Based on kernel version 4.9. Page generated on 2016-12-21 14:37 EST.
1 MORE NOTES ON HD-AUDIO DRIVER 2 ============================= 3 Takashi Iwai <tiwai@suse.de> 4 5 6 GENERAL 7 ------- 8 9 HD-audio is the new standard on-board audio component on modern PCs 10 after AC97. Although Linux has been supporting HD-audio since long 11 time ago, there are often problems with new machines. A part of the 12 problem is broken BIOS, and the rest is the driver implementation. 13 This document explains the brief trouble-shooting and debugging 14 methods for the HD-audio hardware. 15 16 The HD-audio component consists of two parts: the controller chip and 17 the codec chips on the HD-audio bus. Linux provides a single driver 18 for all controllers, snd-hda-intel. Although the driver name contains 19 a word of a well-known hardware vendor, it's not specific to it but for 20 all controller chips by other companies. Since the HD-audio 21 controllers are supposed to be compatible, the single snd-hda-driver 22 should work in most cases. But, not surprisingly, there are known 23 bugs and issues specific to each controller type. The snd-hda-intel 24 driver has a bunch of workarounds for these as described below. 25 26 A controller may have multiple codecs. Usually you have one audio 27 codec and optionally one modem codec. In theory, there might be 28 multiple audio codecs, e.g. for analog and digital outputs, and the 29 driver might not work properly because of conflict of mixer elements. 30 This should be fixed in future if such hardware really exists. 31 32 The snd-hda-intel driver has several different codec parsers depending 33 on the codec. It has a generic parser as a fallback, but this 34 functionality is fairly limited until now. Instead of the generic 35 parser, usually the codec-specific parser (coded in patch_*.c) is used 36 for the codec-specific implementations. The details about the 37 codec-specific problems are explained in the later sections. 38 39 If you are interested in the deep debugging of HD-audio, read the 40 HD-audio specification at first. The specification is found on 41 Intel's web page, for example: 42 43 - http://www.intel.com/standards/hdaudio/ 44 45 46 HD-AUDIO CONTROLLER 47 ------------------- 48 49 DMA-Position Problem 50 ~~~~~~~~~~~~~~~~~~~~ 51 The most common problem of the controller is the inaccurate DMA 52 pointer reporting. The DMA pointer for playback and capture can be 53 read in two ways, either via a LPIB register or via a position-buffer 54 map. As default the driver tries to read from the io-mapped 55 position-buffer, and falls back to LPIB if the position-buffer appears 56 dead. However, this detection isn't perfect on some devices. In such 57 a case, you can change the default method via `position_fix` option. 58 59 `position_fix=1` means to use LPIB method explicitly. 60 `position_fix=2` means to use the position-buffer. 61 `position_fix=3` means to use a combination of both methods, needed 62 for some VIA controllers. The capture stream position is corrected 63 by comparing both LPIB and position-buffer values. 64 `position_fix=4` is another combination available for all controllers, 65 and uses LPIB for the playback and the position-buffer for the capture 66 streams. 67 0 is the default value for all other 68 controllers, the automatic check and fallback to LPIB as described in 69 the above. If you get a problem of repeated sounds, this option might 70 help. 71 72 In addition to that, every controller is known to be broken regarding 73 the wake-up timing. It wakes up a few samples before actually 74 processing the data on the buffer. This caused a lot of problems, for 75 example, with ALSA dmix or JACK. Since 2.6.27 kernel, the driver puts 76 an artificial delay to the wake up timing. This delay is controlled 77 via `bdl_pos_adj` option. 78 79 When `bdl_pos_adj` is a negative value (as default), it's assigned to 80 an appropriate value depending on the controller chip. For Intel 81 chips, it'd be 1 while it'd be 32 for others. Usually this works. 82 Only in case it doesn't work and you get warning messages, you should 83 change this parameter to other values. 84 85 86 Codec-Probing Problem 87 ~~~~~~~~~~~~~~~~~~~~~ 88 A less often but a more severe problem is the codec probing. When 89 BIOS reports the available codec slots wrongly, the driver gets 90 confused and tries to access the non-existing codec slot. This often 91 results in the total screw-up, and destructs the further communication 92 with the codec chips. The symptom appears usually as error messages 93 like: 94 ------------------------------------------------------------------------ 95 hda_intel: azx_get_response timeout, switching to polling mode: 96 last cmd=0x12345678 97 hda_intel: azx_get_response timeout, switching to single_cmd mode: 98 last cmd=0x12345678 99 ------------------------------------------------------------------------ 100 101 The first line is a warning, and this is usually relatively harmless. 