PCI SSID and looks through the static configuration
161table until any matching entry is found.  If you have a new machine,
162you may see a message like below:
163::
164
165    hda_codec: ALC880: BIOS auto-probing.
166
167Meanwhile, in the earlier versions, you would see a message like:
168::
169
170    hda_codec: Unknown model for ALC880, trying auto-probe from BIOS...
171
172Even if you see such a message, DON'T PANIC.  Take a deep breath and
173keep your towel.  First of all, it's an informational message, no
174warning, no error.  This means that the PCI SSID of your device isn't
175listed in the known preset model (white-)list.  But, this doesn't mean
176that the driver is broken.  Many codec-drivers provide the automatic
177configuration mechanism based on the BIOS setup.
178
179The HD-audio codec has usually "pin" widgets, and BIOS sets the default
180configuration of each pin, which indicates the location, the
181connection type, the jack color, etc.  The HD-audio driver can guess
182the right connection judging from these default configuration values.
183However -- some codec-support codes, such as patch_analog.c, don't
184support the automatic probing (yet as of 2.6.28).  And, BIOS is often,
185yes, pretty often broken.  It sets up wrong values and screws up the
186driver.
187
188The preset model (or recently called as "fix-up") is provided
189basically to overcome such a situation.  When the matching preset
190model is found in the white-list, the driver assumes the static
191configuration of that preset with the correct pin setup, etc.
192Thus, if you have a newer machine with a slightly different PCI SSID
193(or codec SSID) from the existing one, you may have a good chance to
194re-use the same model.  You can pass the ``model`` option to specify the
195preset model instead of PCI (and codec-) SSID look-up.
196
197What ``model`` option values are available depends on the codec chip.
198Check your codec chip from the codec proc file (see "Codec Proc-File"
199section below).  It will show the vendor/product name of your codec
200chip.  Then, see Documentation/sound/hd-audio/models.rst file,
201the section of HD-audio driver.  You can find a list of codecs
202and ``model`` options belonging to each codec.  For example, for Realtek
203ALC262 codec chip, pass ``model=ultra`` for devices that are compatible
204with Samsung Q1 Ultra.
205
206Thus, the first thing you can do for any brand-new, unsupported and
207non-working HD-audio hardware is to check HD-audio codec and several
208different ``model`` option values.  If you have any luck, some of them
209might suit with your device well.
210
211There are a few special model option values:
212
213* when 'nofixup' is passed, the device-specific fixups in the codec
214  parser are skipped.
215* when ``generic`` is passed, the codec-specific parser is skipped and
216  only the generic parser is used.
217
218A new style for the model option that was introduced since 5.15 kernel
219is to pass the PCI or codec SSID in the form of ``model=XXXX:YYYY``
220where XXXX and YYYY are the sub-vendor and sub-device IDs in hex
221numbers, respectively.  This is a kind of aliasing to another device;
222when this form is given, the driver will refer to that SSID as a
223reference to the quirk table.  It'd be useful especially when the
224target quirk isn't listed in the model table.  For example, passing
225model=103c:8862 will apply the quirk for HP ProBook 445 G8 (which
226isn't found in the model table as of writing) as long as the device is
227handled equivalently by the same driver.
228
229
230Speaker and Headphone Output
231----------------------------
232One of the most frequent (and obvious) bugs with HD-audio is the
233silent output from either or both of a built-in speaker and a
234headphone jack.  In general, you should try a headphone output at
235first.  A speaker output often requires more additional controls like
236the external amplifier bits.  Thus a headphone output has a slightly
237better chance.
238
239Before making a bug report, double-check whether the mixer is set up
240correctly.  The recent version of snd-hda-intel driver provides mostly
241"Master" volume control as well as "Front" volume (where Front
242indicates the front-channels).  In addition, there can be individual
243"Headphone" and "Speaker" controls.
244
245Ditto for the speaker output.  There can be "External Amplifier"
246switch on some codecs.  Turn on this if present.
247
248Another related problem is the automatic mute of speaker output by
249headphone plugging.  This feature is implemented in most cases, but
250not on every preset model or codec-support code.
251
252In anyway, try a different model option if you have such a problem.
