freebsd-dev/sys/dev/sound/macio/aoa.c

391 lines
8.9 KiB
C
Raw Normal View History

/*-
* Copyright 2008 by Marco Trillo. 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 ``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 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.
*
* $FreeBSD$
*/
/*
* Apple Onboard Audio (AOA).
*/
#include <sys/cdefs.h>
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/bus.h>
#include <sys/malloc.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <machine/dbdma.h>
#include <machine/resource.h>
#include <machine/bus.h>
#include <sys/rman.h>
#include <dev/ofw/ofw_bus.h>
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/macio/aoa.h>
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
#include "mixer_if.h"
struct aoa_dma {
struct mtx mutex;
struct resource *reg; /* DBDMA registers */
dbdma_channel_t *channel; /* DBDMA channel */
bus_dma_tag_t tag; /* bus_dma tag */
struct pcm_channel *pcm; /* PCM channel */
struct snd_dbuf *buf; /* PCM buffer */
u_int slots; /* # of slots */
u_int slot; /* current slot */
u_int bufsz; /* buffer size */
u_int blksz; /* block size */
int running;
};
static void
aoa_dma_set_program(struct aoa_dma *dma)
{
u_int32_t addr;
int i;
addr = (u_int32_t) sndbuf_getbufaddr(dma->buf);
KASSERT(dma->bufsz == sndbuf_getsize(dma->buf), ("bad size"));
dma->slots = dma->bufsz / dma->blksz;
for (i = 0; i < dma->slots; ++i) {
dbdma_insert_command(dma->channel,
i, /* slot */
DBDMA_OUTPUT_MORE, /* command */
0, /* stream */
addr, /* data */
dma->blksz, /* count */
DBDMA_ALWAYS, /* interrupt */
DBDMA_COND_TRUE, /* branch */
DBDMA_NEVER, /* wait */
dma->slots + 1 /* branch_slot */
);
addr += dma->blksz;
}
/* Branch back to beginning. */
dbdma_insert_branch(dma->channel, dma->slots, 0);
/* STOP command to branch when S0 is asserted. */
dbdma_insert_stop(dma->channel, dma->slots + 1);
/* Set S0 as the condition to branch to STOP. */
dbdma_set_branch_selector(dma->channel, 1 << 0, 1 << 0);
dbdma_set_device_status(dma->channel, 1 << 0, 0);
dbdma_sync_commands(dma->channel, BUS_DMASYNC_PREWRITE);
}
#define AOA_BUFFER_SIZE 65536
static struct aoa_dma *
aoa_dma_create(struct aoa_softc *sc)
{
struct aoa_dma *dma;
bus_dma_tag_t tag;
int err;
device_t self;
self = sc->sc_dev;
err = bus_dma_tag_create(bus_get_dma_tag(self),
4, 0, BUS_SPACE_MAXADDR_32BIT, BUS_SPACE_MAXADDR, NULL, NULL,
AOA_BUFFER_SIZE, 1, AOA_BUFFER_SIZE, 0, NULL, NULL, &tag);
if (err != 0)
return (NULL);
dma = malloc(sizeof(*dma), M_DEVBUF, M_WAITOK | M_ZERO);
dma->tag = tag;
dma->bufsz = AOA_BUFFER_SIZE;
dma->blksz = PAGE_SIZE; /* initial blocksize */
mtx_init(&dma->mutex, "AOA", NULL, MTX_DEF);
sc->sc_intrp = dma;
return (dma);
}
static void
aoa_dma_delete(struct aoa_dma *dma)
{
bus_dma_tag_destroy(dma->tag);
mtx_destroy(&dma->mutex);
free(dma, M_DEVBUF);
}
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
aoa_chan_setblocksize(kobj_t obj, void *data, u_int32_t blocksz)
{
struct aoa_dma *dma = data;
int err, lz;
DPRINTF(("aoa_chan_setblocksize: blocksz = %u, dma->blksz = %u\n",
blocksz, dma->blksz));
KASSERT(!dma->running, ("dma is running"));
KASSERT(blocksz > 0, ("bad blocksz"));
/* Round blocksz down to a power of two... */
__asm volatile ("cntlzw %0,%1" : "=r"(lz) : "r"(blocksz));
blocksz = 1 << (31 - lz);
DPRINTF(("blocksz = %u\n", blocksz));
/* ...but no more than the buffer. */
if (blocksz > dma->bufsz)
blocksz = dma->bufsz;
err = sndbuf_resize(dma->buf, dma->bufsz / blocksz, blocksz);
if (err != 0) {
DPRINTF(("sndbuf_resize returned %d\n", err));
return (0);
}
if (blocksz == dma->blksz)
return (dma->blksz);
/* One slot per block plus branch to 0 plus STOP. */
err = dbdma_resize_channel(dma->channel, 2 + dma->bufsz / blocksz);
if (err != 0) {
DPRINTF(("dbdma_resize_channel returned %d\n", err));
return (0);
}
/* Set the new blocksize. */
dma->blksz = blocksz;
aoa_dma_set_program(dma);
