blob: 55aab1168236f27ccb80c380f12c580095ac9fd0 [file] [log] [blame]
Takashi Iwaia7fe49b2008-12-03 18:26:35 +01001MORE NOTES ON HD-AUDIO DRIVER
2=============================
3 Takashi Iwai <tiwai@suse.de>
4
5
6GENERAL
7-------
8
9HD-audio is the new standard on-board audio component on modern PCs
10after AC97. Although Linux has been supporting HD-audio since long
11time ago, there are often problems with new machines. A part of the
Takashi Iwaid2afbe72008-12-10 09:28:15 +010012problem is broken BIOS, and the rest is the driver implementation.
13This document explains the brief trouble-shooting and debugging
14methods for the HD-audio hardware.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010015
16The HD-audio component consists of two parts: the controller chip and
17the codec chips on the HD-audio bus. Linux provides a single driver
Takashi Iwaid2afbe72008-12-10 09:28:15 +010018for all controllers, snd-hda-intel. Although the driver name contains
19a word of a well-known harware vendor, it's not specific to it but for
20all controller chips by other companies. Since the HD-audio
21controllers are supposed to be compatible, the single snd-hda-driver
22should work in most cases. But, not surprisingly, there are known
23bugs and issues specific to each controller type. The snd-hda-intel
24driver has a bunch of workarounds for these as described below.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010025
26A controller may have multiple codecs. Usually you have one audio
Takashi Iwaid2afbe72008-12-10 09:28:15 +010027codec and optionally one modem codec. In theory, there might be
28multiple audio codecs, e.g. for analog and digital outputs, and the
29driver might not work properly because of conflict of mixer elements.
30This should be fixed in future if such hardware really exists.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010031
32The snd-hda-intel driver has several different codec parsers depending
33on the codec. It has a generic parser as a fallback, but this
34functionality is fairly limited until now. Instead of the generic
35parser, usually the codec-specific parser (coded in patch_*.c) is used
36for the codec-specific implementations. The details about the
37codec-specific problems are explained in the later sections.
38
39If you are interested in the deep debugging of HD-audio, read the
40HD-audio specification at first. The specification is found on
41Intel's web page, for example:
42
43- http://www.intel.com/standards/hdaudio/
44
45
46HD-AUDIO CONTROLLER
47-------------------
48
49DMA-Position Problem
50~~~~~~~~~~~~~~~~~~~~
51The most common problem of the controller is the inaccurate DMA
52pointer reporting. The DMA pointer for playback and capture can be
53read in two ways, either via a LPIB register or via a position-buffer
Takashi Iwaid2afbe72008-12-10 09:28:15 +010054map. As default the driver tries to read from the io-mapped
55position-buffer, and falls back to LPIB if the position-buffer appears
56dead. However, this detection isn't perfect on some devices. In such
57a case, you can change the default method via `position_fix` option.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010058
59`position_fix=1` means to use LPIB method explicitly.
60`position_fix=2` means to use the position-buffer. 0 is the default
Takashi Iwaid2afbe72008-12-10 09:28:15 +010061value, the automatic check and fallback to LPIB as described in the
62above. If you get a problem of repeated sounds, this option might
63help.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010064
65In addition to that, every controller is known to be broken regarding
66the wake-up timing. It wakes up a few samples before actually
67processing the data on the buffer. This caused a lot of problems, for
68example, with ALSA dmix or JACK. Since 2.6.27 kernel, the driver puts
69an artificial delay to the wake up timing. This delay is controlled
70via `bdl_pos_adj` option.
71
72When `bdl_pos_adj` is a negative value (as default), it's assigned to
73an appropriate value depending on the controller chip. For Intel
Takashi Iwaid2afbe72008-12-10 09:28:15 +010074chips, it'd be 1 while it'd be 32 for others. Usually this works.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010075Only in case it doesn't work and you get warning messages, you should
Takashi Iwaid2afbe72008-12-10 09:28:15 +010076change this parameter to other values.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010077
78
79Codec-Probing Problem
80~~~~~~~~~~~~~~~~~~~~~
81A less often but a more severe problem is the codec probing. When
82BIOS reports the available codec slots wrongly, the driver gets
83confused and tries to access the non-existing codec slot. This often
Takashi Iwaid2afbe72008-12-10 09:28:15 +010084results in the total screw-up, and destructs the further communication
85with the codec chips. The symptom appears usually as error messages
86like:
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010087------------------------------------------------------------------------
Takashi Iwaid2afbe72008-12-10 09:28:15 +010088 hda_intel: azx_get_response timeout, switching to polling mode:
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010089 last cmd=0x12345678
Takashi Iwaid2afbe72008-12-10 09:28:15 +010090 hda_intel: azx_get_response timeout, switching to single_cmd mode:
Takashi Iwaia7fe49b2008-12-03 18:26:35 +010091 last cmd=0x12345678
92------------------------------------------------------------------------
93
94The first line is a warning, and this is usually relatively harmless.