102 It means that the codec response isn't notified via an IRQ. The 103 driver uses explicit polling method to read the response. It gives 104 very slight CPU overhead, but you'd unlikely notice it. 105 106 The second line is, however, a fatal error. If this happens, usually 107 it means that something is really wrong. Most likely you are 108 accessing a non-existing codec slot. 109 110 Thus, if the second error message appears, try to narrow the probed 111 codec slots via `probe_mask` option. It's a bitmask, and each bit 112 corresponds to the codec slot. For example, to probe only the first 113 slot, pass `probe_mask=1`. For the first and the third slots, pass 114 `probe_mask=5` (where 5 = 1 | 4), and so on. 115 116 Since 2.6.29 kernel, the driver has a more robust probing method, so 117 this error might happen rarely, though. 118 119 On a machine with a broken BIOS, sometimes you need to force the 120 driver to probe the codec slots the hardware doesn't report for use. 121 In such a case, turn the bit 8 (0x100) of `probe_mask` option on. 122 Then the rest 8 bits are passed as the codec slots to probe 123 unconditionally. For example, `probe_mask=0x103` will force to probe 124 the codec slots 0 and 1 no matter what the hardware reports. 125 126 127 Interrupt Handling 128 ~~~~~~~~~~~~~~~~~~ 129 HD-audio driver uses MSI as default (if available) since 2.6.33 130 kernel as MSI works better on some machines, and in general, it's 131 better for performance. However, Nvidia controllers showed bad 132 regressions with MSI (especially in a combination with AMD chipset), 133 thus we disabled MSI for them. 134 135 There seem also still other devices that don't work with MSI. If you 136 see a regression wrt the sound quality (stuttering, etc) or a lock-up 137 in the recent kernel, try to pass `enable_msi=0` option to disable 138 MSI. If it works, you can add the known bad device to the blacklist 139 defined in hda_intel.c. In such a case, please report and give the 140 patch back to the upstream developer. 141 142 143 HD-AUDIO CODEC 144 -------------- 145 146 Model Option 147 ~~~~~~~~~~~~ 148 The most common problem regarding the HD-audio driver is the 149 unsupported codec features or the mismatched device configuration. 150 Most of codec-specific code has several preset models, either to 151 override the BIOS setup or to provide more comprehensive features. 152 153 The driver checks PCI SSID and looks through the static configuration 154 table until any matching entry is found. If you have a new machine, 155 you may see a message like below: 156 ------------------------------------------------------------------------ 157 hda_codec: ALC880: BIOS auto-probing. 158 ------------------------------------------------------------------------ 159 Meanwhile, in the earlier versions, you would see a message like: 160 ------------------------------------------------------------------------ 161 hda_codec: Unknown model for ALC880, trying auto-probe from BIOS... 162 ------------------------------------------------------------------------ 163 Even if you see such a message, DON'T PANIC. Take a deep breath and 164 keep your towel. First of all, it's an informational message, no 165 warning, no error. This means that the PCI SSID of your device isn't 166 listed in the known preset model (white-)list. But, this doesn't mean 167 that the driver is broken. Many codec-drivers provide the automatic 168 configuration mechanism based on the BIOS setup. 169 170 The HD-audio codec has usually "pin" widgets, and BIOS sets the default 171 configuration of each pin, which indicates the location, the 172 connection type, the jack color, etc. The HD-audio driver can guess 173 the right connection judging from these default configuration values. 174 However -- some codec-support codes, such as patch_analog.c, don't 175 support the automatic probing (yet as of 2.6.28). And, BIOS is often, 176 yes, pretty often broken. It sets up wrong values and screws up the 177 driver. 178 179 The preset model (or recently called as "fix-up") is provided 180 basically to overcome such a situation. When the matching preset 181 model is found in the white-list, the driver assumes the static 182 configuration of that preset with the correct pin setup, etc. 183 Thus, if you have a newer machine with a slightly different PCI SSID 184 (or codec SSID) from the existing one, you may have a good chance to 185 re-use the same model. You can pass the `model` option to specify the 186 preset model instead of PCI (and codec-) SSID look-up. 187 188 What `model` option values are available depends on the codec chip. 189 Check your codec chip from the codec proc file (see "Codec Proc-File" 190 section below). It will show the vendor/product name of your codec 191 chip. Then, see Documentation/sound/alsa/HD-Audio-Models.txt file, 192 the section of HD-audio driver. You can find a list of codecs 193 and `model` options belonging to each codec. For example, for Realtek 194 ALC262 codec chip, pass `model=ultra` for devices that are compatible 195 with Samsung Q1 Ultra. 