253Some other models may match better and give you more matching
254functionality.  If none of the available models works, send a bug
255report.  See the bug report section for details.
256
257If you are masochistic enough to debug the driver problem, note the
258following:
259
260* The speaker (and the headphone, too) output often requires the
261  external amplifier.  This can be set usually via EAPD verb or a
262  certain GPIO.  If the codec pin supports EAPD, you have a better
263  chance via SET_EAPD_BTL verb (0x70c).  On others, GPIO pin (mostly
264  it's either GPIO0 or GPIO1) may turn on/off EAPD.
265* Some Realtek codecs require special vendor-specific coefficients to
266  turn on the amplifier.  See patch_realtek.c.
267* IDT codecs may have extra power-enable/disable controls on each
268  analog pin.  See patch_sigmatel.c.
269* Very rare but some devices don't accept the pin-detection verb until
270  triggered.  Issuing GET_PIN_SENSE verb (0xf09) may result in the
271  codec-communication stall.  Some examples are found in
272  patch_realtek.c.
273
274
275Capture Problems
276----------------
277The capture problems are often because of missing setups of mixers.
278Thus, before submitting a bug report, make sure that you set up the
279mixer correctly.  For example, both "Capture Volume" and "Capture
280Switch" have to be set properly in addition to the right "Capture
281Source" or "Input Source" selection.  Some devices have "Mic Boost"
282volume or switch.
283
284When the PCM device is opened via "default" PCM (without pulse-audio
285plugin), you'll likely have "Digital Capture Volume" control as well.
286This is provided for the extra gain/attenuation of the signal in
287software, especially for the inputs without the hardware volume
288control such as digital microphones.  Unless really needed, this
289should be set to exactly 50%, corresponding to 0dB -- neither extra
290gain nor attenuation.  When you use "hw" PCM, i.e., a raw access PCM,
291this control will have no influence, though.
292
293It's known that some codecs / devices have fairly bad analog circuits,
294and the recorded sound contains a certain DC-offset.  This is no bug
295of the driver.
296
297Most of modern laptops have no analog CD-input connection.  Thus, the
298recording from CD input won't work in many cases although the driver
299provides it as the capture source.  Use CDDA instead.
300
301The automatic switching of the built-in and external mic per plugging
302is implemented on some codec models but not on every model.  Partly
303because of my laziness but mostly lack of testers.  Feel free to
304submit the improvement patch to the author.
305
306
307Direct Debugging
308----------------
309If no model option gives you a better result, and you are a tough guy
310to fight against evil, try debugging via hitting the raw HD-audio
311codec verbs to the device.  Some tools are available: hda-emu and
312hda-analyzer.  The detailed description is found in the sections
313below.  You'd need to enable hwdep for using these tools.  See "Kernel
314Configuration" section.
315
316
317Other Issues
318============
319
320Kernel Configuration
321--------------------
322In general, I recommend you to enable the sound debug option,
323``CONFIG_SND_DEBUG=y``, no matter whether you are debugging or not.
324This enables snd_printd() macro and others, and you'll get additional
325kernel messages at probing.
326
327In addition, you can enable ``CONFIG_SND_DEBUG_VERBOSE=y``.  But this
328will give you far more messages.  Thus turn this on only when you are
329sure to want it.
330
331Don't forget to turn on the appropriate ``CONFIG_SND_HDA_CODEC_*``
332options.  Note that each of them corresponds to the codec chip, not
333the controller chip.  Thus, even if lspci shows the Nvidia controller,
334you may need to choose the option for other vendors.  If you are
335unsure, just select all yes.
336
337``CONFIG_SND_HDA_HWDEP`` is a useful option for debugging the driver.
338When this is enabled, the driver creates hardware-dependent devices
339(one per each codec), and you have a raw access to the device via
340these device files.  For example, ``hwC0D2`` will be created for the
341codec slot #2 of the first card (#0).  For debug-tools such as
342hda-verb and hda-analyzer, the hwdep device has to be enabled.
343Thus, it'd be better to turn this on always.
