freebsd-nq/sys/dev/sound/pci/neomagic.c

823 lines
19 KiB
C
Raw Normal View History

/*-
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
*
2003-09-07 16:28:03 +00:00
* Copyright (c) 1999 Cameron Grant <cg@freebsd.org>
* All rights reserved.
*
* Derived from the public domain Linux driver
*
* 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, WHETHERIN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THEPOSSIBILITY OF
* SUCH DAMAGE.
*/
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
#ifdef HAVE_KERNEL_OPTION_HEADERS
#include "opt_snd.h"
#endif
#include <dev/sound/pcm/sound.h>
#include <dev/sound/pcm/ac97.h>
#include <dev/sound/pci/neomagic.h>
#include <dev/sound/pci/neomagic-coeff.h>
#include <dev/pci/pcireg.h>
#include <dev/pci/pcivar.h>
SND_DECLARE_FILE("$FreeBSD$");
/* -------------------------------------------------------------------- */
#define NM_BUFFSIZE 16384
#define NM256AV_PCI_ID 0x800510c8
#define NM256ZX_PCI_ID 0x800610c8
struct sc_info;
/* channel registers */
struct sc_chinfo {
int active, spd, dir, fmt;
u_int32_t blksize, wmark;
struct snd_dbuf *buffer;
struct pcm_channel *channel;
struct sc_info *parent;
};
/* device private data */
struct sc_info {
device_t dev;
u_int32_t type;
struct resource *reg, *irq, *buf;
int regid, irqid, bufid;
void *ih;
u_int32_t ac97_base, ac97_status, ac97_busy;
u_int32_t buftop, pbuf, rbuf, cbuf, acbuf;
u_int32_t playint, recint, misc1int, misc2int;
u_int32_t irsz, badintr;
struct sc_chinfo pch, rch;
};
/* -------------------------------------------------------------------- */
/*
* prototypes
*/
/* stuff */
static int nm_loadcoeff(struct sc_info *sc, int dir, int num);
static int nm_setch(struct sc_chinfo *ch);
static int nm_init(struct sc_info *);
static void nm_intr(void *);
/* talk to the card */
static u_int32_t nm_rd(struct sc_info *, int, int);
static void nm_wr(struct sc_info *, int, u_int32_t, int);
static u_int32_t nm_rdbuf(struct sc_info *, int, int);
static void nm_wrbuf(struct sc_info *, int, u_int32_t, int);
static u_int32_t badcards[] = {
0x0007103c,
0x008f1028,
0x00dd1014,
0x8005110a,
};
#define NUM_BADCARDS (sizeof(badcards) / sizeof(u_int32_t))
/* The actual rates supported by the card. */
static int samplerates[9] = {
8000,
11025,
16000,
22050,
24000,
32000,
44100,
48000,
99999999
};
/* -------------------------------------------------------------------- */
static u_int32_t nm_fmt[] = {
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
SND_FORMAT(AFMT_U8, 1, 0),
SND_FORMAT(AFMT_U8, 2, 0),
SND_FORMAT(AFMT_S16_LE, 1, 0),
SND_FORMAT(AFMT_S16_LE, 2, 0),
0
};
static struct pcmchan_caps nm_caps = {4000, 48000, nm_fmt, 0};
/* -------------------------------------------------------------------- */
/* Hardware */
static u_int32_t
nm_rd(struct sc_info *sc, int regno, int size)
{
bus_space_tag_t st = rman_get_bustag(sc->reg);
bus_space_handle_t sh = rman_get_bushandle(sc->reg);
switch (size) {
case 1:
return bus_space_read_1(st, sh, regno);
case 2:
