d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
9fc371f86b
freebsd-nq/sys/dev/sound/pcm/ac97.c

825 lines
22 KiB
C
Raw Normal View History

Start each of the license/copyright comments with /*-, minor shuffle of lines
2005-01-06 01:43:34 +00:00
/*-
update my email address.
2003-09-07 16:28:03 +00:00
* Copyright (c) 1999 Cameron Grant <cg@freebsd.org>
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
repo-copied to make way for newmidi, this commit updates include paths
1999-11-20 16:50:33 +00:00
#include <dev/sound/pcm/sound.h>
#include <dev/sound/pcm/ac97.h>
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
#include <dev/sound/pcm/ac97_patch.h>
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
#include "mixer_if.h"
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
many changes: * add new channels to the end of the list so channels used in order of addition * de-globalise definition of struct snddev_info and provide accessor functions where necessary. * move the $FreeBSD$ tag in each .c file into a macro and allow the /dev/sndstat handler to display these when set to maximum verbosity to aid debugging. * allow each device to register its own sndstat handler to reduce the amount of groping sndstat must do in foreign structs.
2001-08-23 11:30:52 +00:00
SND_DECLARE_FILE("$FreeBSD$");
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
MALLOC_DEFINE(M_AC97, "ac97", "ac97 codec");
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
move a private struct definition from ac97.h to ac97.c add proper handling of cards which take eapd=1 to mean 'produce sound' instead of 'external amplifier power down'
2001-06-23 17:32:40 +00:00
struct ac97mixtable_entry {
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
int reg:8; /* register index */
/* reg < 0 if inverted polarity */
unsigned bits:4; /* width of control field */
unsigned ofs:4; /* offset (only if stereo=0) */
unsigned stereo:1; /* set for stereo controls */
unsigned mute:1; /* bit15 is MUTE */
unsigned recidx:4; /* index in rec mux */
unsigned mask:1; /* use only masked bits */
unsigned enable:1; /* entry is enabled */
move a private struct definition from ac97.h to ac97.c add proper handling of cards which take eapd=1 to mean 'produce sound' instead of 'external amplifier power down'
2001-06-23 17:32:40 +00:00
};
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
#define AC97_NAMELEN 16
struct ac97_info {
kobj_t methods;
device_t dev;
void *devinfo;
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
u_int32_t id;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
unsigned count, caps, se, extcaps, extid, extstat, noext:1;
move a private struct definition from ac97.h to ac97.c add proper handling of cards which take eapd=1 to mean 'produce sound' instead of 'external amplifier power down'
2001-06-23 17:32:40 +00:00
u_int32_t flags;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
struct ac97mixtable_entry mix[32];
char name[AC97_NAMELEN];
(hopefully) fix build breakage some people are seeing Approved by: re
2002-11-26 18:16:27 +00:00
struct mtx *lock;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
};
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
struct ac97_vendorid {
u_int32_t id;
const char *name;
};
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
struct ac97_codecid {
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
u_int32_t id;
u_int8_t stepmask;
u_int8_t noext:1;
char *name;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_patch patch;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
};
static const struct ac97mixtable_entry ac97mixtable_default[32] = {
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
/* [offset] reg bits of st mu re mk en */
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
[SOUND_MIXER_VOLUME] = { AC97_MIX_MASTER, 5, 0, 1, 1, 6, 0, 1 },
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
[SOUND_MIXER_OGAIN] = { AC97_MIX_AUXOUT, 5, 0, 1, 1, 0, 0, 0 },
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
[SOUND_MIXER_PHONEOUT] = { AC97_MIX_MONO, 5, 0, 0, 1, 7, 0, 0 },
[SOUND_MIXER_BASS] = { AC97_MIX_TONE, 4, 8, 0, 0, 0, 1, 0 },
[SOUND_MIXER_TREBLE] = { AC97_MIX_TONE, 4, 0, 0, 0, 0, 1, 0 },
[SOUND_MIXER_PCM] = { AC97_MIX_PCM, 5, 0, 1, 1, 0, 0, 1 },
[SOUND_MIXER_SPEAKER] = { AC97_MIX_BEEP, 4, 1, 0, 1, 0, 0, 0 },
[SOUND_MIXER_LINE] = { AC97_MIX_LINE, 5, 0, 1, 1, 5, 0, 1 },
[SOUND_MIXER_PHONEIN] = { AC97_MIX_PHONE, 5, 0, 0, 1, 8, 0, 0 },
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
[SOUND_MIXER_MIC] = { AC97_MIX_MIC, 5, 0, 0, 1, 1, 1, 1 },
Comment out the mixer entry for igain for the mic 20dB boost for now since it can cause feedback problems on some laptops. Reviewed by: orion
2003-04-23 16:49:53 +00:00
#if 0
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
/* use igain for the mic 20dB boost */
[SOUND_MIXER_IGAIN] = { -AC97_MIX_MIC, 1, 6, 0, 0, 0, 1, 1 },
Comment out the mixer entry for igain for the mic 20dB boost for now since it can cause feedback problems on some laptops. Reviewed by: orion
2003-04-23 16:49:53 +00:00
#endif
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
[SOUND_MIXER_CD] = { AC97_MIX_CD, 5, 0, 1, 1, 2, 0, 1 },
[SOUND_MIXER_LINE1] = { AC97_MIX_AUX, 5, 0, 1, 1, 4, 0, 0 },
[SOUND_MIXER_VIDEO] = { AC97_MIX_VIDEO, 5, 0, 1, 1, 3, 0, 0 },
[SOUND_MIXER_RECLEV] = { -AC97_MIX_RGAIN, 4, 0, 1, 1, 0, 0, 1 }
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
};
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
static const struct ac97_vendorid ac97vendorid[] = {
{ 0x41445300, "Analog Devices" },
{ 0x414b4d00, "Asahi Kasei" },
{ 0x414c4300, "Realtek" },
{ 0x414c4700, "Avance Logic" },
{ 0x43525900, "Cirrus Logic" },
{ 0x434d4900, "C-Media Electronics" },
{ 0x43585400, "Conexant" },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x44543000, "Diamond Technology" },
Add eMicro codec ids.