344
345``CONFIG_SND_HDA_RECONFIG`` is a new option, and this depends on the
346hwdep option above.  When enabled, you'll have some sysfs files under
347the corresponding hwdep directory.  See "HD-audio reconfiguration"
348section below.
349
350``CONFIG_SND_HDA_POWER_SAVE`` option enables the power-saving feature.
351See "Power-saving" section below.
352
353
354Codec Proc-File
355---------------
356The codec proc-file is a treasure-chest for debugging HD-audio.
357It shows most of useful information of each codec widget.
358
359The proc file is located in /proc/asound/card*/codec#*, one file per
360each codec slot.  You can know the codec vendor, product id and
361names, the type of each widget, capabilities and so on.
362This file, however, doesn't show the jack sensing state, so far.  This
363is because the jack-sensing might be depending on the trigger state.
364
365This file will be picked up by the debug tools, and also it can be fed
366to the emulator as the primary codec information.  See the debug tools
367section below.
368
369This proc file can be also used to check whether the generic parser is
370used.  When the generic parser is used, the vendor/product ID name
371will appear as "Realtek ID 0262", instead of "Realtek ALC262".
372
373
374HD-Audio Reconfiguration
375------------------------
376This is an experimental feature to allow you re-configure the HD-audio
377codec dynamically without reloading the driver.  The following sysfs
378files are available under each codec-hwdep device directory (e.g.
379/sys/class/sound/hwC0D0):
380
381vendor_id
382    Shows the 32bit codec vendor-id hex number.  You can change the
383    vendor-id value by writing to this file.
384subsystem_id
385    Shows the 32bit codec subsystem-id hex number.  You can change the
386    subsystem-id value by writing to this file.
387revision_id
388    Shows the 32bit codec revision-id hex number.  You can change the
389    revision-id value by writing to this file.
390afg
391    Shows the AFG ID.  This is read-only.
392mfg
393    Shows the MFG ID.  This is read-only.
394name
395    Shows the codec name string.  Can be changed by writing to this
396    file.
397modelname
398    Shows the currently set ``model`` option.  Can be changed by writing
399    to this file.
400init_verbs
401    The extra verbs to execute at initialization.  You can add a verb by
402    writing to this file.  Pass three numbers: nid, verb and parameter
403    (separated with a space).
404hints
405    Shows / stores hint strings for codec parsers for any use.
406    Its format is ``key = value``.  For example, passing ``jack_detect = no``
407    will disable the jack detection of the machine completely.
408init_pin_configs
409    Shows the initial pin default config values set by BIOS.
410driver_pin_configs
411    Shows the pin default values set by the codec parser explicitly.
412    This doesn't show all pin values but only the changed values by
413    the parser.  That is, if the parser doesn't change the pin default
414    config values by itself, this will contain nothing.
415user_pin_configs
416    Shows the pin default config values to override the BIOS setup.
417    Writing this (with two numbers, NID and value) appends the new
418    value.  The given will be used instead of the initial BIOS value at
419    the next reconfiguration time.  Note that this config will override
420    even the driver pin configs, too.
421reconfig
422    Triggers the codec re-configuration.  When any value is written to
423    this file, the driver re-initialize and parses the codec tree
424    again.  All the changes done by the sysfs entries above are taken
425    into account.
426clear
427    Resets the codec, removes the mixer elements and PCM stuff of the
428    specified codec, and clear all init verbs and hints.
429
430For example, when you want to change the pin default configuration
431value of the pin widget 0x14 to 0x9993013f, and let the driver
432re-configure based on that state, run like below:
433::
434
435    # echo 0x14 0x9993013f > /sys/class/sound/hwC0D0/user_pin_configs
436    # echo 1 > /sys/class/sound/hwC0D0/reconfig
437
438
439Hint Strings
440------------
441The codec parser have several switches and adjustment knobs for
442matching better with the actual codec or device behavior.  Many of
443them can be adjusted dynamically via "hints" strings as mentioned in
444the section above.  For example, by passing ``jack_detect = no`` string
445via sysfs or a patch file, you can disable the jack detection, thus
446the codec parser will skip the features like auto-mute or mic
447auto-switch.  As a boolean value, either ``yes``, ``no``, ``true``, ``false``,
448``1`` or ``0`` can be passed.