return bus_space_read_2(st, sh, regno);
case 4:
return bus_space_read_4(st, sh, regno);
default:
return 0xffffffff;
}
}
static void
nm_wr(struct sc_info *sc, int regno, u_int32_t data, int size)
{
bus_space_tag_t st = rman_get_bustag(sc->reg);
bus_space_handle_t sh = rman_get_bushandle(sc->reg);
switch (size) {
case 1:
bus_space_write_1(st, sh, regno, data);
break;
case 2:
bus_space_write_2(st, sh, regno, data);
break;
case 4:
bus_space_write_4(st, sh, regno, data);
break;
}
}
static u_int32_t
nm_rdbuf(struct sc_info *sc, int regno, int size)
{
bus_space_tag_t st = rman_get_bustag(sc->buf);
bus_space_handle_t sh = rman_get_bushandle(sc->buf);
switch (size) {
case 1:
return bus_space_read_1(st, sh, regno);
case 2:
return bus_space_read_2(st, sh, regno);
case 4:
return bus_space_read_4(st, sh, regno);
default:
return 0xffffffff;
}
}
static void
nm_wrbuf(struct sc_info *sc, int regno, u_int32_t data, int size)
{
bus_space_tag_t st = rman_get_bustag(sc->buf);
bus_space_handle_t sh = rman_get_bushandle(sc->buf);
switch (size) {
case 1:
bus_space_write_1(st, sh, regno, data);
break;
case 2:
bus_space_write_2(st, sh, regno, data);
break;
case 4:
bus_space_write_4(st, sh, regno, data);
break;
}
}
/* -------------------------------------------------------------------- */
/* ac97 codec */
static int
nm_waitcd(struct sc_info *sc)
{
int cnt = 10;
int fail = 1;
while (cnt-- > 0) {
if (nm_rd(sc, sc->ac97_status, 2) & sc->ac97_busy) {
DELAY(100);
} else {
fail = 0;
break;
}
}
return (fail);
}
static u_int32_t
nm_initcd(kobj_t obj, void *devinfo)
{
struct sc_info *sc = (struct sc_info *)devinfo;
nm_wr(sc, 0x6c0, 0x01, 1);
#if 0
/*
* The following code-line may cause a hang for some chipsets, see
* PR 56617.
* In case of a bugreport without this line have a look at the PR and
* conditionize the code-line based upon the specific version of
* the chip.
*/
nm_wr(sc, 0x6cc, 0x87, 1);
#endif
nm_wr(sc, 0x6cc, 0x80, 1);
nm_wr(sc, 0x6cc, 0x00, 1);
return 1;
}
static int
nm_rdcd(kobj_t obj, void *devinfo, int regno)
{
struct sc_info *sc = (struct sc_info *)devinfo;
u_int32_t x;
if (!nm_waitcd(sc)) {
x = nm_rd(sc, sc->ac97_base + regno, 2);
DELAY(1000);
return x;
} else {
device_printf(sc->dev, "ac97 codec not ready\n");
return -1;
}
}
static int
nm_wrcd(kobj_t obj, void *devinfo, int regno, u_int32_t data)
{
struct sc_info *sc = (struct sc_info *)devinfo;
int cnt = 3;
if (!nm_waitcd(sc)) {
while (cnt-- > 0) {
nm_wr(sc, sc->ac97_base + regno, data, 2);
if (!nm_waitcd(sc)) {
DELAY(1000);
return 0;
}
}
}
device_printf(sc->dev, "ac97 codec not ready\n");
return -1;
}
static kobj_method_t nm_ac97_methods[] = {
KOBJMETHOD(ac97_init, nm_initcd),
KOBJMETHOD(ac97_read, nm_rdcd),
KOBJMETHOD(ac97_write, nm_wrcd),
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
KOBJMETHOD_END
};
AC97_DECLARE(nm_ac97);
/* -------------------------------------------------------------------- */
static void
nm_ackint(struct sc_info *sc, u_int32_t num)
{
if (sc->type == NM256AV_PCI_ID) {