196 197 Thus, the first thing you can do for any brand-new, unsupported and 198 non-working HD-audio hardware is to check HD-audio codec and several 199 different `model` option values. If you have any luck, some of them 200 might suit with your device well. 201 202 There are a few special model option values: 203 - when 'nofixup' is passed, the device-specific fixups in the codec 204 parser are skipped. 205 - when `generic` is passed, the codec-specific parser is skipped and 206 only the generic parser is used. 207 208 209 Speaker and Headphone Output 210 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 211 One of the most frequent (and obvious) bugs with HD-audio is the 212 silent output from either or both of a built-in speaker and a 213 headphone jack. In general, you should try a headphone output at 214 first. A speaker output often requires more additional controls like 215 the external amplifier bits. Thus a headphone output has a slightly 216 better chance. 217 218 Before making a bug report, double-check whether the mixer is set up 219 correctly. The recent version of snd-hda-intel driver provides mostly 220 "Master" volume control as well as "Front" volume (where Front 221 indicates the front-channels). In addition, there can be individual 222 "Headphone" and "Speaker" controls. 223 224 Ditto for the speaker output. There can be "External Amplifier" 225 switch on some codecs. Turn on this if present. 226 227 Another related problem is the automatic mute of speaker output by 228 headphone plugging. This feature is implemented in most cases, but 229 not on every preset model or codec-support code. 230 231 In anyway, try a different model option if you have such a problem. 232 Some other models may match better and give you more matching 233 functionality. If none of the available models works, send a bug 234 report. See the bug report section for details. 235 236 If you are masochistic enough to debug the driver problem, note the 237 following: 238 239 - The speaker (and the headphone, too) output often requires the 240 external amplifier. This can be set usually via EAPD verb or a 241 certain GPIO. If the codec pin supports EAPD, you have a better 242 chance via SET_EAPD_BTL verb (0x70c). On others, GPIO pin (mostly 243 it's either GPIO0 or GPIO1) may turn on/off EAPD. 244 - Some Realtek codecs require special vendor-specific coefficients to 245 turn on the amplifier. See patch_realtek.c. 246 - IDT codecs may have extra power-enable/disable controls on each 247 analog pin. See patch_sigmatel.c. 248 - Very rare but some devices don't accept the pin-detection verb until 249 triggered. Issuing GET_PIN_SENSE verb (0xf09) may result in the 250 codec-communication stall. Some examples are found in 251 patch_realtek.c. 252 253 254 Capture Problems 255 ~~~~~~~~~~~~~~~~ 256 The capture problems are often because of missing setups of mixers. 257 Thus, before submitting a bug report, make sure that you set up the 258 mixer correctly. For example, both "Capture Volume" and "Capture 259 Switch" have to be set properly in addition to the right "Capture 260 Source" or "Input Source" selection. Some devices have "Mic Boost" 261 volume or switch. 262 263 When the PCM device is opened via "default" PCM (without pulse-audio 264 plugin), you'll likely have "Digital Capture Volume" control as well. 265 This is provided for the extra gain/attenuation of the signal in 266 software, especially for the inputs without the hardware volume 267 control such as digital microphones. Unless really needed, this 268 should be set to exactly 50%, corresponding to 0dB -- neither extra 269 gain nor attenuation. When you use "hw" PCM, i.e., a raw access PCM, 270 this control will have no influence, though. 271 272 It's known that some codecs / devices have fairly bad analog circuits, 273 and the recorded sound contains a certain DC-offset. This is no bug 274 of the driver. 275 276 Most of modern laptops have no analog CD-input connection. Thus, the 277 recording from CD input won't work in many cases although the driver 278 provides it as the capture source. Use CDDA instead. 279 280 The automatic switching of the built-in and external mic per plugging 281 is implemented on some codec models but not on every model. Partly 282 because of my laziness but mostly lack of testers. Feel free to 283 submit the improvement patch to the author. 284 285 286 Direct Debugging 287 ~~~~~~~~~~~~~~~~ 288 If no model option gives you a better result, and you are a tough guy 289 to fight against evil, try debugging via hitting the raw HD-audio 290 codec verbs to the device. Some tools are available: hda-emu and 291 hda-analyzer. The detailed description is found in the sections 292 below. You'd need to enable hwdep for using these tools. See "Kernel 293 Configuration" section. 