449
450The generic parser supports the following hints:
451
452jack_detect (bool)
453    specify whether the jack detection is available at all on this
454    machine; default true
455inv_jack_detect (bool)
456    indicates that the jack detection logic is inverted
457trigger_sense (bool)
458    indicates that the jack detection needs the explicit call of
459    AC_VERB_SET_PIN_SENSE verb
460inv_eapd (bool)
461    indicates that the EAPD is implemented in the inverted logic
462pcm_format_first (bool)
463    sets the PCM format before the stream tag and channel ID
464sticky_stream (bool)
465    keep the PCM format, stream tag and ID as long as possible;
466    default true
467spdif_status_reset (bool)
468    reset the SPDIF status bits at each time the SPDIF stream is set
469    up
470pin_amp_workaround (bool)
471    the output pin may have multiple amp values
472single_adc_amp (bool)
473    ADCs can have only single input amps
474auto_mute (bool)
475    enable/disable the headphone auto-mute feature; default true
476auto_mic (bool)
477    enable/disable the mic auto-switch feature; default true
478line_in_auto_switch (bool)
479    enable/disable the line-in auto-switch feature; default false
480need_dac_fix (bool)
481    limits the DACs depending on the channel count
482primary_hp (bool)
483    probe headphone jacks as the primary outputs; default true
484multi_io (bool)
485    try probing multi-I/O config (e.g. shared line-in/surround,
486    mic/clfe jacks)
487multi_cap_vol (bool)
488    provide multiple capture volumes
489inv_dmic_split (bool)
490    provide split internal mic volume/switch for phase-inverted
491    digital mics
492indep_hp (bool)
493    provide the independent headphone PCM stream and the corresponding
494    mixer control, if available
495add_stereo_mix_input (bool)
496    add the stereo mix (analog-loopback mix) to the input mux if
497    available
498add_jack_modes (bool)
499    add "xxx Jack Mode" enum controls to each I/O jack for allowing to
500    change the headphone amp and mic bias VREF capabilities
501power_save_node (bool)
502    advanced power management for each widget, controlling the power
503    sate (D0/D3) of each widget node depending on the actual pin and
504    stream states
505power_down_unused (bool)
506    power down the unused widgets, a subset of power_save_node, and
507    will be dropped in future
508add_hp_mic (bool)
509    add the headphone to capture source if possible
510hp_mic_detect (bool)
511    enable/disable the hp/mic shared input for a single built-in mic
512    case; default true
513vmaster (bool)
514    enable/disable the virtual Master control; default true
515mixer_nid (int)
516    specifies the widget NID of the analog-loopback mixer
517
518
519Early Patching
520--------------
521When ``CONFIG_SND_HDA_PATCH_LOADER=y`` is set, you can pass a "patch"
522as a firmware file for modifying the HD-audio setup before
523initializing the codec.  This can work basically like the
524reconfiguration via sysfs in the above, but it does it before the
525first codec configuration.
526
527A patch file is a plain text file which looks like below:
528
529::
530
531    [codec]
532    0x12345678 0xabcd1234 2
533
534    [model]
535    auto
536
537    [pincfg]
538    0x12 0x411111f0
539
540    [verb]
541    0x20 0x500 0x03
542    0x20 0x400 0xff
543
544    [hint]
545    jack_detect = no
546
547
548The file needs to have a line ``[codec]``.  The next line should contain
549three numbers indicating the codec vendor-id (0x12345678 in the
550example), the codec subsystem-id (0xabcd1234) and the address (2) of
551the codec.  The rest patch entries are applied to this specified codec
552until another codec entry is given.  Passing 0 or a negative number to
553the first or the second value will make the check of the corresponding
554field be skipped.  It'll be useful for really broken devices that don't
555initialize SSID properly.
556
557The ``[model]`` line allows to change the model name of the each codec.
558In the example above, it will be changed to model=auto.
559Note that this overrides the module option.
560
561After the ``[pincfg]`` line, the contents are parsed as the initial
562default pin-configurations just like ``user_pin_configs`` sysfs above.
563The values can be shown in user_pin_configs sysfs file, too.