nm_wr(sc, NM_INT_REG, num << 1, 2);
} else if (sc->type == NM256ZX_PCI_ID) {
nm_wr(sc, NM_INT_REG, num, 4);
}
}
static int
nm_loadcoeff(struct sc_info *sc, int dir, int num)
{
int ofs, sz, i;
u_int32_t addr;
addr = (dir == PCMDIR_PLAY)? 0x01c : 0x21c;
if (dir == PCMDIR_REC)
num += 8;
sz = coefficientSizes[num];
ofs = 0;
while (num-- > 0)
ofs+= coefficientSizes[num];
for (i = 0; i < sz; i++)
nm_wrbuf(sc, sc->cbuf + i, coefficients[ofs + i], 1);
nm_wr(sc, addr, sc->cbuf, 4);
if (dir == PCMDIR_PLAY)
sz--;
nm_wr(sc, addr + 4, sc->cbuf + sz, 4);
return 0;
}
static int
nm_setch(struct sc_chinfo *ch)
{
struct sc_info *sc = ch->parent;
u_int32_t base;
u_int8_t x;
for (x = 0; x < 8; x++)
if (ch->spd < (samplerates[x] + samplerates[x + 1]) / 2)
break;
if (x == 8) return 1;
ch->spd = samplerates[x];
nm_loadcoeff(sc, ch->dir, x);
x <<= 4;
x &= NM_RATE_MASK;
if (ch->fmt & AFMT_16BIT) x |= NM_RATE_BITS_16;
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
if (AFMT_CHANNEL(ch->fmt) > 1) x |= NM_RATE_STEREO;
base = (ch->dir == PCMDIR_PLAY)? NM_PLAYBACK_REG_OFFSET : NM_RECORD_REG_OFFSET;
nm_wr(sc, base + NM_RATE_REG_OFFSET, x, 1);
return 0;
}
/* channel interface */
static void *
nmchan_init(kobj_t obj, void *devinfo, struct snd_dbuf *b, struct pcm_channel *c, int dir)
{
struct sc_info *sc = devinfo;
struct sc_chinfo *ch;
u_int32_t chnbuf;
chnbuf = (dir == PCMDIR_PLAY)? sc->pbuf : sc->rbuf;
ch = (dir == PCMDIR_PLAY)? &sc->pch : &sc->rch;
ch->active = 0;
ch->blksize = 0;
ch->wmark = 0;
ch->buffer = b;
sndbuf_setup(ch->buffer, (u_int8_t *)rman_get_virtual(sc->buf) + chnbuf, NM_BUFFSIZE);
2000-01-13 06:00:57 +00:00
if (bootverbose)
device_printf(sc->dev, "%s buf %p\n", (dir == PCMDIR_PLAY)?
"play" : "rec", sndbuf_getbuf(ch->buffer));
ch->parent = sc;
ch->channel = c;
ch->dir = dir;
return ch;
}
static int
nmchan_free(kobj_t obj, void *data)
{
return 0;
}
static int
nmchan_setformat(kobj_t obj, void *data, u_int32_t format)
{
struct sc_chinfo *ch = data;
ch->fmt = format;
return nm_setch(ch);
}
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
static u_int32_t
nmchan_setspeed(kobj_t obj, void *data, u_int32_t speed)
{
struct sc_chinfo *ch = data;
ch->spd = speed;
return nm_setch(ch)? 0 : ch->spd;
}
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
static u_int32_t
nmchan_setblocksize(kobj_t obj, void *data, u_int32_t blocksize)
{
struct sc_chinfo *ch = data;
ch->blksize = blocksize;
return blocksize;
}
static int
nmchan_trigger(kobj_t obj, void *data, int go)
{
struct sc_chinfo *ch = data;
struct sc_info *sc = ch->parent;
int ssz;
if (!PCMTRIG_COMMON(go))
return 0;
ssz = (ch->fmt & AFMT_16BIT)? 2 : 1;
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
if (AFMT_CHANNEL(ch->fmt) > 1)
ssz <<= 1;
if (ch->dir == PCMDIR_PLAY) {
if (go == PCMTRIG_START) {
ch->active = 1;
ch->wmark = ch->blksize;
nm_wr(sc, NM_PBUFFER_START, sc->pbuf, 4);
nm_wr(sc, NM_PBUFFER_END, sc->pbuf + NM_BUFFSIZE - ssz, 4);
nm_wr(sc, NM_PBUFFER_CURRP, sc->pbuf, 4);