294 295 296 OTHER ISSUES 297 ------------ 298 299 Kernel Configuration 300 ~~~~~~~~~~~~~~~~~~~~ 301 In general, I recommend you to enable the sound debug option, 302 `CONFIG_SND_DEBUG=y`, no matter whether you are debugging or not. 303 This enables snd_printd() macro and others, and you'll get additional 304 kernel messages at probing. 305 306 In addition, you can enable `CONFIG_SND_DEBUG_VERBOSE=y`. But this 307 will give you far more messages. Thus turn this on only when you are 308 sure to want it. 309 310 Don't forget to turn on the appropriate `CONFIG_SND_HDA_CODEC_*` 311 options. Note that each of them corresponds to the codec chip, not 312 the controller chip. Thus, even if lspci shows the Nvidia controller, 313 you may need to choose the option for other vendors. If you are 314 unsure, just select all yes. 315 316 `CONFIG_SND_HDA_HWDEP` is a useful option for debugging the driver. 317 When this is enabled, the driver creates hardware-dependent devices 318 (one per each codec), and you have a raw access to the device via 319 these device files. For example, `hwC0D2` will be created for the 320 codec slot #2 of the first card (#0). For debug-tools such as 321 hda-verb and hda-analyzer, the hwdep device has to be enabled. 322 Thus, it'd be better to turn this on always. 323 324 `CONFIG_SND_HDA_RECONFIG` is a new option, and this depends on the 325 hwdep option above. When enabled, you'll have some sysfs files under 326 the corresponding hwdep directory. See "HD-audio reconfiguration" 327 section below. 328 329 `CONFIG_SND_HDA_POWER_SAVE` option enables the power-saving feature. 330 See "Power-saving" section below. 331 332 333 Codec Proc-File 334 ~~~~~~~~~~~~~~~ 335 The codec proc-file is a treasure-chest for debugging HD-audio. 336 It shows most of useful information of each codec widget. 337 338 The proc file is located in /proc/asound/card*/codec#*, one file per 339 each codec slot. You can know the codec vendor, product id and 340 names, the type of each widget, capabilities and so on. 341 This file, however, doesn't show the jack sensing state, so far. This 342 is because the jack-sensing might be depending on the trigger state. 343 344 This file will be picked up by the debug tools, and also it can be fed 345 to the emulator as the primary codec information. See the debug tools 346 section below. 347 348 This proc file can be also used to check whether the generic parser is 349 used. When the generic parser is used, the vendor/product ID name 350 will appear as "Realtek ID 0262", instead of "Realtek ALC262". 351 352 353 HD-Audio Reconfiguration 354 ~~~~~~~~~~~~~~~~~~~~~~~~ 355 This is an experimental feature to allow you re-configure the HD-audio 356 codec dynamically without reloading the driver. The following sysfs 357 files are available under each codec-hwdep device directory (e.g. 358 /sys/class/sound/hwC0D0): 359 360 vendor_id:: 361 Shows the 32bit codec vendor-id hex number. You can change the 362 vendor-id value by writing to this file. 363 subsystem_id:: 364 Shows the 32bit codec subsystem-id hex number. You can change the 365 subsystem-id value by writing to this file. 366 revision_id:: 367 Shows the 32bit codec revision-id hex number. You can change the 368 revision-id value by writing to this file. 369 afg:: 370 Shows the AFG ID. This is read-only. 371 mfg:: 372 Shows the MFG ID. This is read-only. 373 name:: 374 Shows the codec name string. Can be changed by writing to this 375 file. 376 modelname:: 377 Shows the currently set `model` option. Can be changed by writing 378 to this file. 379 init_verbs:: 380 The extra verbs to execute at initialization. You can add a verb by 381 writing to this file. Pass three numbers: nid, verb and parameter 382 (separated with a space). 383 hints:: 384 Shows / stores hint strings for codec parsers for any use. 385 Its format is `key = value`. For example, passing `jack_detect = no` 386 will disable the jack detection of the machine completely. 387 init_pin_configs:: 388 Shows the initial pin default config values set by BIOS. 389 driver_pin_configs:: 390 Shows the pin default values set by the codec parser explicitly. 391 This doesn't show all pin values but only the changed values by 392 the parser. That is, if the parser doesn't change the pin default 393 config values by itself, this will contain nothing. 394 user_pin_configs:: 395 Shows the pin default config values to override the BIOS setup. 396 Writing this (with two numbers, NID and value) appends the new 397 value. The given will be used instead of the initial BIOS value at 398 the next reconfiguration time. Note that this config will override 399 even the driver pin configs, too. 400 reconfig:: 401 Triggers the codec re-configuration. When any value is written to 402 this file, the driver re-initialize and parses the codec tree 403 again. All the changes done by the sysfs entries above are taken 404 into account. 