freebsd-dev/sys/dev/sound/pcm/sound.h

812 lines
25 KiB
C
Raw Normal View History

/*-
2003-09-07 16:28:03 +00:00
* Copyright (c) 1999 Cameron Grant <cg@freebsd.org>
* Copyright by Hannu Savolainen 1995
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
1999-09-01 06:58:27 +00:00
* $FreeBSD$
*/
/*
* first, include kernel header files.
*/
#ifndef _OS_H_
#define _OS_H_
#ifdef _KERNEL
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/ioccom.h>
#include <sys/filio.h>
#include <sys/sockio.h>
#include <sys/fcntl.h>
#include <sys/selinfo.h>
#include <sys/proc.h>
#include <sys/kernel.h> /* for DATA_SET */
#include <sys/module.h>
#include <sys/conf.h>
#include <sys/file.h>
#include <sys/uio.h>
#include <sys/syslog.h>
#include <sys/errno.h>
#include <sys/malloc.h>
#include <sys/bus.h>
#if __FreeBSD_version < 500000
#include <sys/buf.h>
#endif
#include <machine/resource.h>
#include <machine/bus.h>
#include <sys/rman.h>
Welcome to Once-a-year Sound Mega-Commit. Enjoy numerous updates and fixes in every sense. General ------- - Multichannel safe, endian safe, format safe * Large part of critical pcm filters such as vchan.c, feeder_rate.c, feeder_volume.c, feeder_fmt.c and feeder.c has been rewritten so that using them does not cause the pcm data to be converted to 16bit little endian. * Macrosses for accessing pcm data safely are defined within sound.h in the form of PCM_READ_* / PCM_WRITE_* * Currently, most of them are probably limited for mono/stereo handling, but the future addition of true multichannel will be much easier. - Low latency operation * Well, this require lot more works to do not just within sound driver, but we're heading towards right direction. Buffer/block sizing within channel.c is rewritten to calculate precise allocation for various combination of sample/data/rate size. As a result, applying correct SNDCTL_DSP_POLICY value will achive expected latency behaviour simmilar to what commercial 4front driver do. * Signal handling fix. ctrl+c of "cat /dev/zero > /dev/dsp" does not result long delay. * Eliminate sound truncation if the sound data is too small. DIY: 1) Download / extract http://people.freebsd.org/~ariff/lowlatency/shortfiles.tar.gz 2) Do a comparison between "cat state*.au > /dev/dsp" and "for x in state*.au ; do cat $x > /dev/dsp ; done" - there should be no "perceivable" differences. Double close for PR kern/31445. CAVEAT: Low latency come with (unbearable) price especially for poorly written applications. Applications that trying to act smarter by requesting (wrong) blocksize/blockcount will suffer the most. Fixup samples/patches can be found at: http://people.freebsd.org/~ariff/ports/ - Switch minimum/maximum sampling rate limit to "1" and "2016000" (48k * 42) due to closer compatibility with 4front driver. Discussed with: marcus@ (long time ago?) - All driver specific sysctls in the form of "hw.snd.pcm%d.*" have been moved to their own dev sysctl nodes, notably: hw.snd.pcm%d.vchans -> dev.pcm.%d.vchans Bump __FreeBSD_version. Driver specific --------------- - Ditto for sysctls. - snd_atiixp, snd_es137x, snd_via8233, snd_hda * Numerous cleanups and fixes. * _EXPERIMENTAL_ polling mode support using simple callout_* mechanisme. This was intended for pure debugging and latency measurement, but proven good enough in few unexpected and rare cases (such as problematic shared IRQ with GIANT devices - USB). Polling can be enabled/disabled through dev.pcm.0.polling. Disabled by default. - snd_ich * Fix possible overflow during speed calibration. Delay final initialization (pcm_setstatus) after calibration finished. PR: kern/100169 Tested by: Kevin Overman <oberman@es.net> * Inverted EAPD for few Nec VersaPro. PR: kern/104715 Submitted by: KAWATA Masahiko <kawata@mta.biglobe.ne.jp> Thanks to various people, notably Joel Dahl, Yuriy Tsibizov, Kevin Oberman, those at #freebsd-azalia @ freenode and others for testing. Joel Dahl will do the manpage update.