nm_wr(sc, NM_PBUFFER_WMARK, sc->pbuf + ch->wmark, 4);
nm_wr(sc, NM_PLAYBACK_ENABLE_REG, NM_PLAYBACK_FREERUN |
NM_PLAYBACK_ENABLE_FLAG, 1);
nm_wr(sc, NM_AUDIO_MUTE_REG, 0, 2);
} else {
ch->active = 0;
nm_wr(sc, NM_PLAYBACK_ENABLE_REG, 0, 1);
nm_wr(sc, NM_AUDIO_MUTE_REG, NM_AUDIO_MUTE_BOTH, 2);
}
} else {
if (go == PCMTRIG_START) {
ch->active = 1;
ch->wmark = ch->blksize;
nm_wr(sc, NM_RECORD_ENABLE_REG, NM_RECORD_FREERUN |
NM_RECORD_ENABLE_FLAG, 1);
nm_wr(sc, NM_RBUFFER_START, sc->rbuf, 4);
nm_wr(sc, NM_RBUFFER_END, sc->rbuf + NM_BUFFSIZE, 4);
nm_wr(sc, NM_RBUFFER_CURRP, sc->rbuf, 4);
nm_wr(sc, NM_RBUFFER_WMARK, sc->rbuf + ch->wmark, 4);
} else {
ch->active = 0;
nm_wr(sc, NM_RECORD_ENABLE_REG, 0, 1);
}
}
return 0;
}
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
static u_int32_t
nmchan_getptr(kobj_t obj, void *data)
{
struct sc_chinfo *ch = data;
struct sc_info *sc = ch->parent;
if (ch->dir == PCMDIR_PLAY)
return nm_rd(sc, NM_PBUFFER_CURRP, 4) - sc->pbuf;
else
return nm_rd(sc, NM_RBUFFER_CURRP, 4) - sc->rbuf;
}
static struct pcmchan_caps *
nmchan_getcaps(kobj_t obj, void *data)
{
return &nm_caps;
}
static kobj_method_t nmchan_methods[] = {
KOBJMETHOD(channel_init, nmchan_init),
KOBJMETHOD(channel_free, nmchan_free),
KOBJMETHOD(channel_setformat, nmchan_setformat),
KOBJMETHOD(channel_setspeed, nmchan_setspeed),
KOBJMETHOD(channel_setblocksize, nmchan_setblocksize),
KOBJMETHOD(channel_trigger, nmchan_trigger),
KOBJMETHOD(channel_getptr, nmchan_getptr),
KOBJMETHOD(channel_getcaps, nmchan_getcaps),
Sound Mega-commit. Expect further cleanup until code freeze. For a slightly thorough explaination, please refer to [1] http://people.freebsd.org/~ariff/SOUND_4.TXT.html . Summary of changes includes: 1 Volume Per-Channel (vpc). Provides private / standalone volume control unique per-stream pcm channel without touching master volume / pcm. Applications can directly use SNDCTL_DSP_[GET|SET][PLAY|REC]VOL, or for backwards compatibility, SOUND_MIXER_PCM through the opened dsp device instead of /dev/mixer. Special "bypass" mode is enabled through /dev/mixer which will automatically detect if the adjustment is made through /dev/mixer and forward its request to this private volume controller. Changes to this volume object will not interfere with other channels. Requirements: - SNDCTL_DSP_[GET|SET][PLAY|REC]_VOL are newer ioctls (OSSv4) which require specific application modifications (preferred). - No modifications required for using bypass mode, so applications like mplayer or xmms should work out of the box. Kernel hints: - hint.pcm.%d.vpc (0 = disable vpc). Kernel sysctls: - hw.snd.vpc_mixer_bypass (default: 1). Enable or disable /dev/mixer bypass mode. - hw.snd.vpc_autoreset (default: 1). By default, closing/opening /dev/dsp will reset the volume back to 0 db gain/attenuation. Setting this to 0 will preserve its settings across device closing/opening. - hw.snd.vpc_reset (default: 0). Panic/reset button to reset all volume settings back to 0 db. - hw.snd.vpc_0db (default: 45). 