405 clear:: 406 Resets the codec, removes the mixer elements and PCM stuff of the 407 specified codec, and clear all init verbs and hints. 408 409 For example, when you want to change the pin default configuration 410 value of the pin widget 0x14 to 0x9993013f, and let the driver 411 re-configure based on that state, run like below: 412 ------------------------------------------------------------------------ 413 # echo 0x14 0x9993013f > /sys/class/sound/hwC0D0/user_pin_configs 414 # echo 1 > /sys/class/sound/hwC0D0/reconfig 415 ------------------------------------------------------------------------ 416 417 418 Hint Strings 419 ~~~~~~~~~~~~ 420 The codec parser have several switches and adjustment knobs for 421 matching better with the actual codec or device behavior. Many of 422 them can be adjusted dynamically via "hints" strings as mentioned in 423 the section above. For example, by passing `jack_detect = no` string 424 via sysfs or a patch file, you can disable the jack detection, thus 425 the codec parser will skip the features like auto-mute or mic 426 auto-switch. As a boolean value, either `yes`, `no`, `true`, `false`, 427 `1` or `0` can be passed. 428 429 The generic parser supports the following hints: 430 431 - jack_detect (bool): specify whether the jack detection is available 432 at all on this machine; default true 433 - inv_jack_detect (bool): indicates that the jack detection logic is 434 inverted 435 - trigger_sense (bool): indicates that the jack detection needs the 436 explicit call of AC_VERB_SET_PIN_SENSE verb 437 - inv_eapd (bool): indicates that the EAPD is implemented in the 438 inverted logic 439 - pcm_format_first (bool): sets the PCM format before the stream tag 440 and channel ID 441 - sticky_stream (bool): keep the PCM format, stream tag and ID as long 442 as possible; default true 443 - spdif_status_reset (bool): reset the SPDIF status bits at each time 444 the SPDIF stream is set up 445 - pin_amp_workaround (bool): the output pin may have multiple amp 446 values 447 - single_adc_amp (bool): ADCs can have only single input amps 448 - auto_mute (bool): enable/disable the headphone auto-mute feature; 449 default true 450 - auto_mic (bool): enable/disable the mic auto-switch feature; default 451 true 452 - line_in_auto_switch (bool): enable/disable the line-in auto-switch 453 feature; default false 454 - need_dac_fix (bool): limits the DACs depending on the channel count 455 - primary_hp (bool): probe headphone jacks as the primary outputs; 456 default true 457 - multi_io (bool): try probing multi-I/O config (e.g. shared 458 line-in/surround, mic/clfe jacks) 459 - multi_cap_vol (bool): provide multiple capture volumes 460 - inv_dmic_split (bool): provide split internal mic volume/switch for 461 phase-inverted digital mics 462 - indep_hp (bool): provide the independent headphone PCM stream and 463 the corresponding mixer control, if available 464 - add_stereo_mix_input (bool): add the stereo mix (analog-loopback 465 mix) to the input mux if available 466 - add_jack_modes (bool): add "xxx Jack Mode" enum controls to each 467 I/O jack for allowing to change the headphone amp and mic bias VREF 468 capabilities 469 - power_save_node (bool): advanced power management for each widget, 470 controlling the power sate (D0/D3) of each widget node depending on 471 the actual pin and stream states 472 - power_down_unused (bool): power down the unused widgets, a subset of 473 power_save_node, and will be dropped in future 474 - add_hp_mic (bool): add the headphone to capture source if possible 475 - hp_mic_detect (bool): enable/disable the hp/mic shared input for a 476 single built-in mic case; default true 477 - mixer_nid (int): specifies the widget NID of the analog-loopback 478 mixer 479 480 481 Early Patching 482 ~~~~~~~~~~~~~~ 483 When CONFIG_SND_HDA_PATCH_LOADER=y is set, you can pass a "patch" as a 484 firmware file for modifying the HD-audio setup before initializing the 485 codec. This can work basically like the reconfiguration via sysfs in 486 the above, but it does it before the first codec configuration. 487 488 A patch file is a plain text file which looks like below: 489 490 ------------------------------------------------------------------------ 491 [codec] 492 0x12345678 0xabcd1234 2 493 494 [model] 495 auto 496 497 [pincfg] 498 0x12 0x411111f0 499 500 [verb] 501 0x20 0x500 0x03 502 0x20 0x400 0xff 503 504 [hint] 505 jack_detect = no 506 ------------------------------------------------------------------------ 507 508 The file needs to have a line `[codec]`. The next line should contain 509 three numbers indicating the codec vendor-id (0x12345678 in the 510 example), the codec subsystem-id (0xabcd1234) and the address (2) of 511 the codec. The rest patch entries are applied to this specified codec 512 until another codec entry is given. Passing 0 or a negative number to 513 the first or the second value will make the check of the corresponding 514 field be skipped. It'll be useful for really broken devices that don't 515 initialize SSID properly. 