2006-11-26 12:24:06 +00:00
#include <sys/limits.h>
#include <sys/mman.h>
#include <sys/poll.h>
#include <sys/sbuf.h>
#include <sys/soundcard.h>
#include <sys/sysctl.h>
#include <sys/kobj.h>
#include <vm/vm.h>
#include <vm/pmap.h>
#undef USING_MUTEX
#undef USING_DEVFS
2001-02-27 06:58:55 +00:00
#if __FreeBSD_version > 500000
#include <sys/lock.h>
#include <sys/mutex.h>
MFp4 the sound Google Summer of Code project: The goal was to sync with the OSSv4 API 4Front Technologies uses in their proprietary OSS driver. This was successful as far as possible. The part of the API which is stable is implemented, for the rest there are some stubs already. New system ioctls: - SNDCTL_SYSINFO - obtain audio system info (version, # of audio/midi/ mixer devices, etc.) - SNDCTL_AUDIOINFO - fetch details about a specific audio device - SNDCTL_MIXERINFO - fetch details about a specific mixer device New audio ioctls: - Sync groups (SNDCTL_DSP_SYNCGROUP/SNDCTL_DSP_SYNCSTART) which allow triggered playback/recording on multiple devices (even across processes simultaneously). - Peak meters (SNDCTL_DSP_GETIPEAKS/SNDCTL_DSP_GETOPEAKS) - can query audio drivers for peak levels (needs driver support, disabled for now). - Per channel playback/recording levels - SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL. Note that these are still in name only, just wrapping around the AC97-style mixer at the moment. The next step is to push them down to the drivers. Audio ioctls still under development by 4Front (for which stubs may exist in this commit): - SNDCTL_GETNAME, SNDCTL_{GET,SET}{SONG,LABEL} - SNDCTL_DSP_{GET,SET}_CHNORDER - SNDCTL_MIX_ENUMINFO, SNDCTL_MIX_EXTINFO - (might be documented enough in the OSS releases to work on this. These ioctls cover the cool "twiddle any knob on your card" features.) Missing: - SNDCTL_DSP_COOKEDMODE -- this ioctl is used to give applications direct access to a card's buffers, bypassing the feeder architecture. It's a toughy -- "someone" needs to decide : (a) if this is desireable, and (b) if it's reasonably feasible. Updates for driver writers: So far, only two routines to the channel class (in channel_if.m) are added. One is for fetching a list of discrete supported playback/recording rates of a channel, and the other is for fetching peak level info (useful for drawing peak meters). Interested parties may want to help pushing down SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL into the drivers. To use the new stuff you need to rebuild the sound drivers or your kernel (depending on if you use modules or not) and to install soundcard.h (a buildworld/installworld handles this). Sponsored by: Google SoC 2006 Submitted by: ryanb Many thanks to: 4Front Technologies for their cooperation, explanations and the nice license of their soundcard.h.
2006-09-23 20:45:47 +00:00
#include <sys/condvar.h>
2001-02-27 06:58:55 +00:00
#define USING_MUTEX
#define USING_DEVFS
#else
#define INTR_TYPE_AV INTR_TYPE_TTY
#define INTR_MPSAFE 0
#endif
#define SND_DYNSYSCTL
struct pcm_channel;
struct pcm_feeder;
struct snd_dbuf;
struct snd_mixer;
#include <dev/sound/pcm/buffer.h>
#include <dev/sound/pcm/channel.h>
#include <dev/sound/pcm/feeder.h>
#include <dev/sound/pcm/mixer.h>
MFp4 the sound Google Summer of Code project: The goal was to sync with the OSSv4 API 4Front Technologies uses in their proprietary OSS driver. This was successful as far as possible. The part of the API which is stable is implemented, for the rest there are some stubs already. New system ioctls: - SNDCTL_SYSINFO - obtain audio system info (version, # of audio/midi/ mixer devices, etc.) - SNDCTL_AUDIOINFO - fetch details about a specific audio device - SNDCTL_MIXERINFO - fetch details about a specific mixer device New audio ioctls: - Sync groups (SNDCTL_DSP_SYNCGROUP/SNDCTL_DSP_SYNCSTART) which allow triggered playback/recording on multiple devices (even across processes simultaneously). - Peak meters (SNDCTL_DSP_GETIPEAKS/SNDCTL_DSP_GETOPEAKS) - can query audio drivers for peak levels (needs driver support, disabled for now). - Per channel playback/recording levels - SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL. Note that these are still in name only, just wrapping around the AC97-style mixer at the moment. The next step is to push them down to the drivers. Audio ioctls still under development by 4Front (for which stubs may exist in this commit): - SNDCTL_GETNAME, SNDCTL_{GET,SET}{SONG,LABEL} - SNDCTL_DSP_{GET,SET}_CHNORDER - SNDCTL_MIX_ENUMINFO, SNDCTL_MIX_EXTINFO - (might be documented enough in the OSS releases to work on this. These ioctls cover the cool "twiddle any knob on your card" features.) Missing: - SNDCTL_DSP_COOKEDMODE -- this ioctl is used to give applications direct access to a card's buffers, bypassing the feeder architecture. It's a toughy -- "someone" needs to decide : (a) if this is desireable, and (b) if it's reasonably feasible. Updates for driver writers: So far, only two routines to the channel class (in channel_if.m) are added. One is for fetching a list of discrete supported playback/recording rates of a channel, and the other is for fetching peak level info (useful for drawing peak meters). Interested parties may want to help pushing down SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL into the drivers. To use the new stuff you need to rebuild the sound drivers or your kernel (depending on if you use modules or not) and to install soundcard.h (a buildworld/installworld handles this). Sponsored by: Google SoC 2006 Submitted by: ryanb Many thanks to: 4Front Technologies for their cooperation, explanations and the nice license of their soundcard.h.
2006-09-23 20:45:47 +00:00
#include <dev/sound/pcm/dsp.h>
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
#include <dev/sound/clone.h>
#include <dev/sound/unit.h>
#define PCM_SOFTC_SIZE 512
#define SND_STATUSLEN 64
2000-08-09 01:22:09 +00:00
#define SOUND_MODVER 2
#define SOUND_MINVER SOUND_MODVER
#define SOUND_PREFVER SOUND_MODVER
#define SOUND_MAXVER SOUND_MODVER
/*
* We're abusing the fact that MAXMINOR still have enough room
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
* for our bit twiddling and nobody ever need 512 unique soundcards,
* 32 unique device types and 1024 unique cloneable devices for the
* next 100 years...