0 db relative to linear mixer value. 2 High quality fixed-point Bandlimited SINC sampling rate converter, based on Julius O'Smith's Digital Audio Resampling - http://ccrma.stanford.edu/~jos/resample/. It includes a filter design script written in awk (the clumsiest joke I've ever written) - 100% 32bit fixed-point, 64bit accumulator. - Possibly among the fastest (if not fastest) of its kind. - Resampling quality is tunable, either runtime or during kernel compilation (FEEDER_RATE_PRESETS). - Quality can be further customized during kernel compilation by defining FEEDER_RATE_PRESETS in /etc/make.conf. Kernel sysctls: - hw.snd.feeder_rate_quality. 0 - Zero-order Hold (ZOH). Fastest, bad quality. 1 - Linear Interpolation (LINEAR). Slightly slower than ZOH, better quality but still does not eliminate aliasing. 2 - (and above) - Sinc Interpolation(SINC). Best quality. SINC quality always start from 2 and above. Rough quality comparisons: - http://people.freebsd.org/~ariff/z_comparison/ 3 Bit-perfect mode. Bypasses all feeder/dsp effects. Pure sound will be directly fed into the hardware. 4 Parametric (compile time) Software Equalizer (Bass/Treble mixer). Can be customized by defining FEEDER_EQ_PRESETS in /etc/make.conf. 5 Transparent/Adaptive Virtual Channel. Now you don't have to disable vchans in order to make digital format pass through. It also makes vchans more dynamic by choosing a better format/rate among all the concurrent streams, which means that dev.pcm.X.play.vchanformat/rate becomes sort of optional. 6 Exclusive Stream, with special open() mode O_EXCL. This will "mute" other concurrent vchan streams and only allow a single channel with O_EXCL set to keep producing sound. Other Changes: * most feeder_* stuffs are compilable in userland. Let's not speculate whether we should go all out for it (save that for FreeBSD 16.0-RELEASE). * kobj signature fixups, thanks to Andriy Gapon <avg@freebsd.org> * pull out channel mixing logic out of vchan.c and create its own feeder_mixer for world justice. * various refactoring here and there, for good or bad. * activation of few more OSSv4 ioctls() (see [1] above). * opt_snd.h for possible compile time configuration: (mostly for debugging purposes, don't try these at home) SND_DEBUG SND_DIAGNOSTIC SND_FEEDER_MULTIFORMAT SND_FEEDER_FULL_MULTIFORMAT SND_FEEDER_RATE_HP SND_PCM_64 SND_OLDSTEREO Manual page updates are on the way. Tested by: joel, Olivier SMEDTS <olivier at gid0 d org>, too many unsung / unnamed heroes.
2009-06-07 19:12:08 +00:00
KOBJMETHOD_END
};
CHANNEL_DECLARE(nmchan);
/* The interrupt handler */
static void
nm_intr(void *p)
{
struct sc_info *sc = (struct sc_info *)p;
int status, x;
status = nm_rd(sc, NM_INT_REG, sc->irsz);
if (status == 0)
return;
if (status & sc->playint) {
status &= ~sc->playint;
sc->pch.wmark += sc->pch.blksize;
sc->pch.wmark %= NM_BUFFSIZE;
nm_wr(sc, NM_PBUFFER_WMARK, sc->pbuf + sc->pch.wmark, 4);