2003-04-21 04:48:40 +00:00
{ 0x454d4300, "eMicro" },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x45838300, "ESS Technology" },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x48525300, "Intersil" },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x49434500, "ICEnsemble" },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x49544500, "ITE, Inc." },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x4e534300, "National Semiconductor" },
{ 0x50534300, "Philips Semiconductor" },
{ 0x83847600, "SigmaTel" },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x53494c00, "Silicon Laboratories" },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x54524100, "TriTech" },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x54584e00, "Texas Instruments" },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x56494100, "VIA Technologies" },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x57454300, "Winbond" },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x574d4c00, "Wolfson" },
{ 0x594d4800, "Yamaha" },
Add Creative EV1938. Submitted by: David Xu <davidxu@FreeBSD.org> PR: kern/54810
2003-08-29 03:24:08 +00:00
{ 0x01408300, "Creative" },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x00000000, NULL }
};
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
static struct ac97_codecid ac97codecid[] = {
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x41445303, 0x00, 0, "AD1819", 0 },
{ 0x41445340, 0x00, 0, "AD1881", 0 },
{ 0x41445348, 0x00, 0, "AD1881A", 0 },
{ 0x41445360, 0x00, 0, "AD1885", 0 },
{ 0x41445361, 0x00, 0, "AD1886", ad1886_patch },
Additional codec ids.
2003-04-04 07:37:25 +00:00
{ 0x41445362, 0x00, 0, "AD1887", 0 },
{ 0x41445363, 0x00, 0, "AD1886A", 0 },
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
{ 0x41445370, 0x00, 0, "AD1980", ad198x_patch },
Additional codec ids.
2003-04-04 07:37:25 +00:00
{ 0x41445372, 0x00, 0, "AD1981A", 0 },
{ 0x41445374, 0x00, 0, "AD1981B", 0 },
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
{ 0x41445375, 0x00, 0, "AD1985", ad198x_patch },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x414b4d00, 0x00, 1, "AK4540", 0 },
{ 0x414b4d01, 0x00, 1, "AK4542", 0 },
{ 0x414b4d02, 0x00, 1, "AK4543", 0 },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x414b4d06, 0x00, 0, "AK4544A", 0 },
{ 0x454b4d07, 0x00, 0, "AK4545", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x414c4320, 0x0f, 0, "ALC100", 0 },
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
{ 0x414c4730, 0x0f, 0, "ALC101", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x414c4710, 0x0f, 0, "ALC200", 0 },
{ 0x414c4740, 0x0f, 0, "ALC202", 0 },
{ 0x414c4720, 0x0f, 0, "ALC650", 0 },
Recognize the Avance Logic ALC655 codec found on some ICH4/5-based motherboards, such as the Gigabyte I848P. PR: kern/54176 Submitted by: Chris Keladis <chris@cmc.optus.net.au> Forgotten by: orion
2003-10-12 11:51:24 +00:00
{ 0x414c4760, 0x0f, 0, "ALC655", 0 },
Add ID for ALC658 CODEC. Tested on: GIGABYTE GA-8S655FX-L Reviewed by: orion
2003-11-11 22:15:17 +00:00
{ 0x414c4780, 0x0f, 0, "ALC658", 0 },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x414c4790, 0x0f, 0, "ALC850", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x43525900, 0x07, 0, "CS4297", 0 },
{ 0x43525910, 0x07, 0, "CS4297A", 0 },
{ 0x43525920, 0x07, 0, "CS4294/98", 0 },
Add support CS4294 PR: kern/66280 Submitted by: Christian Brueffer <chris@unixpages.org>
2004-05-08 03:41:40 +00:00
{ 0x4352592d, 0x07, 0, "CS4294", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x43525930, 0x07, 0, "CS4299", 0 },
{ 0x43525940, 0x07, 0, "CS4201", 0 },
Correct CS4205 base vendor id. Reported by: Christophe Juniet and Mark Santcroos.
2003-03-12 15:11:25 +00:00
{ 0x43525958, 0x07, 0, "CS4205", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x43525960, 0x07, 0, "CS4291A", 0 },
{ 0x434d4961, 0x00, 0, "CMI9739", 0 },
{ 0x434d4941, 0x00, 0, "CMI9738", 0 },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x43585421, 0x00, 0, "HSD11246", 0 },
{ 0x43585428, 0x07, 0, "CX20468", 0 },
{ 0x44543000, 0x00, 0, "DT0398", 0 },
Add eMicro codec ids.