*/
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
#define PCMMAXUNIT (snd_max_u())
#define PCMMAXDEV (snd_max_d())
#define PCMMAXCHAN (snd_max_c())
#define PCMMAXCLONE PCMMAXCHAN
#define PCMUNIT(x) (snd_unit2u(dev2unit(x)))
#define PCMDEV(x) (snd_unit2d(dev2unit(x)))
#define PCMCHAN(x) (snd_unit2c(dev2unit(x)))
/*
* By design, limit possible channels for each direction.
*/
#define SND_MAXHWCHAN 256
#define SND_MAXVCHANS SND_MAXHWCHAN
#define SD_F_SIMPLEX 0x00000001
#define SD_F_AUTOVCHAN 0x00000002
#define SD_F_SOFTPCMVOL 0x00000004
#define SD_F_PSWAPLR 0x00000008
#define SD_F_RSWAPLR 0x00000010
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
#define SD_F_DYING 0x00000020
#define SD_F_SUICIDE 0x00000040
Last (again ?!?) major commit for RELENG_7, featuring total Giant eradication in/from userland path, countless locking fixes, etc. - General sleep call through msleep(9) has been converted to condvar(9) with better consistencies. - Heavily guard every possible "slow path" entries (open(), close(), few ioctl()s, sysctls), but once it entering "fast path" (io, interrupt started), they are free to fly on their own. - Rearrange locking sequences, resulting better concurrency and serialization. Large part doesn't even need locking at all, and will be removed in future. Less clutter, except in few places due to lock ordering. - Anonymous mixer object creation/deletion to simplify mixer handling beyond typical mixer ioctls. Submitted by: chibis (with modifications) - Add few mix_[get|set|..] functions to avoid calling mixer_ioctl() directly using cryptic arguments. - Locking fixes to avoid possible deadlock with (still under Giant) USB. - Better simplex/duplex device handling. - Recover mmap() functionality for recording, which has been lost since 2.2.x - 3.x (the introduction of newpcm). Full-duplex mmap still doesn't work (due to VM/page design), but people still can mmap both by opening each direction separately. mmaped playback is guarantee to work either way. - New sysctl: "hw.snd.compat_linux_mmap" to allow PROT_EXEC page mapping, due to recent changes in linux compatibility layer which require it. All linux applications that using sound + mmap() (mostly games) require this to be enabled. Disabled by default. - Other goodies.. too many, that will increase releng7 shareholder value and make users of releng6 (and below) cry ;) * This commit should be atomic. If anything goes wrong (not counting problem originated from elsewhere), I will not hesitate to revert everything back within 12 hours. This substantial changes itself not a rocket science and the process has begun for almost 2 years, and lots of incremental changes are already in place during that period of time. * Some issues does occur in snd_emu10kx (note the 'x') due to various internal locking issues and it is currently being worked on by chibis. Tested by: chibis (Yuriy Tsibizov), joel, Alexandre Vieira, many innocent souls...
2007-06-16 03:37:28 +00:00
#define SD_F_BUSY 0x00000080
#define SD_F_MPSAFE 0x00000100
#define SD_F_REGISTERED 0x00000200
#define SD_F_PRIO_RD 0x10000000
#define SD_F_PRIO_WR 0x20000000
#define SD_F_PRIO_SET (SD_F_PRIO_RD | SD_F_PRIO_WR)
#define SD_F_DIR_SET 0x40000000
#define SD_F_TRANSIENT 0xf0000000
Last (again ?!?) major commit for RELENG_7, featuring total Giant eradication in/from userland path, countless locking fixes, etc. - General sleep call through msleep(9) has been converted to condvar(9) with better consistencies. - Heavily guard every possible "slow path" entries (open(), close(), few ioctl()s, sysctls), but once it entering "fast path" (io, interrupt started), they are free to fly on their own. - Rearrange locking sequences, resulting better concurrency and serialization. Large part doesn't even need locking at all, and will be removed in future. Less clutter, except in few places due to lock ordering. - Anonymous mixer object creation/deletion to simplify mixer handling beyond typical mixer ioctls. Submitted by: chibis (with modifications) - Add few mix_[get|set|..] functions to avoid calling mixer_ioctl() directly using cryptic arguments. - Locking fixes to avoid possible deadlock with (still under Giant) USB. - Better simplex/duplex device handling. - Recover mmap() functionality for recording, which has been lost since 2.2.x - 3.x (the introduction of newpcm). Full-duplex mmap still doesn't work (due to VM/page design), but people still can mmap both by opening each direction separately. mmaped playback is guarantee to work either way. - New sysctl: "hw.snd.compat_linux_mmap" to allow PROT_EXEC page mapping, due to recent changes in linux compatibility layer which require it. All linux applications that using sound + mmap() (mostly games) require this to be enabled. Disabled by default. - Other goodies.. too many, that will increase releng7 shareholder value and make users of releng6 (and below) cry ;) * This commit should be atomic. If anything goes wrong (not counting problem originated from elsewhere), I will not hesitate to revert everything back within 12 hours. This substantial changes itself not a rocket science and the process has begun for almost 2 years, and lots of incremental changes are already in place during that period of time. * Some issues does occur in snd_emu10kx (note the 'x') due to various internal locking issues and it is currently being worked on by chibis. Tested by: chibis (Yuriy Tsibizov), joel, Alexandre Vieira, many innocent souls...