nm_ackint(sc, sc->playint);
chn_intr(sc->pch.channel);
}
if (status & sc->recint) {
status &= ~sc->recint;
sc->rch.wmark += sc->rch.blksize;
sc->rch.wmark %= NM_BUFFSIZE;
nm_wr(sc, NM_RBUFFER_WMARK, sc->rbuf + sc->rch.wmark, 4);
nm_ackint(sc, sc->recint);
chn_intr(sc->rch.channel);
}
if (status & sc->misc1int) {
status &= ~sc->misc1int;
nm_ackint(sc, sc->misc1int);
x = nm_rd(sc, 0x400, 1);
nm_wr(sc, 0x400, x | 2, 1);
device_printf(sc->dev, "misc int 1\n");
}
if (status & sc->misc2int) {
status &= ~sc->misc2int;
nm_ackint(sc, sc->misc2int);
x = nm_rd(sc, 0x400, 1);
nm_wr(sc, 0x400, x & ~2, 1);
device_printf(sc->dev, "misc int 2\n");
}
if (status) {
nm_ackint(sc, status);
device_printf(sc->dev, "unknown int\n");
}
}
/* -------------------------------------------------------------------- */
/*
* Probe and attach the card
*/
static int
nm_init(struct sc_info *sc)
{
u_int32_t ofs, i;
if (sc->type == NM256AV_PCI_ID) {
sc->ac97_base = NM_MIXER_OFFSET;
sc->ac97_status = NM_MIXER_STATUS_OFFSET;
sc->ac97_busy = NM_MIXER_READY_MASK;
sc->buftop = 2560 * 1024;
sc->irsz = 2;
sc->playint = NM_PLAYBACK_INT;
sc->recint = NM_RECORD_INT;
sc->misc1int = NM_MISC_INT_1;
sc->misc2int = NM_MISC_INT_2;
} else if (sc->type == NM256ZX_PCI_ID) {
sc->ac97_base = NM_MIXER_OFFSET;
sc->ac97_status = NM2_MIXER_STATUS_OFFSET;
sc->ac97_busy = NM2_MIXER_READY_MASK;
sc->buftop = (nm_rd(sc, 0xa0b, 2)? 6144 : 4096) * 1024;
sc->irsz = 4;
sc->playint = NM2_PLAYBACK_INT;
sc->recint = NM2_RECORD_INT;
sc->misc1int = NM2_MISC_INT_1;
sc->misc2int = NM2_MISC_INT_2;
} else return -1;
sc->badintr = 0;
ofs = sc->buftop - 0x0400;
sc->buftop -= 0x1400;
if (bootverbose)
device_printf(sc->dev, "buftop is 0x%08x\n", sc->buftop);
if ((nm_rdbuf(sc, ofs, 4) & NM_SIG_MASK) == NM_SIGNATURE) {
i = nm_rdbuf(sc, ofs + 4, 4);
if (i != 0 && i != 0xffffffff) {
if (bootverbose)
device_printf(sc->dev, "buftop is changed to 0x%08x\n", i);
sc->buftop = i;
}
}
sc->cbuf = sc->buftop - NM_MAX_COEFFICIENT;
sc->rbuf = sc->cbuf - NM_BUFFSIZE;
sc->pbuf = sc->rbuf - NM_BUFFSIZE;
sc->acbuf = sc->pbuf - (NM_TOTAL_COEFF_COUNT * 4);
nm_wr(sc, 0, 0x11, 1);
nm_wr(sc, NM_RECORD_ENABLE_REG, 0, 1);
nm_wr(sc, 0x214, 0, 2);
return 0;
}
static int
nm_pci_probe(device_t dev)
{
struct sc_info *sc = NULL;
char *s = NULL;
u_int32_t subdev, i;
subdev = (pci_get_subdevice(dev) << 16) | pci_get_subvendor(dev);
switch (pci_get_devid(dev)) {
case NM256AV_PCI_ID:
i = 0;
while ((i < NUM_BADCARDS) && (badcards[i] != subdev))
i++;
/* Try to catch other non-ac97 cards */
if (i == NUM_BADCARDS) {
if (!(sc = malloc(sizeof(*sc), M_DEVBUF, M_NOWAIT | M_ZERO))) {
device_printf(dev, "cannot allocate softc\n");
return ENXIO;
}
sc->regid = PCIR_BAR(1);
sc->reg = bus_alloc_resource_any(dev, SYS_RES_MEMORY,
&sc->regid,
RF_ACTIVE);
if (!sc->reg) {
device_printf(dev, "unable to map register space\n");
free(sc, M_DEVBUF);
return ENXIO;
}
/*
* My Panasonic CF-M2EV needs resetting device
* before checking mixer is present or not.
* t.ichinoseki@nifty.com.