95It means that the codec response isn't notified via an IRQ. The
96driver uses explicit polling method to read the response. It gives
97very slight CPU overhead, but you'd unlikely notice it.
98
99The second line is, however, a fatal error. If this happens, usually
100it means that something is really wrong. Most likely you are
101accessing a non-existing codec slot.
102
103Thus, if the second error message appears, try to narrow the probed
104codec slots via `probe_mask` option. It's a bitmask, and each bit
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100105corresponds to the codec slot. For example, to probe only the first
106slot, pass `probe_mask=1`. For the first and the third slots, pass
107`probe_mask=5` (where 5 = 1 | 4), and so on.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100108
109Since 2.6.29 kernel, the driver has a more robust probing method, so
110this error might happen rarely, though.
111
Takashi Iwaiae374d62009-02-13 08:33:55 +0100112On a machine with a broken BIOS, sometimes you need to force the
113driver to probe the codec slots the hardware doesn't report for use.
114In such a case, turn the bit 8 (0x100) of `probe_mask` option on.
115Then the rest 8 bits are passed as the codec slots to probe
116unconditionally. For example, `probe_mask=0x103` will force to probe
117the codec slots 0 and 1 no matter what the hardware reports.
118
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100119
120Interrupt Handling
121~~~~~~~~~~~~~~~~~~
122In rare but some cases, the interrupt isn't properly handled as
123default. You would notice this by the DMA transfer error reported by
124ALSA PCM core, for example. Using MSI might help in such a case.
125Pass `enable_msi=1` option for enabling MSI.
126
127
128HD-AUDIO CODEC
129--------------
130
131Model Option
132~~~~~~~~~~~~
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100133The most common problem regarding the HD-audio driver is the
134unsupported codec features or the mismatched device configuration.
135Most of codec-specific code has several preset models, either to
136override the BIOS setup or to provide more comprehensive features.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100137
138The driver checks PCI SSID and looks through the static configuration
139table until any matching entry is found. If you have a new machine,
140you may see a message like below:
141------------------------------------------------------------------------
142 hda_codec: Unknown model for ALC880, trying auto-probe from BIOS...
143------------------------------------------------------------------------
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100144Even if you see such a message, DON'T PANIC. Take a deep breath and
145keep your towel. First of all, it's an informational message, no
146warning, no error. This means that the PCI SSID of your device isn't
147listed in the known preset model (white-)list. But, this doesn't mean
148that the driver is broken. Many codec-drivers provide the automatic
149configuration mechanism based on the BIOS setup.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100150
151The HD-audio codec has usually "pin" widgets, and BIOS sets the default
152configuration of each pin, which indicates the location, the
153connection type, the jack color, etc. The HD-audio driver can guess
154the right connection judging from these default configuration values.
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100155However -- some codec-support codes, such as patch_analog.c, don't
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100156support the automatic probing (yet as of 2.6.28). And, BIOS is often,
157yes, pretty often broken. It sets up wrong values and screws up the
158driver.
159
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100160The preset model is provided basically to overcome such a situation.
161When the matching preset model is found in the white-list, the driver
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100162assumes the static configuration of that preset and builds the mixer
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100163elements and PCM streams based on the static information. Thus, if
164you have a newer machine with a slightly different PCI SSID from the
165existing one, you may have a good chance to re-use the same model.
166You can pass the `model` option to specify the preset model instead of
167PCI SSID look-up.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100168
169What `model` option values are available depends on the codec chip.