2007-06-16 03:37:28 +00:00
#define PCM_ALIVE(x) ((x) != NULL && (x)->lock != NULL && \
!((x)->flags & SD_F_DYING))
#define PCM_REGISTERED(x) (PCM_ALIVE(x) && \
((x)->flags & SD_F_REGISTERED))
/* many variables should be reduced to a range. Here define a macro */
#define RANGE(var, low, high) (var) = \
(((var)<(low))? (low) : ((var)>(high))? (high) : (var))
#define DSP_BUFFSIZE (8192)
Welcome to Once-a-year Sound Mega-Commit. Enjoy numerous updates and fixes in every sense. General ------- - Multichannel safe, endian safe, format safe * Large part of critical pcm filters such as vchan.c, feeder_rate.c, feeder_volume.c, feeder_fmt.c and feeder.c has been rewritten so that using them does not cause the pcm data to be converted to 16bit little endian. * Macrosses for accessing pcm data safely are defined within sound.h in the form of PCM_READ_* / PCM_WRITE_* * Currently, most of them are probably limited for mono/stereo handling, but the future addition of true multichannel will be much easier. - Low latency operation * Well, this require lot more works to do not just within sound driver, but we're heading towards right direction. Buffer/block sizing within channel.c is rewritten to calculate precise allocation for various combination of sample/data/rate size. As a result, applying correct SNDCTL_DSP_POLICY value will achive expected latency behaviour simmilar to what commercial 4front driver do. * Signal handling fix. ctrl+c of "cat /dev/zero > /dev/dsp" does not result long delay. * Eliminate sound truncation if the sound data is too small. DIY: 1) Download / extract http://people.freebsd.org/~ariff/lowlatency/shortfiles.tar.gz 2) Do a comparison between "cat state*.au > /dev/dsp" and "for x in state*.au ; do cat $x > /dev/dsp ; done" - there should be no "perceivable" differences. Double close for PR kern/31445. CAVEAT: Low latency come with (unbearable) price especially for poorly written applications. Applications that trying to act smarter by requesting (wrong) blocksize/blockcount will suffer the most. Fixup samples/patches can be found at: http://people.freebsd.org/~ariff/ports/ - Switch minimum/maximum sampling rate limit to "1" and "2016000" (48k * 42) due to closer compatibility with 4front driver. Discussed with: marcus@ (long time ago?) - All driver specific sysctls in the form of "hw.snd.pcm%d.*" have been moved to their own dev sysctl nodes, notably: hw.snd.pcm%d.vchans -> dev.pcm.%d.vchans Bump __FreeBSD_version. Driver specific --------------- - Ditto for sysctls. - snd_atiixp, snd_es137x, snd_via8233, snd_hda * Numerous cleanups and fixes. * _EXPERIMENTAL_ polling mode support using simple callout_* mechanisme. This was intended for pure debugging and latency measurement, but proven good enough in few unexpected and rare cases (such as problematic shared IRQ with GIANT devices - USB). Polling can be enabled/disabled through dev.pcm.0.polling. Disabled by default. - snd_ich * Fix possible overflow during speed calibration. Delay final initialization (pcm_setstatus) after calibration finished. PR: kern/100169 Tested by: Kevin Overman <oberman@es.net> * Inverted EAPD for few Nec VersaPro. PR: kern/104715 Submitted by: KAWATA Masahiko <kawata@mta.biglobe.ne.jp> Thanks to various people, notably Joel Dahl, Yuriy Tsibizov, Kevin Oberman, those at #freebsd-azalia @ freenode and others for testing. Joel Dahl will do the manpage update.
2006-11-26 12:24:06 +00:00
/*
* Macros for reading/writing PCM sample / int values from bytes array.
* Since every process is done using signed integer (and to make our life
* less miserable), unsigned sample will be converted to its signed
* counterpart and restored during writing back. To avoid overflow,
* we truncate 32bit (and only 32bit) samples down to 24bit (see below
* for the reason), unless PCM_USE_64BIT_ARITH is defined.
*/
/*
* Automatically turn on 64bit arithmetic on suitable archs
* (amd64 64bit, ia64, etc..) for wider 32bit samples / integer processing.