*/
nm_wr(sc, 0, 0x11, 1); /* reset device */
if ((nm_rd(sc, NM_MIXER_PRESENCE, 2) &
NM_PRESENCE_MASK) != NM_PRESENCE_VALUE) {
i = 0; /* non-ac97 card, but not listed */
DEB(device_printf(dev, "subdev = 0x%x - badcard?\n",
subdev));
}
bus_release_resource(dev, SYS_RES_MEMORY, sc->regid,
sc->reg);
free(sc, M_DEVBUF);
}
if (i == NUM_BADCARDS)
s = "NeoMagic 256AV";
2000-01-13 06:00:57 +00:00
DEB(else)
DEB(device_printf(dev, "this is a non-ac97 NM256AV, not attaching\n"));
break;
case NM256ZX_PCI_ID:
s = "NeoMagic 256ZX";
break;
}
if (s) device_set_desc(dev, s);
return s? 0 : ENXIO;
}
static int
nm_pci_attach(device_t dev)
{
struct sc_info *sc;
struct ac97_info *codec = NULL;
char status[SND_STATUSLEN];
sc = malloc(sizeof(*sc), M_DEVBUF, M_WAITOK | M_ZERO);
sc->dev = dev;
sc->type = pci_get_devid(dev);
pci_enable_busmaster(dev);
sc->bufid = PCIR_BAR(0);
sc->buf = bus_alloc_resource_any(dev, SYS_RES_MEMORY, &sc->bufid,
RF_ACTIVE);
sc->regid = PCIR_BAR(1);
sc->reg = bus_alloc_resource_any(dev, SYS_RES_MEMORY, &sc->regid,
RF_ACTIVE);
if (!sc->buf || !sc->reg) {
device_printf(dev, "unable to map register space\n");
goto bad;
}
if (nm_init(sc) == -1) {
device_printf(dev, "unable to initialize the card\n");
goto bad;
}
codec = AC97_CREATE(dev, sc, nm_ac97);
if (codec == NULL) goto bad;
if (mixer_init(dev, ac97_getmixerclass(), codec) == -1) goto bad;
sc->irqid = 0;
sc->irq = bus_alloc_resource_any(dev, SYS_RES_IRQ, &sc->irqid,
RF_ACTIVE | RF_SHAREABLE);
if (!sc->irq || snd_setup_intr(dev, sc->irq, 0, nm_intr, sc, &sc->ih)) {
device_printf(dev, "unable to map interrupt\n");
goto bad;
}
Use uintmax_t (typedef'd to rman_res_t type) for rman ranges. On some architectures, u_long isn't large enough for resource definitions. Particularly, powerpc and arm allow 36-bit (or larger) physical addresses, but type `long' is only 32-bit. This extends rman's resources to uintmax_t. With this change, any resource can feasibly be placed anywhere in physical memory (within the constraints of the driver). Why uintmax_t and not something machine dependent, or uint64_t? Though it's possible for uintmax_t to grow, it's highly unlikely it will become 128-bit on 32-bit architectures. 64-bit architectures should have plenty of RAM to absorb the increase on resource sizes if and when this occurs, and the number of resources on memory-constrained systems should be sufficiently small as to not pose a drastic overhead. That being said, uintmax_t was chosen for source clarity. If it's specified as uint64_t, all printf()-like calls would either need casts to uintmax_t, or be littered with PRI*64 macros. Casts to uintmax_t aren't horrible, but it would also bake into the API for resource_list_print_type() either a hidden assumption that entries get cast to uintmax_t for printing, or these calls would need the PRI*64 macros. Since source code is meant to be read more often than written, I chose the clearest path of simply using uintmax_t. Tested on a PowerPC p5020-based board, which places all device resources in 0xfxxxxxxxx, and has 8GB RAM. Regression tested on qemu-system-i386 Regression tested on qemu-system-mips (malta profile) Tested PAE and devinfo on virtualbox (live CD) Special thanks to bz for his testing on ARM. Reviewed By: bz, jhb (previous) Relnotes: Yes Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D4544