170Check your codec chip from the codec proc file (see "Codec Proc-File"
171section below). It will show the vendor/product name of your codec
Justin Mattock83b20862009-04-14 14:31:21 -0700172chip. Then, see Documentation/sound/alsa/HD-Audio-Models.txt file,
Takashi Iwai692f9042008-12-19 12:44:46 +0100173the section of HD-audio driver. You can find a list of codecs
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100174and `model` options belonging to each codec. For example, for Realtek
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100175ALC262 codec chip, pass `model=ultra` for devices that are compatible
176with Samsung Q1 Ultra.
177
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100178Thus, the first thing you can do for any brand-new, unsupported and
179non-working HD-audio hardware is to check HD-audio codec and several
Justin Mattock83b20862009-04-14 14:31:21 -0700180different `model` option values. If you have any luck, some of them
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100181might suit with your device well.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100182
183Some codecs such as ALC880 have a special model option `model=test`.
184This configures the driver to provide as many mixer controls as
185possible for every single pin feature except for the unsolicited
186events (and maybe some other specials). Adjust each mixer element and
187try the I/O in the way of trial-and-error until figuring out the whole
188I/O pin mappings.
189
190Note that `model=generic` has a special meaning. It means to use the
191generic parser regardless of the codec. Usually the codec-specific
192parser is much better than the generic parser (as now). Thus this
193option is more about the debugging purpose.
194
195
196Speaker and Headphone Output
197~~~~~~~~~~~~~~~~~~~~~~~~~~~~
198One of the most frequent (and obvious) bugs with HD-audio is the
199silent output from either or both of a built-in speaker and a
200headphone jack. In general, you should try a headphone output at
201first. A speaker output often requires more additional controls like
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100202the external amplifier bits. Thus a headphone output has a slightly
203better chance.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100204
205Before making a bug report, double-check whether the mixer is set up
206correctly. The recent version of snd-hda-intel driver provides mostly
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100207"Master" volume control as well as "Front" volume (where Front
208indicates the front-channels). In addition, there can be individual
209"Headphone" and "Speaker" controls.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100210
211Ditto for the speaker output. There can be "External Amplifier"
212switch on some codecs. Turn on this if present.
213
214Another related problem is the automatic mute of speaker output by
215headphone plugging. This feature is implemented in most cases, but
216not on every preset model or codec-support code.
217
218In anyway, try a different model option if you have such a problem.
219Some other models may match better and give you more matching
220functionality. If none of the available models works, send a bug
221report. See the bug report section for details.
222
223If you are masochistic enough to debug the driver problem, note the
224following:
225
226- The speaker (and the headphone, too) output often requires the
227 external amplifier. This can be set usually via EAPD verb or a
228 certain GPIO. If the codec pin supports EAPD, you have a better
229 chance via SET_EAPD_BTL verb (0x70c). On others, GPIO pin (mostly
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100230 it's either GPIO0 or GPIO1) may turn on/off EAPD.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100231- Some Realtek codecs require special vendor-specific coefficients to
232 turn on the amplifier. See patch_realtek.c.
233- IDT codecs may have extra power-enable/disable controls on each
234 analog pin. See patch_sigmatel.c.
235- Very rare but some devices don't accept the pin-detection verb until
236 triggered. Issuing GET_PIN_SENSE verb (0xf09) may result in the
237 codec-communication stall. Some examples are found in
238 patch_realtek.c.
239
240
241Capture Problems
242~~~~~~~~~~~~~~~~
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100243The capture problems are often because of missing setups of mixers.
244Thus, before submitting a bug report, make sure that you set up the
245mixer correctly. For example, both "Capture Volume" and "Capture
246Switch" have to be set properly in addition to the right "Capture
247Source" or "Input Source" selection. Some devices have "Mic Boost"
248volume or switch.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100249
250When the PCM device is opened via "default" PCM (without pulse-audio
251plugin), you'll likely have "Digital Capture Volume" control as well.
252This is provided for the extra gain/attenuation of the signal in
253software, especially for the inputs without the hardware volume
254control such as digital microphones. Unless really needed, this
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100255should be set to exactly 50%, corresponding to 0dB -- neither extra
256gain nor attenuation. When you use "hw" PCM, i.e., a raw access PCM,
257this control will have no influence, though.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100258
259It's known that some codecs / devices have fairly bad analog circuits,
260and the recorded sound contains a certain DC-offset. This is no bug
261of the driver.