*/
#if LONG_BIT >= 64
#undef PCM_USE_64BIT_ARITH
#define PCM_USE_64BIT_ARITH 1
#else
#if 0
#undef PCM_USE_64BIT_ARITH
#define PCM_USE_64BIT_ARITH 1
#endif
#endif
#ifdef PCM_USE_64BIT_ARITH
typedef int64_t intpcm_t;
#else
typedef int32_t intpcm_t;
#endif
/* 32bit fixed point shift */
#define PCM_FXSHIFT 8
#define PCM_S8_MAX 0x7f
#define PCM_S8_MIN -0x80
#define PCM_S16_MAX 0x7fff
#define PCM_S16_MIN -0x8000
#define PCM_S24_MAX 0x7fffff
#define PCM_S24_MIN -0x800000
#ifdef PCM_USE_64BIT_ARITH
#if LONG_BIT >= 64
#define PCM_S32_MAX 0x7fffffffL
#define PCM_S32_MIN -0x80000000L
#else
#define PCM_S32_MAX 0x7fffffffLL
#define PCM_S32_MIN -0x80000000LL
#endif
#else
#define PCM_S32_MAX 0x7fffffff
#define PCM_S32_MIN (-0x7fffffff - 1)
#endif
/* Bytes-per-sample definition */
#define PCM_8_BPS 1
#define PCM_16_BPS 2
#define PCM_24_BPS 3
#define PCM_32_BPS 4
#if BYTE_ORDER == LITTLE_ENDIAN
#define PCM_READ_S16_LE(b8) *((int16_t *)(b8))
#define _PCM_READ_S32_LE(b8) *((int32_t *)(b8))
#define PCM_READ_S16_BE(b8) \
((int32_t)((b8)[1] | ((int8_t)((b8)[0])) << 8))
#define _PCM_READ_S32_BE(b8) \
((int32_t)((b8)[3] | (b8)[2] << 8 | (b8)[1] << 16 | \
((int8_t)((b8)[0])) << 24))
#define PCM_WRITE_S16_LE(b8, val) *((int16_t *)(b8)) = (val)
#define _PCM_WRITE_S32_LE(b8, val) *((int32_t *)(b8)) = (val)
#define PCM_WRITE_S16_BE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[1] = val; \
b8[0] = val >> 8; \
} while(0)
#define _PCM_WRITE_S32_BE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[3] = val; \
b8[2] = val >> 8; \
b8[1] = val >> 16; \
b8[0] = val >> 24; \
} while(0)
#define PCM_READ_U16_LE(b8) ((int16_t)(*((uint16_t *)(b8)) ^ 0x8000))
#define _PCM_READ_U32_LE(b8) ((int32_t)(*((uint32_t *)(b8)) ^ 0x80000000))
#define PCM_READ_U16_BE(b8) \
((int32_t)((b8)[1] | ((int8_t)((b8)[0] ^ 0x80)) << 8))
#define _PCM_READ_U32_BE(b8) \
((int32_t)((b8)[3] | (b8)[2] << 8 | (b8)[1] << 16 | \
((int8_t)((b8)[0] ^ 0x80)) << 24))
#define PCM_WRITE_U16_LE(b8, val) *((uint16_t *)(b8)) = (val) ^ 0x8000
#define _PCM_WRITE_U32_LE(b8, val) *((uint32_t *)(b8)) = (val) ^ 0x80000000
#define PCM_WRITE_U16_BE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[1] = val; \
b8[0] = (val >> 8) ^ 0x80; \
} while(0)
#define _PCM_WRITE_U32_BE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[3] = val; \
b8[2] = val >> 8; \
b8[1] = val >> 16; \
b8[0] = (val >> 24) ^ 0x80; \
} while(0)
#else /* !LITTLE_ENDIAN */
#define PCM_READ_S16_LE(b8) \
((int32_t)((b8)[0] | ((int8_t)((b8)[1])) << 8))
#define _PCM_READ_S32_LE(b8) \
((int32_t)((b8)[0] | (b8)[1] << 8 | (b8)[2] << 16 | \
((int8_t)((b8)[3])) << 24))
#define PCM_READ_S16_BE(b8) *((int16_t *)(b8))
#define _PCM_READ_S32_BE(b8) *((int32_t *)(b8))
#define PCM_WRITE_S16_LE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[0] = val; \
b8[1] = val >> 8; \
} while(0)
#define _PCM_WRITE_S32_LE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[0] = val; \
b8[1] = val >> 8; \
b8[2] = val >> 16; \
b8[3] = val >> 24; \
} while(0)
#define PCM_WRITE_S16_BE(b8, val) *((int16_t *)(b8)) = (val)
#define _PCM_WRITE_S32_BE(b8, val) *((int32_t *)(b8)) = (val)
#define PCM_READ_U16_LE(b8) \
((int32_t)((b8)[0] | ((int8_t)((b8)[1] ^ 0x80)) << 8))
#define _PCM_READ_U32_LE(b8) \
((int32_t)((b8)[0] | (b8)[1] << 8 | (b8)[2] << 16 | \
((int8_t)((b8)[3] ^ 0x80)) << 24))
#define PCM_READ_U16_BE(b8) ((int16_t)(*((uint16_t *)(b8)) ^ 0x8000))
#define _PCM_READ_U32_BE(b8) ((int32_t)(*((uint32_t *)(b8)) ^ 0x80000000))
#define PCM_WRITE_U16_LE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[0] = val; \
b8[1] = (val >> 8) ^ 0x80; \
} while(0)
#define _PCM_WRITE_U32_LE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[0] = val; \
b8[1] = val >> 8; \
b8[2] = val >> 16; \
b8[3] = (val >> 24) ^ 0x80; \
} while(0)
#define PCM_WRITE_U16_BE(b8, val) *((uint16_t *)(b8)) = (val) ^ 0x8000
#define _PCM_WRITE_U32_BE(b8, val) *((uint32_t *)(b8)) = (val) ^ 0x80000000
#endif
#define PCM_READ_S24_LE(b8) \
((int32_t)((b8)[0] | (b8)[1] << 8 | ((int8_t)((b8)[2])) << 16))
#define PCM_READ_S24_BE(b8) \
((int32_t)((b8)[2] | (b8)[1] << 8 | ((int8_t)((b8)[0])) << 16))
#define PCM_WRITE_S24_LE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[0] = val; \
b8[1] = val >> 8; \
b8[2] = val >> 16; \
} while(0)
#define PCM_WRITE_S24_BE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[2] = val; \
b8[1] = val >> 8; \
b8[0] = val >> 16; \
} while(0)
#define PCM_READ_U24_LE(b8) \
((int32_t)((b8)[0] | (b8)[1] << 8 | \