564
565Similarly, the lines after ``[verb]`` are parsed as ``init_verbs``
566sysfs entries, and the lines after ``[hint]`` are parsed as ``hints``
567sysfs entries, respectively.
568
569Another example to override the codec vendor id from 0x12345678 to
5700xdeadbeef is like below:
571::
572
573    [codec]
574    0x12345678 0xabcd1234 2
575
576    [vendor_id]
577    0xdeadbeef
578
579
580In the similar way, you can override the codec subsystem_id via
581``[subsystem_id]``, the revision id via ``[revision_id]`` line.
582Also, the codec chip name can be rewritten via ``[chip_name]`` line.
583::
584
585    [codec]
586    0x12345678 0xabcd1234 2
587
588    [subsystem_id]
589    0xffff1111
590
591    [revision_id]
592    0x10
593
594    [chip_name]
595    My-own NEWS-0002
596
597
598The hd-audio driver reads the file via request_firmware().  Thus,
599a patch file has to be located on the appropriate firmware path,
600typically, /lib/firmware.  For example, when you pass the option
601``patch=hda-init.fw``, the file /lib/firmware/hda-init.fw must be
602present.
603
604The patch module option is specific to each card instance, and you
605need to give one file name for each instance, separated by commas.
606For example, if you have two cards, one for an on-board analog and one
607for an HDMI video board, you may pass patch option like below:
608::
609
610    options snd-hda-intel patch=on-board-patch,hdmi-patch
611
612
613Power-Saving
614------------
615The power-saving is a kind of auto-suspend of the device.  When the
616device is inactive for a certain time, the device is automatically
617turned off to save the power.  The time to go down is specified via
618``power_save`` module option, and this option can be changed dynamically
619via sysfs.
620
621The power-saving won't work when the analog loopback is enabled on
622some codecs.  Make sure that you mute all unneeded signal routes when
623you want the power-saving.
624
625The power-saving feature might cause audible click noises at each
626power-down/up depending on the device.  Some of them might be
627solvable, but some are hard, I'm afraid.  Some distros such as
628openSUSE enables the power-saving feature automatically when the power
629cable is unplugged.  Thus, if you hear noises, suspect first the
630power-saving.  See /sys/module/snd_hda_intel/parameters/power_save to
631check the current value.  If it's non-zero, the feature is turned on.
632
633The recent kernel supports the runtime PM for the HD-audio controller
634chip, too.  It means that the HD-audio controller is also powered up /
635down dynamically.  The feature is enabled only for certain controller
636chips like Intel LynxPoint.  You can enable/disable this feature
637forcibly by setting ``power_save_controller`` option, which is also
638available at /sys/module/snd_hda_intel/parameters directory.
639
640
641Tracepoints
642-----------
643The hd-audio driver gives a few basic tracepoints.
644``hda:hda_send_cmd`` traces each CORB write while ``hda:hda_get_response``
645traces the response from RIRB (only when read from the codec driver).
646``hda:hda_bus_reset`` traces the bus-reset due to fatal error, etc,
647``hda:hda_unsol_event`` traces the unsolicited events, and
648``hda:hda_power_down`` and ``hda:hda_power_up`` trace the power down/up
649via power-saving behavior.
650
651Enabling all tracepoints can be done like
652::
653
654    # echo 1 > /sys/kernel/debug/tracing/events/hda/enable
655
656then after some commands, you can traces from
657/sys/kernel/debug/tracing/trace file.  For example, when you want to
658trace what codec command is sent, enable the tracepoint like:
659::
660
661    # cat /sys/kernel/debug/tracing/trace
662    # tracer: nop
663    #
664    #       TASK-PID    CPU#    TIMESTAMP  FUNCTION
665    #          | |       |          |         |
666	   <...>-7807  [002] 105147.774889: hda_send_cmd: [0:0] val=e3a019
667	   <...>-7807  [002] 105147.774893: hda_send_cmd: [0:0] val=e39019
668	   <...>-7807  [002] 105147.999542: hda_send_cmd: [0:0] val=e3a01a
669	   <...>-7807  [002] 105147.999543: hda_send_cmd: [0:0] val=e3901a
670	   <...>-26764 [001] 349222.837143: hda_send_cmd: [0:0] val=e3a019
671	   <...>-26764 [001] 349222.837148: hda_send_cmd: [0:0] val=e39019
672	   <...>-26764 [001] 349223.058539: hda_send_cmd: [0:0] val=e3a01a
673	   <...>-26764 [001] 349223.058541: hda_send_cmd: [0:0] val=e3901a
674
675Here ``[0:0]`` indicates the card number and the codec address, and
676``val`` shows the value sent to the codec, respectively.  The value is
677a packed value, and you can decode it via hda-decode-verb program
678included in hda-emu package below.  For example, the value e3a019 is
679to set the left output-amp value to 25.