2003-04-21 04:48:40 +00:00
{ 0x454d4323, 0x00, 0, "EM28023", 0 },
{ 0x454d4328, 0x00, 0, "EM28028", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x45838308, 0x00, 0, "ES1988", 0 }, /* Formerly ES1921(?) */
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x48525300, 0x00, 0, "HMP9701", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x49434501, 0x00, 0, "ICE1230", 0 },
{ 0x49434511, 0x00, 0, "ICE1232", 0 },
{ 0x49434514, 0x00, 0, "ICE1232A", 0 },
Additional VT1616 id. Submitted by: Greg Lewis
2003-08-21 23:22:51 +00:00
{ 0x49434551, 0x03, 0, "VT1616", 0 }, /* Via badged ICE */
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x49544520, 0x00, 0, "ITE2226E", 0 },
{ 0x49544560, 0x07, 0, "ITE2646E", 0 }, /* XXX: patch needed */
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x4e534340, 0x00, 0, "LM4540", 0 }, /* Spec blank on revid */
{ 0x4e534343, 0x00, 0, "LM4543", 0 }, /* Ditto */
{ 0x4e534346, 0x00, 0, "LM4546A", 0 },
{ 0x4e534348, 0x00, 0, "LM4548A", 0 },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x4e534331, 0x00, 0, "LM4549", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x4e534349, 0x00, 0, "LM4549A", 0 },
{ 0x4e534350, 0x00, 0, "LM4550", 0 },
{ 0x50534301, 0x00, 0, "UCB1510", 0 },
{ 0x50534304, 0x00, 0, "UCB1400", 0 },
{ 0x83847600, 0x00, 0, "STAC9700/83/84", 0 },
{ 0x83847604, 0x00, 0, "STAC9701/03/04/05", 0 },
Add eMicro codec ids.
2003-04-21 04:48:40 +00:00
{ 0x83847605, 0x00, 0, "STAC9704", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x83847608, 0x00, 0, "STAC9708/11", 0 },
{ 0x83847609, 0x00, 0, "STAC9721/23", 0 },
{ 0x83847644, 0x00, 0, "STAC9744/45", 0 },
{ 0x83847650, 0x00, 0, "STAC9750/51", 0 },
{ 0x83847652, 0x00, 0, "STAC9752/53", 0 },
{ 0x83847656, 0x00, 0, "STAC9756/57", 0 },
{ 0x83847658, 0x00, 0, "STAC9758/59", 0 },
{ 0x83847660, 0x00, 0, "STAC9760/61", 0 }, /* Extrapolated */
{ 0x83847662, 0x00, 0, "STAC9762/63", 0 }, /* Extrapolated */
{ 0x53494c22, 0x00, 0, "Si3036", 0 },
{ 0x53494c23, 0x00, 0, "Si3038", 0 },
{ 0x54524103, 0x00, 0, "TR28023", 0 }, /* Extrapolated */
{ 0x54524106, 0x00, 0, "TR28026", 0 },
{ 0x54524108, 0x00, 0, "TR28028", 0 },
{ 0x54524123, 0x00, 0, "TR28602", 0 },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x54524e03, 0x07, 0, "TLV320AIC27", 0 },
{ 0x54584e20, 0x00, 0, "TLC320AD90", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x56494161, 0x00, 0, "VIA1612A", 0 },
{ 0x574d4c00, 0x00, 0, "WM9701A", 0 },
{ 0x574d4c03, 0x00, 0, "WM9703/4/7/8", 0 },
{ 0x574d4c04, 0x00, 0, "WM9704Q", 0 },
{ 0x574d4c05, 0x00, 0, "WM9705/10", 0 },
Sync with DFBSD v.1.16. Add new codecs IDs, fix some spelling. Approved by: des (interim mentor)
2004-01-23 16:38:54 +00:00
{ 0x574d4d09, 0x00, 0, "WM9709", 0 },
{ 0x574d4c12, 0x00, 0, "WM9711/12", 0 }, /* XXX: patch needed */
{ 0x57454301, 0x00, 0, "W83971D", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0x594d4800, 0x00, 0, "YMF743", 0 },
{ 0x594d4802, 0x00, 0, "YMF752", 0 },
{ 0x594d4803, 0x00, 0, "YMF753", 0 },
Add Creative EV1938. Submitted by: David Xu <davidxu@FreeBSD.org> PR: kern/54810
2003-08-29 03:24:08 +00:00
{ 0x01408384, 0x00, 0, "EV1938", 0 },
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
{ 0, 0, 0, NULL, 0 }
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
};
static char *ac97enhancement[] = {
if we have a codec init routine, fail the attach if init fails
2000-05-26 21:42:50 +00:00
"no 3D Stereo Enhancement",
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
"Analog Devices Phat Stereo",
"Creative Stereo Enhancement",
"National Semi 3D Stereo Enhancement",
"Yamaha Ymersion",
"BBE 3D Stereo Enhancement",
"Crystal Semi 3D Stereo Enhancement",
"Qsound QXpander",
"Spatializer 3D Stereo Enhancement",
"SRS 3D Stereo Enhancement",
"Platform Tech 3D Stereo Enhancement",
"AKM 3D Audio",
"Aureal Stereo Enhancement",
"Aztech 3D Enhancement",
"Binaura 3D Audio Enhancement",
"ESS Technology Stereo Enhancement",
"Harman International VMAx",
"Nvidea 3D Stereo Enhancement",
"Philips Incredible Sound",
"Texas Instruments 3D Stereo Enhancement",
"VLSI Technology 3D Stereo Enhancement",
"TriTech 3D Stereo Enhancement",
"Realtek 3D Stereo Enhancement",
"Samsung 3D Stereo Enhancement",
"Wolfson Microelectronics 3D Enhancement",
"Delta Integration 3D Enhancement",
"SigmaTel 3D Enhancement",
"Reserved 27",
"Rockwell 3D Stereo Enhancement",
"Reserved 29",
"Reserved 30",
"Reserved 31"
};
static char *ac97feature[] = {
"mic channel",
"reserved",
"tone",
"simulated stereo",
"headphone",
"bass boost",
"18 bit DAC",
"20 bit DAC",
"18 bit ADC",
"20 bit ADC"
};
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
static char *ac97extfeature[] = {
"variable rate PCM",
"double rate PCM",
"reserved 1",
"variable rate mic",
"reserved 2",
"reserved 3",
"center DAC",
"surround DAC",
"LFE DAC",
"AMAP",
"reserved 4",
"reserved 5",
"reserved 6",
"reserved 7",
};
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
u_int16_t
ac97_rdcd(struct ac97_info *codec, int reg)
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
{
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
return AC97_READ(codec->methods, codec->devinfo, reg);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
}
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
void
ac97_wrcd(struct ac97_info *codec, int reg, u_int16_t val)
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
{
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
AC97_WRITE(codec->methods, codec->devinfo, reg, val);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
}
Add ac97_reset function that polls ready bits in power register. One component in tweaks required for the ICH3 controller in the thinkpad x22 reported by Colin Perkins on -multimedia.