((int8_t)((b8)[2] ^ 0x80)) << 16))
#define PCM_READ_U24_BE(b8) \
((int32_t)((b8)[2] | (b8)[1] << 8 | \
((int8_t)((b8)[0] ^ 0x80)) << 16))
#define PCM_WRITE_U24_LE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[0] = val; \
b8[1] = val >> 8; \
b8[2] = (val >> 16) ^ 0x80; \
} while(0)
#define PCM_WRITE_U24_BE(bb8, vval) do { \
int32_t val = (vval); \
uint8_t *b8 = (bb8); \
b8[2] = val; \
b8[1] = val >> 8; \
b8[0] = (val >> 16) ^ 0x80; \
} while(0)
#ifdef PCM_USE_64BIT_ARITH
#define PCM_READ_S32_LE(b8) _PCM_READ_S32_LE(b8)
#define PCM_READ_S32_BE(b8) _PCM_READ_S32_BE(b8)
#define PCM_WRITE_S32_LE(b8, val) _PCM_WRITE_S32_LE(b8, val)
#define PCM_WRITE_S32_BE(b8, val) _PCM_WRITE_S32_BE(b8, val)
#define PCM_READ_U32_LE(b8) _PCM_READ_U32_LE(b8)
#define PCM_READ_U32_BE(b8) _PCM_READ_U32_BE(b8)
#define PCM_WRITE_U32_LE(b8, val) _PCM_WRITE_U32_LE(b8, val)
#define PCM_WRITE_U32_BE(b8, val) _PCM_WRITE_U32_BE(b8, val)
#else /* !PCM_USE_64BIT_ARITH */
/*
* 24bit integer ?!? This is quite unfortunate, eh? Get the fact straight:
* Dynamic range for:
* 1) Human =~ 140db
* 2) 16bit = 96db (close enough)
* 3) 24bit = 144db (perfect)
* 4) 32bit = 196db (way too much)
* 5) Bugs Bunny = Gazillion!@%$Erbzzztt-EINVAL db
* Since we're not Bugs Bunny ..uh..err.. avoiding 64bit arithmetic, 24bit
* is pretty much sufficient for our signed integer processing.
*/
#define PCM_READ_S32_LE(b8) (_PCM_READ_S32_LE(b8) >> PCM_FXSHIFT)
#define PCM_READ_S32_BE(b8) (_PCM_READ_S32_BE(b8) >> PCM_FXSHIFT)
#define PCM_WRITE_S32_LE(b8, val) _PCM_WRITE_S32_LE(b8, (val) << PCM_FXSHIFT)
#define PCM_WRITE_S32_BE(b8, val) _PCM_WRITE_S32_BE(b8, (val) << PCM_FXSHIFT)
#define PCM_READ_U32_LE(b8) (_PCM_READ_U32_LE(b8) >> PCM_FXSHIFT)
#define PCM_READ_U32_BE(b8) (_PCM_READ_U32_BE(b8) >> PCM_FXSHIFT)
#define PCM_WRITE_U32_LE(b8, val) _PCM_WRITE_U32_LE(b8, (val) << PCM_FXSHIFT)
#define PCM_WRITE_U32_BE(b8, val) _PCM_WRITE_U32_BE(b8, (val) << PCM_FXSHIFT)
#endif
/*
* 8bit sample is pretty much useless since it doesn't provide
* sufficient dynamic range throughout our filtering process.
* For the sake of completeness, declare it anyway.
*/
#define PCM_READ_S8(b8) *((int8_t *)(b8))
#define PCM_READ_S8_NE(b8) PCM_READ_S8(b8)
#define PCM_READ_U8(b8) ((int8_t)(*((uint8_t *)(b8)) ^ 0x80))
#define PCM_READ_U8_NE(b8) PCM_READ_U8(b8)
#define PCM_WRITE_S8(b8, val) *((int8_t *)(b8)) = (val)
#define PCM_WRITE_S8_NE(b8, val) PCM_WRITE_S8(b8, val)
#define PCM_WRITE_U8(b8, val) *((uint8_t *)(b8)) = (val) ^ 0x80
#define PCM_WRITE_U8_NE(b8, val) PCM_WRITE_U8(b8, val)
#define PCM_CLAMP_S8(val) \
(((val) > PCM_S8_MAX) ? PCM_S8_MAX : \
(((val) < PCM_S8_MIN) ? PCM_S8_MIN : (val)))
#define PCM_CLAMP_S16(val) \
(((val) > PCM_S16_MAX) ? PCM_S16_MAX : \
(((val) < PCM_S16_MIN) ? PCM_S16_MIN : (val)))
#define PCM_CLAMP_S24(val) \
(((val) > PCM_S24_MAX) ? PCM_S24_MAX : \
(((val) < PCM_S24_MIN) ? PCM_S24_MIN : (val)))
#ifdef PCM_USE_64BIT_ARITH
#define PCM_CLAMP_S32(val) \
(((val) > PCM_S32_MAX) ? PCM_S32_MAX : \
(((val) < PCM_S32_MIN) ? PCM_S32_MIN : (val)))
#else
#define PCM_CLAMP_S32(val) \
(((val) > PCM_S24_MAX) ? PCM_S32_MAX : \
(((val) < PCM_S24_MIN) ? PCM_S32_MIN : \
((val) << PCM_FXSHIFT)))
#endif
#define PCM_CLAMP_U8(val) PCM_CLAMP_S8(val)
#define PCM_CLAMP_U16(val) PCM_CLAMP_S16(val)
#define PCM_CLAMP_U24(val) PCM_CLAMP_S24(val)
#define PCM_CLAMP_U32(val) PCM_CLAMP_S32(val)
/* make figuring out what a format is easier. got AFMT_STEREO already */
#define AFMT_32BIT (AFMT_S32_LE | AFMT_S32_BE | AFMT_U32_LE | AFMT_U32_BE)
Whats New: 1. Support wide range sampling rate, as low as 1hz up to int32 max (which is, insane) through new feeder_rate, multiple precisions choice (32/64 bit converter). This is indeed, quite insane, but it does give us more room and flexibility. Plenty sysctl options to adjust resampling characteristics. 2. Support 24/32 bit pcm format conversion through new, much improved, simplified and optimized feeder_fmt. Changes: 1. buffer.c / dsp.c / sound.h * Support for 24/32 AFMT. 2. feeder_rate.c * New implementation of sampling rate conversion with 32/64 bit precision, 1 - int32max hz (which is, ridiculous, yet very addictive). Much improved / smarter buffer management to not cause any missing samples at the end of conversion process * Tunable sysctls for various aspect: hw.snd.feeder_rate_ratemin - minimum allowable sampling rate (default to 4000) hw.snd.feeder_rate_ratemax - maximum allowable sampling rate (default to 1102500) hw.snd.feeder_rate_buffersize - conversion buffer size (default to 8192) hw.snd.feeder_rate_scaling - scaling / conversion method (please refer to the source for explaination). Default to previous implementation type. 3. feeder_fmt.c / sound.h * New implementation, support for 24/32bit conversion, optimized, and simplified. Few routines has been removed (8 to xlaw, 16 to 8). It just doesn't make sense. 