680::
681
682    % hda-decode-verb 0xe3a019
683    raw value = 0x00e3a019
684    cid = 0, nid = 0x0e, verb = 0x3a0, parm = 0x19
685    raw value: verb = 0x3a0, parm = 0x19
686    verbname = set_amp_gain_mute
687    amp raw val = 0xa019
688    output, left, idx=0, mute=0, val=25
689
690
691Development Tree
692----------------
693The latest development codes for HD-audio are found on sound git tree:
694
695* git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git
696
697The master branch or for-next branches can be used as the main
698development branches in general while the development for the current
699and next kernels are found in for-linus and for-next branches,
700respectively.
701
702
703Sending a Bug Report
704--------------------
705If any model or module options don't work for your device, it's time
706to send a bug report to the developers.  Give the following in your
707bug report:
708
709* Hardware vendor, product and model names
710* Kernel version (and ALSA-driver version if you built externally)
711* ``alsa-info.sh`` output; run with ``--no-upload`` option.  See the
712  section below about alsa-info
713
714If it's a regression, at best, send alsa-info outputs of both working
715and non-working kernels.  This is really helpful because we can
716compare the codec registers directly.
717
718Send a bug report either the following:
719
720kernel-bugzilla
721    https://bugzilla.kernel.org/
722alsa-devel ML
723    alsa-devel@alsa-project.org
724
725
726Debug Tools
727===========
728
729This section describes some tools available for debugging HD-audio
730problems.
731
732alsa-info
733---------
734The script ``alsa-info.sh`` is a very useful tool to gather the audio
735device information.  It's included in alsa-utils package.  The latest
736version can be found on git repository:
737
738* git://git.alsa-project.org/alsa-utils.git
739
740The script can be fetched directly from the following URL, too:
741
742* https://www.alsa-project.org/alsa-info.sh
743
744Run this script as root, and it will gather the important information
745such as the module lists, module parameters, proc file contents
746including the codec proc files, mixer outputs and the control
747elements.  As default, it will store the information onto a web server
748on alsa-project.org.  But, if you send a bug report, it'd be better to
749run with ``--no-upload`` option, and attach the generated file.
750
751There are some other useful options.  See ``--help`` option output for
752details.
753
754When a probe error occurs or when the driver obviously assigns a
755mismatched model, it'd be helpful to load the driver with
756``probe_only=1`` option (at best after the cold reboot) and run
757alsa-info at this state.  With this option, the driver won't configure
758the mixer and PCM but just tries to probe the codec slot.  After
759probing, the proc file is available, so you can get the raw codec
760information before modified by the driver.  Of course, the driver
761isn't usable with ``probe_only=1``.  But you can continue the
762configuration via hwdep sysfs file if hda-reconfig option is enabled.
763Using ``probe_only`` mask 2 skips the reset of HDA codecs (use
764``probe_only=3`` as module option). The hwdep interface can be used
765to determine the BIOS codec initialization.
766
767
768hda-verb
769--------
770hda-verb is a tiny program that allows you to access the HD-audio
771codec directly.  You can execute a raw HD-audio codec verb with this.
772This program accesses the hwdep device, thus you need to enable the
773kernel config ``CONFIG_SND_HDA_HWDEP=y`` beforehand.
774
775The hda-verb program takes four arguments: the hwdep device file, the
776widget NID, the verb and the parameter.  When you access to the codec
777on the slot 2 of the card 0, pass /dev/snd/hwC0D2 to the first
778argument, typically.  (However, the real path name depends on the
779system.)