2002-04-26 15:27:56 +00:00
static void
ac97_reset(struct ac97_info *codec)
{
u_int32_t i, ps;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REG_RESET, 0);
Add ac97_reset function that polls ready bits in power register. One component in tweaks required for the ICH3 controller in the thinkpad x22 reported by Colin Perkins on -multimedia.
2002-04-26 15:27:56 +00:00
for (i = 0; i < 500; i++) {
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ps = ac97_rdcd(codec, AC97_REG_POWER) & AC97_POWER_STATUS;
(hopefully) fix build breakage some people are seeing Approved by: re
2002-11-26 18:16:27 +00:00
if (ps == AC97_POWER_STATUS)
Add ac97_reset function that polls ready bits in power register. One component in tweaks required for the ICH3 controller in the thinkpad x22 reported by Colin Perkins on -multimedia.
2002-04-26 15:27:56 +00:00
return;
DELAY(1000);
}
Print ac97 name/id on normal boot. Fix typo reported in pr misc/47794.
2003-02-03 01:38:43 +00:00
device_printf(codec->dev, "AC97 reset timed out.\n");
Add ac97_reset function that polls ready bits in power register. One component in tweaks required for the ICH3 controller in the thinkpad x22 reported by Colin Perkins on -multimedia.
2002-04-26 15:27:56 +00:00
}
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
int
ac97_setrate(struct ac97_info *codec, int which, int rate)
{
u_int16_t v;
switch(which) {
(hopefully) fix build breakage some people are seeing Approved by: re
2002-11-26 18:16:27 +00:00
case AC97_REGEXT_FDACRATE:
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
case AC97_REGEXT_SDACRATE:
case AC97_REGEXT_LDACRATE:
(hopefully) fix build breakage some people are seeing Approved by: re
2002-11-26 18:16:27 +00:00
case AC97_REGEXT_LADCRATE:
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
case AC97_REGEXT_MADCRATE:
break;
default:
return -1;
}
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (rate != 0) {
v = rate;
if (codec->extstat & AC97_EXTCAP_DRA)
v >>= 1;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, which, v);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
}
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
v = ac97_rdcd(codec, which);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (codec->extstat & AC97_EXTCAP_DRA)
v <<= 1;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
return v;
}
int
ac97_setextmode(struct ac97_info *codec, u_int16_t mode)
{
mode &= AC97_EXTCAPS;
Emit a warning if invalid mode passed to ac97_setextmode. MFC after: 3 days
2001-11-21 04:08:10 +00:00
if ((mode & ~codec->extcaps) != 0) {
(hopefully) fix build breakage some people are seeing Approved by: re
2002-11-26 18:16:27 +00:00
device_printf(codec->dev, "ac97 invalid mode set 0x%04x\n",
Emit a warning if invalid mode passed to ac97_setextmode. MFC after: 3 days
2001-11-21 04:08:10 +00:00
mode);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
return -1;
Emit a warning if invalid mode passed to ac97_setextmode. MFC after: 3 days
2001-11-21 04:08:10 +00:00
}
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REGEXT_STAT, mode);
codec->extstat = ac97_rdcd(codec, AC97_REGEXT_STAT) & AC97_EXTCAPS;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
return (mode == codec->extstat)? 0 : -1;
}
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
u_int16_t
ac97_getextmode(struct ac97_info *codec)
{
return codec->extstat;
}
u_int16_t
ac97_getextcaps(struct ac97_info *codec)
{
return codec->extcaps;
}
add a method for retrieving codec capabilities
2001-09-18 14:34:59 +00:00
u_int16_t
ac97_getcaps(struct ac97_info *codec)
{
return codec->caps;
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
static int
ac97_setrecsrc(struct ac97_info *codec, int channel)
{
struct ac97mixtable_entry *e = &codec->mix[channel];
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
if (e->recidx > 0) {
int val = e->recidx - 1;
val |= val << 8;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REG_RECSEL, val);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return 0;
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
} else
return -1;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
}
static int
ac97_setmixer(struct ac97_info *codec, unsigned channel, unsigned left, unsigned right)
{
struct ac97mixtable_entry *e = &codec->mix[channel];
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
if (e->reg && e->enable && e->bits) {
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
int mask, max, val, reg;
reg = (e->reg >= 0) ? e->reg : -e->reg; /* AC97 register */
max = (1 << e->bits) - 1; /* actual range */
mask = (max << 8) | max; /* bits of interest */
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (!e->stereo)
right = left;
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
/*
* Invert the range if the polarity requires so,
* then scale to 0..max-1 to compute the value to
* write into the codec, and scale back to 0..100
* for the return value.