4. channel.c * Support for 24/32 AFMT * Fix wrong xruns increment, causing incorrect underruns statistic while using vchans. 5. vchan.c * Support for 24/32 AFMT * Proper speed / rate detection especially for fixed rate ac97. User can override it using kernel hint: hint.pcm.<unit>.vchanrate="xxxx". Notes / Issues: * Virtual Channels (vchans) Enabling vchans can really, really help to solve overrun issues. This is quite understandable, because it operates entirely within its own buffering system without relying on hardware interrupt / state. Even if you don't need vchan, just enable single channel can help much. Few soundcards (notably via8233x, sblive, possibly others) have their own hardware multi channel, and this is unfortunately beyond vchan reachability. * The arrival of 24/32 also come with a price. Applications that can do 24/32bit playback need to be recompiled (notably mplayer). Use (recompiled) mplayer to experiment / test / debug this various format using -af format=fmt. Note that 24bit seeking in mplayer is a little bit broken, sometimes can cause silence or loud static noise. Pausing / seeking few times can solve this problem. You don't have to rebuild world entirely for this. Simply copy /usr/src/sys/sys/soundcard.h to /usr/include/sys/soundcard.h would suffice. Few drivers also need recompilation, and this can be done via /usr/src/sys/modules/sound/. Support for 24bit hardware playback is beyond the scope of this changes. That would require spessific hardware driver changes. * Don't expect playing 9999999999hz is a wise decision. Be reasonable. The new feeder_rate implemention provide flexibility, not insanity. You can easily chew up your CPU with this kind of mind instability. Please use proper mosquito repellent device for this obvious cracked brain attempt. As for testing purposes, you can use (again) mplayer to generate / play with different sampling rate. Use something like "mplayer -af resample=192000:0:0 <files>". Submitted by: Ariff Abdullah <skywizard@MyBSD.org.my> Tested by: multimedia@
2005-07-31 16:16:22 +00:00
#define AFMT_24BIT (AFMT_S24_LE | AFMT_S24_BE | AFMT_U24_LE | AFMT_U24_BE)
#define AFMT_16BIT (AFMT_S16_LE | AFMT_S16_BE | AFMT_U16_LE | AFMT_U16_BE)
#define AFMT_8BIT (AFMT_MU_LAW | AFMT_A_LAW | AFMT_U8 | AFMT_S8)
Whats New: 1. Support wide range sampling rate, as low as 1hz up to int32 max (which is, insane) through new feeder_rate, multiple precisions choice (32/64 bit converter). This is indeed, quite insane, but it does give us more room and flexibility. Plenty sysctl options to adjust resampling characteristics. 2. Support 24/32 bit pcm format conversion through new, much improved, simplified and optimized feeder_fmt. Changes: 1. buffer.c / dsp.c / sound.h * Support for 24/32 AFMT. 2. feeder_rate.c * New implementation of sampling rate conversion with 32/64 bit precision, 1 - int32max hz (which is, ridiculous, yet very addictive). Much improved / smarter buffer management to not cause any missing samples at the end of conversion process * Tunable sysctls for various aspect: hw.snd.feeder_rate_ratemin - minimum allowable sampling rate (default to 4000) hw.snd.feeder_rate_ratemax - maximum allowable sampling rate (default to 1102500) hw.snd.feeder_rate_buffersize - conversion buffer size (default to 8192) hw.snd.feeder_rate_scaling - scaling / conversion method (please refer to the source for explaination). Default to previous implementation type. 3. feeder_fmt.c / sound.h * New implementation, support for 24/32bit conversion, optimized, and simplified. Few routines has been removed (8 to xlaw, 16 to 8). It just doesn't make sense. 4. channel.c * Support for 24/32 AFMT * Fix wrong xruns increment, causing incorrect underruns statistic while using vchans. 5. vchan.c * Support for 24/32 AFMT * Proper speed / rate detection especially for fixed rate ac97. User can override it using kernel hint: hint.pcm.