516 517 The `[model]` line allows to change the model name of the each codec. 518 In the example above, it will be changed to model=auto. 519 Note that this overrides the module option. 520 521 After the `[pincfg]` line, the contents are parsed as the initial 522 default pin-configurations just like `user_pin_configs` sysfs above. 523 The values can be shown in user_pin_configs sysfs file, too. 524 525 Similarly, the lines after `[verb]` are parsed as `init_verbs` 526 sysfs entries, and the lines after `[hint]` are parsed as `hints` 527 sysfs entries, respectively. 528 529 Another example to override the codec vendor id from 0x12345678 to 530 0xdeadbeef is like below: 531 ------------------------------------------------------------------------ 532 [codec] 533 0x12345678 0xabcd1234 2 534 535 [vendor_id] 536 0xdeadbeef 537 ------------------------------------------------------------------------ 538 539 In the similar way, you can override the codec subsystem_id via 540 `[subsystem_id]`, the revision id via `[revision_id]` line. 541 Also, the codec chip name can be rewritten via `[chip_name]` line. 542 ------------------------------------------------------------------------ 543 [codec] 544 0x12345678 0xabcd1234 2 545 546 [subsystem_id] 547 0xffff1111 548 549 [revision_id] 550 0x10 551 552 [chip_name] 553 My-own NEWS-0002 554 ------------------------------------------------------------------------ 555 556 The hd-audio driver reads the file via request_firmware(). Thus, 557 a patch file has to be located on the appropriate firmware path, 558 typically, /lib/firmware. For example, when you pass the option 559 `patch=hda-init.fw`, the file /lib/firmware/hda-init.fw must be 560 present. 561 562 The patch module option is specific to each card instance, and you 563 need to give one file name for each instance, separated by commas. 564 For example, if you have two cards, one for an on-board analog and one 565 for an HDMI video board, you may pass patch option like below: 566 ------------------------------------------------------------------------ 567 options snd-hda-intel patch=on-board-patch,hdmi-patch 568 ------------------------------------------------------------------------ 569 570 571 Power-Saving 572 ~~~~~~~~~~~~ 573 The power-saving is a kind of auto-suspend of the device. When the 574 device is inactive for a certain time, the device is automatically 575 turned off to save the power. The time to go down is specified via 576 `power_save` module option, and this option can be changed dynamically 577 via sysfs. 578 579 The power-saving won't work when the analog loopback is enabled on 580 some codecs. Make sure that you mute all unneeded signal routes when 581 you want the power-saving. 582 583 The power-saving feature might cause audible click noises at each 584 power-down/up depending on the device. Some of them might be 585 solvable, but some are hard, I'm afraid. Some distros such as 586 openSUSE enables the power-saving feature automatically when the power 587 cable is unplugged. Thus, if you hear noises, suspect first the 588 power-saving. See /sys/module/snd_hda_intel/parameters/power_save to 589 check the current value. If it's non-zero, the feature is turned on. 590 591 The recent kernel supports the runtime PM for the HD-audio controller 592 chip, too. It means that the HD-audio controller is also powered up / 593 down dynamically. The feature is enabled only for certain controller 594 chips like Intel LynxPoint. You can enable/disable this feature 595 forcibly by setting `power_save_controller` option, which is also 596 available at /sys/module/snd_hda_intel/parameters directory. 597 598 599 Tracepoints 600 ~~~~~~~~~~~ 601 The hd-audio driver gives a few basic tracepoints. 602 `hda:hda_send_cmd` traces each CORB write while `hda:hda_get_response` 603 traces the response from RIRB (only when read from the codec driver). 604 `hda:hda_bus_reset` traces the bus-reset due to fatal error, etc, 605 `hda:hda_unsol_event` traces the unsolicited events, and 606 `hda:hda_power_down` and `hda:hda_power_up` trace the power down/up 607 via power-saving behavior. 608 609 Enabling all tracepoints can be done like 610 ------------------------------------------------------------------------ 611 # echo 1 > /sys/kernel/debug/tracing/events/hda/enable 612 ------------------------------------------------------------------------ 613 then after some commands, you can traces from 614 /sys/kernel/debug/tracing/trace file. For example, when you want to 615 trace what codec command is sent, enable the tracepoint like: 616 ------------------------------------------------------------------------ 617 # cat /sys/kernel/debug/tracing/trace 618 # tracer: nop 619 # 620 # TASK-PID CPU# TIMESTAMP FUNCTION 621 # | | | | | 622 <...