262
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100263Most of modern laptops have no analog CD-input connection. Thus, the
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100264recording from CD input won't work in many cases although the driver
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100265provides it as the capture source. Use CDDA instead.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100266
267The automatic switching of the built-in and external mic per plugging
268is implemented on some codec models but not on every model. Partly
269because of my laziness but mostly lack of testers. Feel free to
270submit the improvement patch to the author.
271
272
273Direct Debugging
274~~~~~~~~~~~~~~~~
Takashi Iwaif8bbd062008-12-11 13:12:59 +0100275If no model option gives you a better result, and you are a tough guy
Takashi Iwai623b9f62008-12-11 07:44:18 +0100276to fight against evil, try debugging via hitting the raw HD-audio
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100277codec verbs to the device. Some tools are available: hda-emu and
278hda-analyzer. The detailed description is found in the sections
279below. You'd need to enable hwdep for using these tools. See "Kernel
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100280Configuration" section.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100281
282
283OTHER ISSUES
284------------
285
286Kernel Configuration
287~~~~~~~~~~~~~~~~~~~~
288In general, I recommend you to enable the sound debug option,
289`CONFIG_SND_DEBUG=y`, no matter whether you are debugging or not.
290This enables snd_printd() macro and others, and you'll get additional
291kernel messages at probing.
292
293In addition, you can enable `CONFIG_SND_DEBUG_VERBOSE=y`. But this
294will give you far more messages. Thus turn this on only when you are
295sure to want it.
296
297Don't forget to turn on the appropriate `CONFIG_SND_HDA_CODEC_*`
298options. Note that each of them corresponds to the codec chip, not
299the controller chip. Thus, even if lspci shows the Nvidia controller,
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100300you may need to choose the option for other vendors. If you are
301unsure, just select all yes.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100302
303`CONFIG_SND_HDA_HWDEP` is a useful option for debugging the driver.
304When this is enabled, the driver creates hardware-dependent devices
305(one per each codec), and you have a raw access to the device via
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100306these device files. For example, `hwC0D2` will be created for the
307codec slot #2 of the first card (#0). For debug-tools such as
308hda-verb and hda-analyzer, the hwdep device has to be enabled.
309Thus, it'd be better to turn this on always.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100310
311`CONFIG_SND_HDA_RECONFIG` is a new option, and this depends on the
312hwdep option above. When enabled, you'll have some sysfs files under
313the corresponding hwdep directory. See "HD-audio reconfiguration"
314section below.
315
316`CONFIG_SND_HDA_POWER_SAVE` option enables the power-saving feature.
317See "Power-saving" section below.
318
319
320Codec Proc-File
321~~~~~~~~~~~~~~~
322The codec proc-file is a treasure-chest for debugging HD-audio.
323It shows most of useful information of each codec widget.
324
325The proc file is located in /proc/asound/card*/codec#*, one file per
326each codec slot. You can know the codec vendor, product id and
327names, the type of each widget, capabilities and so on.
328This file, however, doesn't show the jack sensing state, so far. This
329is because the jack-sensing might be depending on the trigger state.
330
331This file will be picked up by the debug tools, and also it can be fed
332to the emulator as the primary codec information. See the debug tools
333section below.
334
335This proc file can be also used to check whether the generic parser is
336used. When the generic parser is used, the vendor/product ID name
337will appear as "Realtek ID 0262", instead of "Realtek ALC262".
338
339
340HD-Audio Reconfiguration
341~~~~~~~~~~~~~~~~~~~~~~~~
342This is an experimental feature to allow you re-configure the HD-audio
343codec dynamically without reloading the driver. The following sysfs
344files are available under each codec-hwdep device directory (e.g.
345/sys/class/sound/hwC0D0):
346
347vendor_id::
348 Shows the 32bit codec vendor-id hex number. You can change the
349 vendor-id value by writing to this file.
350subsystem_id::
351 Shows the 32bit codec subsystem-id hex number. You can change the
352 subsystem-id value by writing to this file.
353revision_id::
354 Shows the 32bit codec revision-id hex number. You can change the
355 revision-id value by writing to this file.
356afg::
357 Shows the AFG ID. This is read-only.
358mfg::
359 Shows the MFG ID. This is read-only.
360name::
361 Shows the codec name string. Can be changed by writing to this
362 file.
363modelname::
364 Shows the currently set `model` option. Can be changed by writing
365 to this file.