2016-03-18 01:28:41 +00:00
snprintf(status, SND_STATUSLEN, "at memory 0x%jx, 0x%jx irq %jd %s",
rman_get_start(sc->buf), rman_get_start(sc->reg),
rman_get_start(sc->irq),PCM_KLDSTRING(snd_neomagic));
if (pcm_register(dev, sc, 1, 1)) goto bad;
pcm_addchan(dev, PCMDIR_REC, &nmchan_class, sc);
pcm_addchan(dev, PCMDIR_PLAY, &nmchan_class, sc);
pcm_setstatus(dev, status);
return 0;
bad:
if (codec) ac97_destroy(codec);
if (sc->buf) bus_release_resource(dev, SYS_RES_MEMORY, sc->bufid, sc->buf);
if (sc->reg) bus_release_resource(dev, SYS_RES_MEMORY, sc->regid, sc->reg);
if (sc->ih) bus_teardown_intr(dev, sc->irq, sc->ih);
if (sc->irq) bus_release_resource(dev, SYS_RES_IRQ, sc->irqid, sc->irq);
free(sc, M_DEVBUF);
return ENXIO;
}
static int
nm_pci_detach(device_t dev)
{
int r;
struct sc_info *sc;
r = pcm_unregister(dev);
if (r)
return r;
sc = pcm_getdevinfo(dev);
bus_release_resource(dev, SYS_RES_MEMORY, sc->bufid, sc->buf);
bus_release_resource(dev, SYS_RES_MEMORY, sc->regid, sc->reg);
bus_teardown_intr(dev, sc->irq, sc->ih);
bus_release_resource(dev, SYS_RES_IRQ, sc->irqid, sc->irq);
free(sc, M_DEVBUF);
return 0;
}
static int
nm_pci_suspend(device_t dev)
{
struct sc_info *sc;
sc = pcm_getdevinfo(dev);
/* stop playing */
if (sc->pch.active) {
nm_wr(sc, NM_PLAYBACK_ENABLE_REG, 0, 1);
nm_wr(sc, NM_AUDIO_MUTE_REG, NM_AUDIO_MUTE_BOTH, 2);
}
/* stop recording */
if (sc->rch.active) {
nm_wr(sc, NM_RECORD_ENABLE_REG, 0, 1);
}
return 0;
}
static int
nm_pci_resume(device_t dev)
{
struct sc_info *sc;
sc = pcm_getdevinfo(dev);
/*
* Reinit audio device.
* Don't call nm_init(). It would change buftop if X ran or
* is running. This makes playing and recording buffer address
* shift but these buffers of channel layer are not changed.
* As a result of this inconsistency, periodic noise will be
* generated while playing.
*/
nm_wr(sc, 0, 0x11, 1);
nm_wr(sc, 0x214, 0, 2);
/* Reinit mixer */
if (mixer_reinit(dev) == -1) {
device_printf(dev, "unable to reinitialize the mixer\n");
return ENXIO;
}
/* restart playing */
if (sc->pch.active) {
nm_wr(sc, NM_PLAYBACK_ENABLE_REG, NM_PLAYBACK_FREERUN |
NM_PLAYBACK_ENABLE_FLAG, 1);
nm_wr(sc, NM_AUDIO_MUTE_REG, 0, 2);
}
/* restart recording */
if (sc->rch.active) {
nm_wr(sc, NM_RECORD_ENABLE_REG, NM_RECORD_FREERUN |
NM_RECORD_ENABLE_FLAG, 1);
}
return 0;
}
static device_method_t nm_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, nm_pci_probe),
DEVMETHOD(device_attach, nm_pci_attach),
2000-09-17 23:46:32 +00:00
DEVMETHOD(device_detach, nm_pci_detach),
DEVMETHOD(device_suspend, nm_pci_suspend),
DEVMETHOD(device_resume, nm_pci_resume),
{ 0, 0 }
};
static driver_t nm_driver = {
"pcm",
nm_methods,
PCM_SOFTC_SIZE,
};
DRIVER_MODULE(snd_neomagic, pci, nm_driver, pcm_devclass, 0, 0);
MODULE_DEPEND(snd_neomagic, sound, SOUND_MINVER, SOUND_PREFVER, SOUND_MAXVER);
MODULE_VERSION(snd_neomagic, 1);