>-7807 [002] 105147.774889: hda_send_cmd: [0:0] val=e3a019 623 <...>-7807 [002] 105147.774893: hda_send_cmd: [0:0] val=e39019 624 <...>-7807 [002] 105147.999542: hda_send_cmd: [0:0] val=e3a01a 625 <...>-7807 [002] 105147.999543: hda_send_cmd: [0:0] val=e3901a 626 <...>-26764 [001] 349222.837143: hda_send_cmd: [0:0] val=e3a019 627 <...>-26764 [001] 349222.837148: hda_send_cmd: [0:0] val=e39019 628 <...>-26764 [001] 349223.058539: hda_send_cmd: [0:0] val=e3a01a 629 <...>-26764 [001] 349223.058541: hda_send_cmd: [0:0] val=e3901a 630 ------------------------------------------------------------------------ 631 Here `[0:0]` indicates the card number and the codec address, and 632 `val` shows the value sent to the codec, respectively. The value is 633 a packed value, and you can decode it via hda-decode-verb program 634 included in hda-emu package below. For example, the value e3a019 is 635 to set the left output-amp value to 25. 636 ------------------------------------------------------------------------ 637 % hda-decode-verb 0xe3a019 638 raw value = 0x00e3a019 639 cid = 0, nid = 0x0e, verb = 0x3a0, parm = 0x19 640 raw value: verb = 0x3a0, parm = 0x19 641 verbname = set_amp_gain_mute 642 amp raw val = 0xa019 643 output, left, idx=0, mute=0, val=25 644 ------------------------------------------------------------------------ 645 646 647 Development Tree 648 ~~~~~~~~~~~~~~~~ 649 The latest development codes for HD-audio are found on sound git tree: 650 651 - git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git 652 653 The master branch or for-next branches can be used as the main 654 development branches in general while the development for the current 655 and next kernels are found in for-linus and for-next branches, 656 respectively. 657 658 659 Sending a Bug Report 660 ~~~~~~~~~~~~~~~~~~~~ 661 If any model or module options don't work for your device, it's time 662 to send a bug report to the developers. Give the following in your 663 bug report: 664 665 - Hardware vendor, product and model names 666 - Kernel version (and ALSA-driver version if you built externally) 667 - `alsa-info.sh` output; run with `--no-upload` option. See the 668 section below about alsa-info 669 670 If it's a regression, at best, send alsa-info outputs of both working 671 and non-working kernels. This is really helpful because we can 672 compare the codec registers directly. 673 674 Send a bug report either the followings: 675 676 kernel-bugzilla:: 677 https://bugzilla.kernel.org/ 678 alsa-devel ML:: 679 alsa-devel@alsa-project.org 680 681 682 DEBUG TOOLS 683 ----------- 684 685 This section describes some tools available for debugging HD-audio 686 problems. 687 688 alsa-info 689 ~~~~~~~~~ 690 The script `alsa-info.sh` is a very useful tool to gather the audio 691 device information. It's included in alsa-utils package. The latest 692 version can be found on git repository: 693 694 - git://git.alsa-project.org/alsa-utils.git 695 696 The script can be fetched directly from the following URL, too: 697 698 - http://www.alsa-project.org/alsa-info.sh 699 700 Run this script as root, and it will gather the important information 701 such as the module lists, module parameters, proc file contents 702 including the codec proc files, mixer outputs and the control 703 elements. As default, it will store the information onto a web server 704 on alsa-project.org. But, if you send a bug report, it'd be better to 705 run with `--no-upload` option, and attach the generated file. 706 707 There are some other useful options. See `--help` option output for 708 details. 709 710 When a probe error occurs or when the driver obviously assigns a 711 mismatched model, it'd be helpful to load the driver with 712 `probe_only=1` option (at best after the cold reboot) and run 713 alsa-info at this state. With this option, the driver won't configure 714 the mixer and PCM but just tries to probe the codec slot. After 715 probing, the proc file is available, so you can get the raw codec 716 information before modified by the driver. Of course, the driver 717 isn't usable with `probe_only=1`. But you can continue the 718 configuration via hwdep sysfs file if hda-reconfig option is enabled. 719 Using `probe_only` mask 2 skips the reset of HDA codecs (use 720 `probe_only=3` as module option). The hwdep interface can be used 721 to determine the BIOS codec initialization. 722 723 724 hda-verb 725 ~~~~~~~~ 726 hda-verb is a tiny program that allows you to access the HD-audio 727 codec directly. You can execute a raw HD-audio codec verb with this. 728 This program accesses the hwdep device, thus you need to enable the 729 kernel config `CONFIG_SND_HDA_HWDEP=y` beforehand. 