*/
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
if (e->reg > 0) {
left = 100 - left;
right = 100 - right;
}
left = (left * max) / 100;
right = (right * max) / 100;
val = (left << 8) | right;
left = (left * 100) / max;
right = (right * 100) / max;
if (e->reg > 0) {
left = 100 - left;
right = 100 - right;
}
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
/*
* For mono controls, trim val and mask, also taking
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
* care of e->ofs (offset of control field).
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
*/
if (e->ofs) {
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
val &= max;
val <<= e->ofs;
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
mask = (max << e->ofs);
}
/*
* If we have a mute bit, add it to the mask and
* update val and set mute if both channels require a
* zero volume.
*/
if (e->mute == 1) {
mask |= AC97_MUTE;
if (left == 0 && right == 0)
val = AC97_MUTE;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
}
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
/*
* If the mask bit is set, do not alter the other bits.
*/
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
Use SOUND_MIXER_IGAIN to toggle 20dB mic boost for want of a better place for it. Assorted comments. Submitted by: luigi
2003-04-21 03:47:28 +00:00
if (e->mask) {
int cur = ac97_rdcd(codec, e->reg);
val |= cur & ~(mask);
}
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, reg, val);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return left | (right << 8);
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
} else {
/* printf("ac97_setmixer: reg=%d, bits=%d, enable=%d\n", e->reg, e->bits, e->enable); */
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
return -1;
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
}
s/AC97_MIX_PHONES/AC97_MIX_AUXOUT/ to match ac97r2.{2,3}. Attempt to determine what function of AUX_OUT is: "True line level out", "Headphone out", or "4-Channel out" and frig OSS mixer label accordingly. Addresses problem raised by Randy Bush on -multimedia of not being able to hear audio on ich2 m/b which was eventually found to be because the mixer monitor value was 0. On this h/w the label "monitor" should now be presented as the marginally more intuitive "ogain".
2002-08-23 06:19:28 +00:00
static void
ac97_fix_auxout(struct ac97_info *codec)
{
When present use ogain instead of master for surround sound channels. The latter has lead to reports of broken audio. Do not swap ogain and master when headphones detected.
2003-08-23 21:39:51 +00:00
int keep_ogain;
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
/*
When present use ogain instead of master for surround sound channels. The latter has lead to reports of broken audio. Do not swap ogain and master when headphones detected.
2003-08-23 21:39:51 +00:00
* By default, The ac97 aux_out register (0x04) corresponds to OSS's
* OGAIN setting.
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
*
When present use ogain instead of master for surround sound channels. The latter has lead to reports of broken audio. Do not swap ogain and master when headphones detected.
2003-08-23 21:39:51 +00:00
* We first check whether aux_out is a valid register. If not
* we may not want to keep ogain.
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
*/
When present use ogain instead of master for surround sound channels. The latter has lead to reports of broken audio. Do not swap ogain and master when headphones detected.
2003-08-23 21:39:51 +00:00
keep_ogain = ac97_rdcd(codec, AC97_MIX_AUXOUT) & 0x8000;
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
/*
* Determine what AUX_OUT really means, it can be:
s/AC97_MIX_PHONES/AC97_MIX_AUXOUT/ to match ac97r2.{2,3}. Attempt to determine what function of AUX_OUT is: "True line level out", "Headphone out", or "4-Channel out" and frig OSS mixer label accordingly. Addresses problem raised by Randy Bush on -multimedia of not being able to hear audio on ich2 m/b which was eventually found to be because the mixer monitor value was 0. On this h/w the label "monitor" should now be presented as the marginally more intuitive "ogain".
2002-08-23 06:19:28 +00:00
*
* 1. Headphone out.
* 2. 4-Channel Out
* 3. True line level out (effectively master volume).
*
* See Sections 5.2.1 and 5.27 for AUX_OUT Options in AC97r2.{2,3}.
*/
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
if (codec->extcaps & AC97_EXTCAP_SDAC &&
ac97_rdcd(codec, AC97_MIXEXT_SURROUND) == 0x8080) {
When present use ogain instead of master for surround sound channels. The latter has lead to reports of broken audio. Do not swap ogain and master when headphones detected.
2003-08-23 21:39:51 +00:00
codec->mix[SOUND_MIXER_OGAIN].reg = AC97_MIXEXT_SURROUND;
keep_ogain = 1;
}
if (keep_ogain == 0) {
bzero(&codec->mix[SOUND_MIXER_OGAIN],
sizeof(codec->mix[SOUND_MIXER_OGAIN]));
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
}
}
static void
ac97_fix_tone(struct ac97_info *codec)
{
/* Hide treble and bass if they don't exist */
if ((codec->caps & AC97_CAP_TONE) == 0) {
bzero(&codec->mix[SOUND_MIXER_BASS],
sizeof(codec->mix[SOUND_MIXER_BASS]));
bzero(&codec->mix[SOUND_MIXER_TREBLE],
sizeof(codec->mix[SOUND_MIXER_TREBLE]));
s/AC97_MIX_PHONES/AC97_MIX_AUXOUT/ to match ac97r2.{2,3}. Attempt to determine what function of AUX_OUT is: "True line level out", "Headphone out", or "4-Channel out" and frig OSS mixer label accordingly. Addresses problem raised by Randy Bush on -multimedia of not being able to hear audio on ich2 m/b which was eventually found to be because the mixer monitor value was 0. On this h/w the label "monitor" should now be presented as the marginally more intuitive "ogain".