return (dma->blksz);
}
static int
aoa_chan_setformat(kobj_t obj, void *data, u_int32_t format)
{
DPRINTF(("aoa_chan_setformat: format = %u\n", format));
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 (format != SND_FORMAT(AFMT_S16_BE, 2, 0))
return (EINVAL);
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
aoa_chan_setspeed(kobj_t obj, void *data, u_int32_t speed)
{
DPRINTF(("aoa_chan_setspeed: speed = %u\n", speed));
return (44100);
}
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
aoa_chan_getptr(kobj_t obj, void *data)
{
struct aoa_dma *dma = data;
if (!dma->running)
return (0);
return (dma->slot * dma->blksz);
}
static void *
aoa_chan_init(kobj_t obj, void *devinfo, struct snd_dbuf *b,
struct pcm_channel *c, int dir)
{
struct aoa_softc *sc = devinfo;
struct aoa_dma *dma;
int max_slots, err;
KASSERT(dir == PCMDIR_PLAY, ("bad dir"));
dma = aoa_dma_create(sc);
if (!dma)
return (NULL);
dma->pcm = c;
dma->buf = b;
dma->reg = sc->sc_odma;
/* One slot per block, plus branch to 0 plus STOP. */
max_slots = 2 + dma->bufsz / dma->blksz;
err = dbdma_allocate_channel(dma->reg, 0, bus_get_dma_tag(sc->sc_dev),
max_slots, &dma->channel );
if (err != 0) {
aoa_dma_delete(dma);
return (NULL);
}
if (sndbuf_alloc(dma->buf, dma->tag, 0, dma->bufsz) != 0) {
dbdma_free_channel(dma->channel);
aoa_dma_delete(dma);
return (NULL);
}
aoa_dma_set_program(dma);
return (dma);
}
static int
aoa_chan_trigger(kobj_t obj, void *data, int go)
{
struct aoa_dma *dma = data;
int i;
switch (go) {
case PCMTRIG_START:
/* Start the DMA. */
dma->running = 1;
dma->slot = 0;
dbdma_set_current_cmd(dma->channel, dma->slot);
dbdma_run(dma->channel);
return (0);
case PCMTRIG_STOP:
case PCMTRIG_ABORT:
mtx_lock(&dma->mutex);
dma->running = 0;
/* Make it branch to the STOP command. */
dbdma_set_device_status(dma->channel, 1 << 0, 1 << 0);
/* XXX should wait for DBDMA_ACTIVE to clear. */
DELAY(40000);
/* Reset the DMA. */
dbdma_stop(dma->channel);
dbdma_set_device_status(dma->channel, 1 << 0, 0);
for (i = 0; i < dma->slots; ++i)
dbdma_clear_cmd_status(dma->channel, i);
mtx_unlock(&dma->mutex);
return (0);
}
return (0);
}
static int
aoa_chan_free(kobj_t obj, void *data)
{
struct aoa_dma *dma = data;
sndbuf_free(dma->buf);
dbdma_free_channel(dma->channel);
aoa_dma_delete(dma);
return (0);
}
void
aoa_interrupt(void *xsc)
{
struct aoa_softc *sc = xsc;
struct aoa_dma *dma;
if (!(dma = sc->sc_intrp) || !dma->running)
return;
mtx_lock(&dma->mutex);
while (dbdma_get_cmd_status(dma->channel, dma->slot)) {
dbdma_clear_cmd_status(dma->channel, dma->slot);
dma->slot = (dma->slot + 1) % dma->slots;
mtx_unlock(&dma->mutex);
chn_intr(dma->pcm);
mtx_lock(&dma->mutex);
}
mtx_unlock(&dma->mutex);
}
static u_int32_t sc_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_S16_BE, 2, 0),
0
};
static struct pcmchan_caps aoa_caps = {44100, 44100, sc_fmt, 0};
static struct pcmchan_caps *
aoa_chan_getcaps(kobj_t obj, void *data)
{
return (&aoa_caps);
}
static kobj_method_t aoa_chan_methods[] = {
KOBJMETHOD(channel_init, aoa_chan_init),
KOBJMETHOD(channel_free, aoa_chan_free),
KOBJMETHOD(channel_setformat, aoa_chan_setformat),
KOBJMETHOD(channel_setspeed, aoa_chan_setspeed),
KOBJMETHOD(channel_setblocksize,aoa_chan_setblocksize),
KOBJMETHOD(channel_trigger, aoa_chan_trigger),
KOBJMETHOD(channel_getptr, aoa_chan_getptr),
KOBJMETHOD(channel_getcaps, aoa_chan_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(aoa_chan);
int
aoa_attach(void *xsc)
{
char status[SND_STATUSLEN];
struct aoa_softc *sc;
device_t self;
int err;
sc = xsc;
self = sc->sc_dev;
if (pcm_register(self, sc, 1, 0))
return (ENXIO);
err = pcm_getbuffersize(self, AOA_BUFFER_SIZE, AOA_BUFFER_SIZE,
AOA_BUFFER_SIZE);
DPRINTF(("pcm_getbuffersize returned %d\n", err));
pcm_addchan(self, PCMDIR_PLAY, &aoa_chan_class, sc);
snprintf(status, sizeof(status), "at %s", ofw_bus_get_name(self));
pcm_setstatus(self, status);
return (0);
}