730 731 The hda-verb program takes four arguments: the hwdep device file, the 732 widget NID, the verb and the parameter. When you access to the codec 733 on the slot 2 of the card 0, pass /dev/snd/hwC0D2 to the first 734 argument, typically. (However, the real path name depends on the 735 system.) 736 737 The second parameter is the widget number-id to access. The third 738 parameter can be either a hex/digit number or a string corresponding 739 to a verb. Similarly, the last parameter is the value to write, or 740 can be a string for the parameter type. 741 742 ------------------------------------------------------------------------ 743 % hda-verb /dev/snd/hwC0D0 0x12 0x701 2 744 nid = 0x12, verb = 0x701, param = 0x2 745 value = 0x0 746 747 % hda-verb /dev/snd/hwC0D0 0x0 PARAMETERS VENDOR_ID 748 nid = 0x0, verb = 0xf00, param = 0x0 749 value = 0x10ec0262 750 751 % hda-verb /dev/snd/hwC0D0 2 set_a 0xb080 752 nid = 0x2, verb = 0x300, param = 0xb080 753 value = 0x0 754 ------------------------------------------------------------------------ 755 756 Although you can issue any verbs with this program, the driver state 757 won't be always updated. For example, the volume values are usually 758 cached in the driver, and thus changing the widget amp value directly 759 via hda-verb won't change the mixer value. 760 761 The hda-verb program is included now in alsa-tools: 762 763 - git://git.alsa-project.org/alsa-tools.git 764 765 Also, the old stand-alone package is found in the ftp directory: 766 767 - ftp://ftp.suse.com/pub/people/tiwai/misc/ 768 769 Also a git repository is available: 770 771 - git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/hda-verb.git 772 773 See README file in the tarball for more details about hda-verb 774 program. 775 776 777 hda-analyzer 778 ~~~~~~~~~~~~ 779 hda-analyzer provides a graphical interface to access the raw HD-audio 780 control, based on pyGTK2 binding. It's a more powerful version of 781 hda-verb. The program gives you an easy-to-use GUI stuff for showing 782 the widget information and adjusting the amp values, as well as the 783 proc-compatible output. 784 785 The hda-analyzer: 786 787 - http://git.alsa-project.org/?p=alsa.git;a=tree;f=hda-analyzer 788 789 is a part of alsa.git repository in alsa-project.org: 790 791 - git://git.alsa-project.org/alsa.git 792 793 Codecgraph 794 ~~~~~~~~~~ 795 Codecgraph is a utility program to generate a graph and visualizes the 796 codec-node connection of a codec chip. It's especially useful when 797 you analyze or debug a codec without a proper datasheet. The program 798 parses the given codec proc file and converts to SVG via graphiz 799 program. 800 801 The tarball and GIT trees are found in the web page at: 802 803 - http://helllabs.org/codecgraph/ 804 805 806 hda-emu 807 ~~~~~~~ 808 hda-emu is an HD-audio emulator. The main purpose of this program is 809 to debug an HD-audio codec without the real hardware. Thus, it 810 doesn't emulate the behavior with the real audio I/O, but it just 811 dumps the codec register changes and the ALSA-driver internal changes 812 at probing and operating the HD-audio driver. 813 814 The program requires a codec proc-file to simulate. Get a proc file 815 for the target codec beforehand, or pick up an example codec from the 816 codec proc collections in the tarball. Then, run the program with the 817 proc file, and the hda-emu program will start parsing the codec file 818 and simulates the HD-audio driver: 819 820 ------------------------------------------------------------------------ 821 % hda-emu codecs/stac9200-dell-d820-laptop 822 # Parsing.. 823 hda_codec: Unknown model for STAC9200, using BIOS defaults 824 hda_codec: pin nid 08 bios pin config 40c003fa 825 .... 826 ------------------------------------------------------------------------ 827 828 The program gives you only a very dumb command-line interface. You 829 can get a proc-file dump at the current state, get a list of control 830 (mixer) elements, set/get the control element value, simulate the PCM 831 operation, the jack plugging simulation, etc. 832 833 The program is found in the git repository below: 834 835 - git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/hda-emu.git 836 837 See README file in the repository for more details about hda-emu 838 program. 839 840 841 hda-jack-retask 842 ~~~~~~~~~~~~~~~ 843 hda-jack-retask is a user-friendly GUI program to manipulate the 844 HD-audio pin control for jack retasking. If you have a problem about 845 the jack assignment, try this program and check whether you can get 846 useful results. Once when you figure out the proper pin assignment, 847 it can be fixed either in the driver code statically or via passing a 848 firmware patch file (see "Early Patching" section). 849 850 The program is included in alsa-tools now: 851 852 - git://git.alsa-project.org/alsa-tools.git