2002-08-23 06:19:28 +00:00
}
}
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
static const char*
ac97_hw_desc(u_int32_t id, const char* vname, const char* cname, char* buf)
{
if (cname == NULL) {
sprintf(buf, "Unknown AC97 Codec (id = 0x%08x)", id);
return buf;
}
if (vname == NULL) vname = "Unknown";
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
if (bootverbose) {
sprintf(buf, "%s %s AC97 Codec (id = 0x%08x)", vname, cname, id);
} else {
sprintf(buf, "%s %s AC97 Codec", vname, cname);
}
return buf;
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
static unsigned
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
ac97_initmixer(struct ac97_info *codec)
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
{
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_patch codec_patch;
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
const char *cname, *vname;
char desc[80];
u_int8_t model, step;
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
unsigned i, j, k, old;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
u_int32_t id;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
codec->count = AC97_INIT(codec->methods, codec->devinfo);
if (codec->count == 0) {
device_printf(codec->dev, "ac97 codec init failed\n");
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
return ENODEV;
}
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REG_POWER, (codec->flags & AC97_F_EAPD_INV)? 0x8000 : 0x0000);
Add ac97_reset function that polls ready bits in power register. One component in tweaks required for the ICH3 controller in the thinkpad x22 reported by Colin Perkins on -multimedia.
2002-04-26 15:27:56 +00:00
ac97_reset(codec);
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REG_POWER, (codec->flags & AC97_F_EAPD_INV)? 0x8000 : 0x0000);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
i = ac97_rdcd(codec, AC97_REG_RESET);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
codec->caps = i & 0x03ff;
codec->se = (i & 0x7c00) >> 10;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
id = (ac97_rdcd(codec, AC97_REG_ID1) << 16) | ac97_rdcd(codec, AC97_REG_ID2);
fail in attach if we seem to have no ac97 codec
2000-04-01 22:24:03 +00:00
if (id == 0 || id == 0xffffffff) {
device_printf(codec->dev, "ac97 codec invalid or not present (id == %x)\n", id);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
fail in attach if we seem to have no ac97 codec
2000-04-01 22:24:03 +00:00
return ENODEV;
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
codec->id = id;
initial support for multiple ac97 codecs
2000-09-05 21:08:01 +00:00
codec->noext = 0;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
codec_patch = NULL;
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
cname = NULL;
model = step = 0;
initial support for multiple ac97 codecs
2000-09-05 21:08:01 +00:00
for (i = 0; ac97codecid[i].id; i++) {
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
u_int32_t modelmask = 0xffffffff ^ ac97codecid[i].stepmask;
if ((ac97codecid[i].id & modelmask) == (id & modelmask)) {
initial support for multiple ac97 codecs
2000-09-05 21:08:01 +00:00
codec->noext = ac97codecid[i].noext;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
codec_patch = ac97codecid[i].patch;
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
cname = ac97codecid[i].name;
model = (id & modelmask) & 0xff;
step = (id & ~modelmask) & 0xff;
break;
}
}
vname = NULL;
for (i = 0; ac97vendorid[i].id; i++) {
if (ac97vendorid[i].id == (id & 0xffffff00)) {
vname = ac97vendorid[i].name;
break;
initial support for multiple ac97 codecs
2000-09-05 21:08:01 +00:00
}
}
don't check extended features on codecs known to hang. should fix neomagic problems.
2000-03-24 03:14:35 +00:00
compensate for broken codecs that return 0xffff on unimplemented registers instead of 0.
2000-12-24 03:33:21 +00:00
codec->extcaps = 0;
codec->extid = 0;
codec->extstat = 0;
if (!codec->noext) {
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
i = ac97_rdcd(codec, AC97_REGEXT_ID);
compensate for broken codecs that return 0xffff on unimplemented registers instead of 0.
2000-12-24 03:33:21 +00:00
if (i != 0xffff) {
codec->extcaps = i & 0x3fff;
codec->extid = (i & 0xc000) >> 14;
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
codec->extstat = ac97_rdcd(codec, AC97_REGEXT_STAT) & AC97_EXTCAPS;
compensate for broken codecs that return 0xffff on unimplemented registers instead of 0.
2000-12-24 03:33:21 +00:00
}
don't check extended features on codecs known to hang. should fix neomagic problems.
2000-03-24 03:14:35 +00:00
}
s/AC97_MIX_PHONES/AC97_MIX_AUXOUT/ to match ac97r2.{2,3}. Attempt to determine what function of AUX_OUT is: "True line level out", "Headphone out", or "4-Channel out" and frig OSS mixer label accordingly. Addresses problem raised by Randy Bush on -multimedia of not being able to hear audio on ich2 m/b which was eventually found to be because the mixer monitor value was 0. On this h/w the label "monitor" should now be presented as the marginally more intuitive "ogain".
2002-08-23 06:19:28 +00:00
for (i = 0; i < 32; i++) {
codec->mix[i] = ac97mixtable_default[i];
}
ac97_fix_auxout(codec);
Apply Rudolf Cejka's patch for: o AD1980 hook. o ac97_fix_auxout. and: o Associate AC97_MIX_AUXOUT with SOUND_MIXER_OGAIN rather than SOUND_MIXER_MONITOR. o Add ac97_fix_tone to remove tone controls from mixer if invalid.
2003-08-21 15:50:32 +00:00
ac97_fix_tone(codec);
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
if (codec_patch)
codec_patch(codec);
s/AC97_MIX_PHONES/AC97_MIX_AUXOUT/ to match ac97r2.{2,3}. Attempt to determine what function of AUX_OUT is: "True line level out", "Headphone out", or "4-Channel out" and frig OSS mixer label accordingly. Addresses problem raised by Randy Bush on -multimedia of not being able to hear audio on ich2 m/b which was eventually found to be because the mixer monitor value was 0. On this h/w the label "monitor" should now be presented as the marginally more intuitive "ogain".