366init_verbs::
367 The extra verbs to execute at initialization. You can add a verb by
Takashi Iwaid02b1f32009-03-02 17:34:51 +0100368 writing to this file. Pass three numbers: nid, verb and parameter
369 (separated with a space).
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100370hints::
Takashi Iwaid02b1f32009-03-02 17:34:51 +0100371 Shows / stores hint strings for codec parsers for any use.
372 Its format is `key = value`. For example, passing `hp_detect = yes`
373 to IDT/STAC codec parser will result in the disablement of the
374 headphone detection.
Takashi Iwaif1085c42009-02-20 14:50:35 +0100375init_pin_configs::
376 Shows the initial pin default config values set by BIOS.
Takashi Iwai346ff702009-02-23 09:42:57 +0100377driver_pin_configs::
Takashi Iwaif1085c42009-02-20 14:50:35 +0100378 Shows the pin default values set by the codec parser explicitly.
379 This doesn't show all pin values but only the changed values by
380 the parser. That is, if the parser doesn't change the pin default
381 config values by itself, this will contain nothing.
Takashi Iwai346ff702009-02-23 09:42:57 +0100382user_pin_configs::
383 Shows the pin default config values to override the BIOS setup.
384 Writing this (with two numbers, NID and value) appends the new
385 value. The given will be used instead of the initial BIOS value at
Takashi Iwai5e7b8e02009-02-23 09:45:59 +0100386 the next reconfiguration time. Note that this config will override
387 even the driver pin configs, too.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100388reconfig::
389 Triggers the codec re-configuration. When any value is written to
390 this file, the driver re-initialize and parses the codec tree
391 again. All the changes done by the sysfs entries above are taken
392 into account.
393clear::
394 Resets the codec, removes the mixer elements and PCM stuff of the
395 specified codec, and clear all init verbs and hints.
396
Takashi Iwai39c28712009-02-23 14:14:51 +0100397For example, when you want to change the pin default configuration
398value of the pin widget 0x14 to 0x9993013f, and let the driver
399re-configure based on that state, run like below:
400------------------------------------------------------------------------
401 # echo 0x14 0x9993013f > /sys/class/sound/hwC0D0/user_pin_configs
402 # echo 1 > /sys/class/sound/hwC0D0/reconfig
403------------------------------------------------------------------------
404
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100405
Takashi Iwai76824822009-06-17 17:17:18 +0200406Early Patching
407~~~~~~~~~~~~~~
408When CONFIG_SND_HDA_PATCH_LOADER=y is set, you can pass a "patch" as a
409firmware file for modifying the HD-audio setup before initializing the
410codec. This can work basically like the reconfiguration via sysfs in
411the above, but it does it before the first codec configuration.
412
Takashi Iwai1e7b8c82009-06-17 17:30:54 +0200413A patch file is a plain text file which looks like below:
Takashi Iwai76824822009-06-17 17:17:18 +0200414
415------------------------------------------------------------------------
416 [codec]
417 0x12345678 0xabcd1234 2
418
419 [model]
420 auto
421
422 [pincfg]
423 0x12 0x411111f0
424
425 [verb]
426 0x20 0x500 0x03
427 0x20 0x400 0xff
428
429 [hint]
430 hp_detect = yes
431------------------------------------------------------------------------
432
433The file needs to have a line `[codec]`. The next line should contain
434three numbers indicating the codec vendor-id (0x12345678 in the
435example), the codec subsystem-id (0xabcd1234) and the address (2) of
436the codec. The rest patch entries are applied to this specified codec
437until another codec entry is given.
438
439The `[model]` line allows to change the model name of the each codec.
440In the example above, it will be changed to model=auto.
441Note that this overrides the module option.
442
443After the `[pincfg]` line, the contents are parsed as the initial
444default pin-configurations just like `user_pin_configs` sysfs above.
445The values can be shown in user_pin_configs sysfs file, too.
446
447Similarly, the lines after `[verb]` are parsed as `init_verbs`
448sysfs entries, and the lines after `[hint]` are parsed as `hints`
449sysfs entries, respectively.
450
Takashi Iwai1e7b8c82009-06-17 17:30:54 +0200451The hd-audio driver reads the file via request_firmware(). Thus,
452a patch file has to be located on the appropriate firmware path,
453typically, /lib/firmware. For example, when you pass the option
454`patch=hda-init.fw`, the file /lib/firmware/hda-init-fw must be
455present.