780
781The second parameter is the widget number-id to access.  The third
782parameter can be either a hex/digit number or a string corresponding
783to a verb.  Similarly, the last parameter is the value to write, or
784can be a string for the parameter type.
785
786::
787
788    % hda-verb /dev/snd/hwC0D0 0x12 0x701 2
789    nid = 0x12, verb = 0x701, param = 0x2
790    value = 0x0
791
792    % hda-verb /dev/snd/hwC0D0 0x0 PARAMETERS VENDOR_ID
793    nid = 0x0, verb = 0xf00, param = 0x0
794    value = 0x10ec0262
795
796    % hda-verb /dev/snd/hwC0D0 2 set_a 0xb080
797    nid = 0x2, verb = 0x300, param = 0xb080
798    value = 0x0
799
800
801Although you can issue any verbs with this program, the driver state
802won't be always updated.  For example, the volume values are usually
803cached in the driver, and thus changing the widget amp value directly
804via hda-verb won't change the mixer value.
805
806The hda-verb program is included now in alsa-tools:
807
808* git://git.alsa-project.org/alsa-tools.git
809
810Also, the old stand-alone package is found in the ftp directory:
811
812* ftp://ftp.suse.com/pub/people/tiwai/misc/
813
814Also a git repository is available:
815
816* git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/hda-verb.git
817
818See README file in the tarball for more details about hda-verb
819program.
820
821
822hda-analyzer
823------------
824hda-analyzer provides a graphical interface to access the raw HD-audio
825control, based on pyGTK2 binding.  It's a more powerful version of
826hda-verb.  The program gives you an easy-to-use GUI stuff for showing
827the widget information and adjusting the amp values, as well as the
828proc-compatible output.
829
830The hda-analyzer:
831
832* https://git.alsa-project.org/?p=alsa.git;a=tree;f=hda-analyzer
833
834is a part of alsa.git repository in alsa-project.org:
835
836* git://git.alsa-project.org/alsa.git
837
838Codecgraph
839----------
840Codecgraph is a utility program to generate a graph and visualizes the
841codec-node connection of a codec chip.  It's especially useful when
842you analyze or debug a codec without a proper datasheet.  The program
843parses the given codec proc file and converts to SVG via graphiz
844program.
845
846The tarball and GIT trees are found in the web page at:
847
848* http://helllabs.org/codecgraph/
849
850
851hda-emu
852-------
853hda-emu is an HD-audio emulator.  The main purpose of this program is
854to debug an HD-audio codec without the real hardware.  Thus, it
855doesn't emulate the behavior with the real audio I/O, but it just
856dumps the codec register changes and the ALSA-driver internal changes
857at probing and operating the HD-audio driver.
858
859The program requires a codec proc-file to simulate.  Get a proc file
860for the target codec beforehand, or pick up an example codec from the
861codec proc collections in the tarball.  Then, run the program with the
862proc file, and the hda-emu program will start parsing the codec file
863and simulates the HD-audio driver:
864
865::
866
867    % hda-emu codecs/stac9200-dell-d820-laptop
868    # Parsing..
869    hda_codec: Unknown model for STAC9200, using BIOS defaults
870    hda_codec: pin nid 08 bios pin config 40c003fa
871    ....
872
873
874The program gives you only a very dumb command-line interface.  You
875can get a proc-file dump at the current state, get a list of control
876(mixer) elements, set/get the control element value, simulate the PCM
877operation, the jack plugging simulation, etc.
878
879The program is found in the git repository below:
880
881* git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/hda-emu.git
882
883See README file in the repository for more details about hda-emu
884program.
885
886
887hda-jack-retask
888---------------
889hda-jack-retask is a user-friendly GUI program to manipulate the
890HD-audio pin control for jack retasking.  If you have a problem about
891the jack assignment, try this program and check whether you can get
892useful results.  Once when you figure out the proper pin assignment,
893it can be fixed either in the driver code statically or via passing a
894firmware patch file (see "Early Patching" section).
895
896The program is included in alsa-tools now:
897
898* git://git.alsa-project.org/alsa-tools.git
899