2002-08-23 06:19:28 +00:00
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
for (i = 0; i < 32; i++) {
only probe mandatory channels on known bad codecs
2000-09-23 22:00:09 +00:00
k = codec->noext? codec->mix[i].enable : 1;
if (k && (codec->mix[i].reg > 0)) {
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
old = ac97_rdcd(codec, codec->mix[i].reg);
ac97_wrcd(codec, codec->mix[i].reg, 0x3f);
j = ac97_rdcd(codec, codec->mix[i].reg);
ac97_wrcd(codec, codec->mix[i].reg, old);
compensate for broken codecs that return 0xffff on unimplemented registers instead of 0.
2000-12-24 03:33:21 +00:00
codec->mix[i].enable = (j != 0 && j != old)? 1 : 0;
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
for (k = 1; j & (1 << k); k++);
codec->mix[i].bits = j? k - codec->mix[i].ofs : 0;
}
/* printf("mixch %d, en=%d, b=%d\n", i, codec->mix[i].enable, codec->mix[i].bits); */
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
device_printf(codec->dev, "<%s>\n",
ac97_hw_desc(codec->id, vname, cname, desc));
Print ac97 name/id on normal boot. Fix typo reported in pr misc/47794.
2003-02-03 01:38:43 +00:00
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
if (bootverbose) {
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
device_printf(codec->dev, "Codec features ");
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
for (i = j = 0; i < 10; i++)
if (codec->caps & (1 << i))
printf("%s%s", j++? ", " : "", ac97feature[i]);
printf("%s%d bit master volume", j++? ", " : "", codec->mix[SOUND_MIXER_VOLUME].bits);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
printf("%s%s\n", j? ", " : "", ac97enhancement[codec->se]);
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (codec->extcaps != 0 || codec->extid) {
o Move vendor ids into separate table. o Add assorted vendor and codec ids.
2003-02-28 18:04:52 +00:00
device_printf(codec->dev, "%s codec",
codec->extid? "Secondary" : "Primary");
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (codec->extcaps)
printf(" extended features ");
for (i = j = 0; i < 14; i++)
if (codec->extcaps & (1 << i))
printf("%s%s", j++? ", " : "", ac97extfeature[i]);
printf("\n");
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
}
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
if ((ac97_rdcd(codec, AC97_REG_POWER) & 2) == 0)
update ac97 layer to use device_printf when printing messages
2000-01-18 17:13:43 +00:00
device_printf(codec->dev, "ac97 codec reports dac not ready\n");
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return 0;
}
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
static unsigned
ac97_reinitmixer(struct ac97_info *codec)
{
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
codec->count = AC97_INIT(codec->methods, codec->devinfo);
if (codec->count == 0) {
device_printf(codec->dev, "ac97 codec init failed\n");
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
return ENODEV;
}
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REG_POWER, (codec->flags & AC97_F_EAPD_INV)? 0x8000 : 0x0000);
Add ac97_reset function that polls ready bits in power register. One component in tweaks required for the ICH3 controller in the thinkpad x22 reported by Colin Perkins on -multimedia.
2002-04-26 15:27:56 +00:00
ac97_reset(codec);
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REG_POWER, (codec->flags & AC97_F_EAPD_INV)? 0x8000 : 0x0000);
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
if (!codec->noext) {
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_wrcd(codec, AC97_REGEXT_STAT, codec->extstat);
if ((ac97_rdcd(codec, AC97_REGEXT_STAT) & AC97_EXTCAPS)
Get rid of irritating (bogus) message: pcm0: ac97 codec failed to reset extended mode (0, got 80) This was due to not masking off the correct extended function bits in the read value from the extended status reg. MFC after: 2 days
2001-12-10 20:27:18 +00:00
!= codec->extstat)
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
device_printf(codec->dev, "ac97 codec failed to reset extended mode (%x, got %x)\n",
Get rid of irritating (bogus) message: pcm0: ac97 codec failed to reset extended mode (0, got 80) This was due to not masking off the correct extended function bits in the read value from the extended status reg. MFC after: 2 days
2001-12-10 20:27:18 +00:00
codec->extstat,
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
ac97_rdcd(codec, AC97_REGEXT_STAT) &
AC97_EXTCAPS);
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
}
Add ac97_patch.[ch] that provide space for ac97 codec specific patches.
2003-01-25 16:54:05 +00:00
if ((ac97_rdcd(codec, AC97_REG_POWER) & 2) == 0)
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
device_printf(codec->dev, "ac97 codec reports dac not ready\n");
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxunlock(codec->lock);
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
return 0;
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
struct ac97_info *
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
ac97_create(device_t dev, void *devinfo, kobj_class_t cls)
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
{
struct ac97_info *codec;
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
codec = (struct ac97_info *)malloc(sizeof *codec, M_AC97, M_NOWAIT);
if (codec == NULL)
return NULL;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snprintf(codec->name, AC97_NAMELEN, "%s:ac97", device_get_nameunit(dev));
Add lock type arguments to callers of snd_mtxcreate().