456
457The patch module option is specific to each card instance, and you
458need to give one file name for each instance, separated by commas.
459For example, if you have two cards, one for an on-board analog and one
460for an HDMI video board, you may pass patch option like below:
461------------------------------------------------------------------------
462 options snd-hda-intel patch=on-board-patch,hdmi-patch
463------------------------------------------------------------------------
464
Takashi Iwai76824822009-06-17 17:17:18 +0200465
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100466Power-Saving
467~~~~~~~~~~~~
468The power-saving is a kind of auto-suspend of the device. When the
469device is inactive for a certain time, the device is automatically
470turned off to save the power. The time to go down is specified via
471`power_save` module option, and this option can be changed dynamically
472via sysfs.
473
474The power-saving won't work when the analog loopback is enabled on
475some codecs. Make sure that you mute all unneeded signal routes when
476you want the power-saving.
477
478The power-saving feature might cause audible click noises at each
479power-down/up depending on the device. Some of them might be
480solvable, but some are hard, I'm afraid. Some distros such as
481openSUSE enables the power-saving feature automatically when the power
482cable is unplugged. Thus, if you hear noises, suspect first the
Takashi Iwai623b9f62008-12-11 07:44:18 +0100483power-saving. See /sys/module/snd_hda_intel/parameters/power_save to
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100484check the current value. If it's non-zero, the feature is turned on.
485
486
Takashi Iwai132bb7c2008-12-11 15:39:52 +0100487Development Tree
488~~~~~~~~~~~~~~~~
489The latest development codes for HD-audio are found on sound git tree:
490
491- git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6.git
492
493The master branch or for-next branches can be used as the main
494development branches in general while the HD-audio specific patches
495are committed in topic/hda branch.
496
497If you are using the latest Linus tree, it'd be better to pull the
498above GIT tree onto it. If you are using the older kernels, an easy
499way to try the latest ALSA code is to build from the snapshot
500tarball. There are daily tarballs and the latest snapshot tarball.
501All can be built just like normal alsa-driver release packages, that
502is, installed via the usual spells: configure, make and make
503install(-modules). See INSTALL in the package. The snapshot tarballs
504are found at:
505
506- ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/
507
508
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100509Sending a Bug Report
510~~~~~~~~~~~~~~~~~~~~
511If any model or module options don't work for your device, it's time
512to send a bug report to the developers. Give the following in your
513bug report:
514
515- Hardware vendor, product and model names
516- Kernel version (and ALSA-driver version if you built externally)
517- `alsa-info.sh` output; run with `--no-upload` option. See the
518 section below about alsa-info
519
520If it's a regression, at best, send alsa-info outputs of both working
521and non-working kernels. This is really helpful because we can
522compare the codec registers directly.
523
524Send a bug report either the followings:
525
526kernel-bugzilla::
527 http://bugme.linux-foundation.org/
528alsa-devel ML::
529 alsa-devel@alsa-project.org
530
531
532DEBUG TOOLS
533-----------
534
535This section describes some tools available for debugging HD-audio
536problems.
537
538alsa-info
539~~~~~~~~~
540The script `alsa-info.sh` is a very useful tool to gather the audio
541device information. You can fetch the latest version from:
542
543- http://www.alsa-project.org/alsa-info.sh
544
545Run this script as root, and it will gather the important information
546such as the module lists, module parameters, proc file contents
547including the codec proc files, mixer outputs and the control
548elements. As default, it will store the information onto a web server
549on alsa-project.org. But, if you send a bug report, it'd be better to
550run with `--no-upload` option, and attach the generated file.
551
552There are some other useful options. See `--help` option output for
553details.
554
Takashi Iwaiae374d62009-02-13 08:33:55 +0100555When a probe error occurs or when the driver obviously assigns a
556mismatched model, it'd be helpful to load the driver with
557`probe_only=1` option (at best after the cold reboot) and run
558alsa-info at this state. With this option, the driver won't configure
559the mixer and PCM but just tries to probe the codec slot. After
560probing, the proc file is available, so you can get the raw codec
561information before modified by the driver. Of course, the driver
562isn't usable with `probe_only=1`. But you can continue the
563configuration via hwdep sysfs file if hda-reconfig option is enabled.