2002-04-04 20:56:47 +00:00
codec->lock = snd_mtxcreate(codec->name, "ac97 codec");
Back out M_* changes, per decision of the TRB. Approved by: trb
2003-02-19 05:47:46 +00:00
codec->methods = kobj_create(cls, M_AC97, M_WAITOK);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
if (codec->methods == NULL) {
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
snd_mtxfree(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
free(codec, M_AC97);
return NULL;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
}
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
codec->dev = dev;
codec->devinfo = devinfo;
move a private struct definition from ac97.h to ac97.c add proper handling of cards which take eapd=1 to mean 'produce sound' instead of 'external amplifier power down'
2001-06-23 17:32:40 +00:00
codec->flags = 0;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return codec;
}
change mixer api slightly change channel interface - kobj implementation coming soonish make pcm_makelinks not panic if modular add pcm_unregister() these changes support newpcm kld unloading, but this is only implemented by ds1.c
2000-09-01 20:09:24 +00:00
void
ac97_destroy(struct ac97_info *codec)
{
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxlock(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
if (codec->methods != NULL)
kobj_delete(codec->methods, M_AC97);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
snd_mtxfree(codec->lock);
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
free(codec, M_AC97);
change mixer api slightly change channel interface - kobj implementation coming soonish make pcm_makelinks not panic if modular add pcm_unregister() these changes support newpcm kld unloading, but this is only implemented by ds1.c
2000-09-01 20:09:24 +00:00
}
move a private struct definition from ac97.h to ac97.c add proper handling of cards which take eapd=1 to mean 'produce sound' instead of 'external amplifier power down'
2001-06-23 17:32:40 +00:00
void
ac97_setflags(struct ac97_info *codec, u_int32_t val)
{
codec->flags = val;
}
u_int32_t
ac97_getflags(struct ac97_info *codec)
{
return codec->flags;
}
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
/* -------------------------------------------------------------------- */
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
ac97mix_init(struct snd_mixer *m)
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
{
struct ac97_info *codec = mix_getdevinfo(m);
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
u_int32_t i, mask;
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (codec == NULL)
return -1;
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
fail in attach if we seem to have no ac97 codec
2000-04-01 22:24:03 +00:00
if (ac97_initmixer(codec))
return -1;
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
mask = 0;
for (i = 0; i < 32; i++)
mask |= codec->mix[i].enable? 1 << i : 0;
mix_setdevs(m, mask);
mask = 0;
for (i = 0; i < 32; i++)
mask |= codec->mix[i].recidx? 1 << i : 0;
mix_setrecdevs(m, mask);
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return 0;
}
change mixer api slightly change channel interface - kobj implementation coming soonish make pcm_makelinks not panic if modular add pcm_unregister() these changes support newpcm kld unloading, but this is only implemented by ds1.c
2000-09-01 20:09:24 +00:00
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
ac97mix_uninit(struct snd_mixer *m)
change mixer api slightly change channel interface - kobj implementation coming soonish make pcm_makelinks not panic if modular add pcm_unregister() these changes support newpcm kld unloading, but this is only implemented by ds1.c
2000-09-01 20:09:24 +00:00
{
struct ac97_info *codec = mix_getdevinfo(m);
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
change mixer api slightly change channel interface - kobj implementation coming soonish make pcm_makelinks not panic if modular add pcm_unregister() these changes support newpcm kld unloading, but this is only implemented by ds1.c
2000-09-01 20:09:24 +00:00
if (codec == NULL)
return -1;
/*
if (ac97_uninitmixer(codec))
return -1;
*/
ac97_destroy(codec);
return 0;
}
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
ac97mix_reinit(struct snd_mixer *m)
add reinit functions to mixers unstaticize chn_start() add reset/resetdone functions to channels
2000-10-26 20:46:58 +00:00
{
struct ac97_info *codec = mix_getdevinfo(m);
if (codec == NULL)
return -1;
return ac97_reinitmixer(codec);
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
ac97mix_set(struct snd_mixer *m, unsigned dev, unsigned left, unsigned right)
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
{
struct ac97_info *codec = mix_getdevinfo(m);
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (codec == NULL)
return -1;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return ac97_setmixer(codec, dev, left, right);
}
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
ac97mix_setrecsrc(struct snd_mixer *m, u_int32_t src)
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
{
int i;
struct ac97_info *codec = mix_getdevinfo(m);
dynamify- probe codec for supported channels add channel mappings for mixer
2000-09-17 23:51:00 +00:00
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if (codec == NULL)
return -1;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
for (i = 0; i < SOUND_MIXER_NRDEVICES; i++)
update the ac97 layer: * add a callback for initialising the mixer interface * support ac97 2.1 variable rate audio feature fix ac97-using drivers for the above add suspend/resume support for neomagic
2000-03-20 15:30:50 +00:00
if ((src & (1 << i)) != 0)
break;
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
return (ac97_setrecsrc(codec, i) == 0)? 1 << i : -1;
}
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
static kobj_method_t ac97mixer_methods[] = {
KOBJMETHOD(mixer_init, ac97mix_init),
KOBJMETHOD(mixer_uninit, ac97mix_uninit),
KOBJMETHOD(mixer_reinit, ac97mix_reinit),
KOBJMETHOD(mixer_set, ac97mix_set),
KOBJMETHOD(mixer_setrecsrc, ac97mix_setrecsrc),
{ 0, 0 }
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
};
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
MIXER_DECLARE(ac97mixer);
/* -------------------------------------------------------------------- */
kobj_class_t
ac97_getmixerclass(void)
{
return &ac97mixer_class;
}
say hello to newpcm. it is not yet enabled, requiring new pnp code from dfr to compile successfully. further details will be provided in the commit enabling newpcm.
1999-09-01 04:08:39 +00:00
Reference in New Issue Copy Permalink