564
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100565
566hda-verb
567~~~~~~~~
568hda-verb is a tiny program that allows you to access the HD-audio
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100569codec directly. You can execute a raw HD-audio codec verb with this.
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100570This program accesses the hwdep device, thus you need to enable the
571kernel config `CONFIG_SND_HDA_HWDEP=y` beforehand.
572
573The hda-verb program takes four arguments: the hwdep device file, the
574widget NID, the verb and the parameter. When you access to the codec
575on the slot 2 of the card 0, pass /dev/snd/hwC0D2 to the first
576argument, typically. (However, the real path name depends on the
577system.)
578
579The second parameter is the widget number-id to access. The third
580parameter can be either a hex/digit number or a string corresponding
581to a verb. Similarly, the last parameter is the value to write, or
582can be a string for the parameter type.
583
584------------------------------------------------------------------------
585 % hda-verb /dev/snd/hwC0D0 0x12 0x701 2
586 nid = 0x12, verb = 0x701, param = 0x2
587 value = 0x0
588
589 % hda-verb /dev/snd/hwC0D0 0x0 PARAMETERS VENDOR_ID
590 nid = 0x0, verb = 0xf00, param = 0x0
591 value = 0x10ec0262
592
593 % hda-verb /dev/snd/hwC0D0 2 set_a 0xb080
594 nid = 0x2, verb = 0x300, param = 0xb080
595 value = 0x0
596------------------------------------------------------------------------
597
598Although you can issue any verbs with this program, the driver state
599won't be always updated. For example, the volume values are usually
600cached in the driver, and thus changing the widget amp value directly
601via hda-verb won't change the mixer value.
602
603The hda-verb program is found in the ftp directory:
604
605- ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/misc/
606
607Also a git repository is available:
608
609- git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/hda-verb.git
610
611See README file in the tarball for more details about hda-verb
612program.
613
614
615hda-analyzer
616~~~~~~~~~~~~
617hda-analyzer provides a graphical interface to access the raw HD-audio
618control, based on pyGTK2 binding. It's a more powerful version of
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100619hda-verb. The program gives you an easy-to-use GUI stuff for showing
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100620the widget information and adjusting the amp values, as well as the
621proc-compatible output.
622
623The hda-analyzer is a part of alsa.git repository in
624alsa-project.org:
625
626- http://git.alsa-project.org/?p=alsa.git;a=tree;f=hda-analyzer
627
628
Takashi Iwai623b9f62008-12-11 07:44:18 +0100629Codecgraph
630~~~~~~~~~~
631Codecgraph is a utility program to generate a graph and visualizes the
632codec-node connection of a codec chip. It's especially useful when
633you analyze or debug a codec without a proper datasheet. The program
634parses the given codec proc file and converts to SVG via graphiz
635program.
636
637The tarball and GIT trees are found in the web page at:
638
639- http://helllabs.org/codecgraph/
640
641
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100642hda-emu
643~~~~~~~
Takashi Iwaid2afbe72008-12-10 09:28:15 +0100644hda-emu is an HD-audio emulator. The main purpose of this program is
Takashi Iwaia7fe49b2008-12-03 18:26:35 +0100645to debug an HD-audio codec without the real hardware. Thus, it
646doesn't emulate the behavior with the real audio I/O, but it just
647dumps the codec register changes and the ALSA-driver internal changes
648at probing and operating the HD-audio driver.
649
650The program requires a codec proc-file to simulate. Get a proc file
651for the target codec beforehand, or pick up an example codec from the
652codec proc collections in the tarball. Then, run the program with the
653proc file, and the hda-emu program will start parsing the codec file
654and simulates the HD-audio driver:
655
656------------------------------------------------------------------------
657 % hda-emu codecs/stac9200-dell-d820-laptop
658 # Parsing..
659 hda_codec: Unknown model for STAC9200, using BIOS defaults
660 hda_codec: pin nid 08 bios pin config 40c003fa
661 ....
662------------------------------------------------------------------------
663
664The program gives you only a very dumb command-line interface. You
665can get a proc-file dump at the current state, get a list of control
666(mixer) elements, set/get the control element value, simulate the PCM
667operation, the jack plugging simulation, etc.
668
669The package is found in:
670
671- ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/misc/
672
673A git repository is available:
674
675- git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/hda-emu.git
676
677See README file in the tarball for more details about hda-emu
678program.