<unit>.vchanrate="xxxx". Notes / Issues: * Virtual Channels (vchans) Enabling vchans can really, really help to solve overrun issues. This is quite understandable, because it operates entirely within its own buffering system without relying on hardware interrupt / state. Even if you don't need vchan, just enable single channel can help much. Few soundcards (notably via8233x, sblive, possibly others) have their own hardware multi channel, and this is unfortunately beyond vchan reachability. * The arrival of 24/32 also come with a price. Applications that can do 24/32bit playback need to be recompiled (notably mplayer). Use (recompiled) mplayer to experiment / test / debug this various format using -af format=fmt. Note that 24bit seeking in mplayer is a little bit broken, sometimes can cause silence or loud static noise. Pausing / seeking few times can solve this problem. You don't have to rebuild world entirely for this. Simply copy /usr/src/sys/sys/soundcard.h to /usr/include/sys/soundcard.h would suffice. Few drivers also need recompilation, and this can be done via /usr/src/sys/modules/sound/. Support for 24bit hardware playback is beyond the scope of this changes. That would require spessific hardware driver changes. * Don't expect playing 9999999999hz is a wise decision. Be reasonable. The new feeder_rate implemention provide flexibility, not insanity. You can easily chew up your CPU with this kind of mind instability. Please use proper mosquito repellent device for this obvious cracked brain attempt. As for testing purposes, you can use (again) mplayer to generate / play with different sampling rate. Use something like "mplayer -af resample=192000:0:0 <files>". Submitted by: Ariff Abdullah <skywizard@MyBSD.org.my> Tested by: multimedia@
2005-07-31 16:16:22 +00:00
#define AFMT_SIGNED (AFMT_S32_LE | AFMT_S32_BE | AFMT_S24_LE | AFMT_S24_BE | \
AFMT_S16_LE | AFMT_S16_BE | AFMT_S8)
#define AFMT_BIGENDIAN (AFMT_S32_BE | AFMT_U32_BE | AFMT_S24_BE | AFMT_U24_BE | \
AFMT_S16_BE | AFMT_U16_BE)
struct pcm_channel *fkchan_setup(device_t dev);
int fkchan_kill(struct pcm_channel *c);
/*
* Minor numbers for the sound driver.
*
* Unfortunately Creative called the codec chip of SB as a DSP. For this
* reason the /dev/dsp is reserved for digitized audio use. There is a
* device for true DSP processors but it will be called something else.
* In v3.0 it's /dev/sndproc but this could be a temporary solution.
*/
#define SND_DEV_CTL 0 /* Control port /dev/mixer */
#define SND_DEV_SEQ 1 /* Sequencer /dev/sequencer */
#define SND_DEV_MIDIN 2 /* Raw midi access */
#define SND_DEV_DSP 3 /* Digitized voice /dev/dsp */
#define SND_DEV_AUDIO 4 /* Sparc compatible /dev/audio */
#define SND_DEV_DSP16 5 /* Like /dev/dsp but 16 bits/sample */
#define SND_DEV_STATUS 6 /* /dev/sndstat */
/* #7 not in use now. */
#define SND_DEV_SEQ2 8 /* /dev/sequencer, level 2 interface */
#define SND_DEV_SNDPROC 9 /* /dev/sndproc for programmable devices */
#define SND_DEV_PSS SND_DEV_SNDPROC /* ? */
#define SND_DEV_NORESET 10
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
#define SND_DEV_DSPHW_PLAY 11 /* specific playback channel */
#define SND_DEV_DSPHW_VPLAY 12 /* specific virtual playback channel */
#define SND_DEV_DSPHW_REC 13 /* specific record channel */
#define SND_DEV_DSPHW_VREC 14 /* specific virtual record channel */
Last (again ?!?) major commit for RELENG_7, featuring total Giant eradication in/from userland path, countless locking fixes, etc. - General sleep call through msleep(9) has been converted to condvar(9) with better consistencies. - Heavily guard every possible "slow path" entries (open(), close(), few ioctl()s, sysctls), but once it entering "fast path" (io, interrupt started), they are free to fly on their own. - Rearrange locking sequences, resulting better concurrency and serialization. Large part doesn't even need locking at all, and will be removed in future. Less clutter, except in few places due to lock ordering. - Anonymous mixer object creation/deletion to simplify mixer handling beyond typical mixer ioctls. Submitted by: chibis (with modifications) - Add few mix_[get|set|..] functions to avoid calling mixer_ioctl() directly using cryptic arguments. - Locking fixes to avoid possible deadlock with (still under Giant) USB. - Better simplex/duplex device handling. - Recover mmap() functionality for recording, which has been lost since 2.2.x - 3.x (the introduction of newpcm). Full-duplex mmap still doesn't work (due to VM/page design), but people still can mmap both by opening each direction separately. mmaped playback is guarantee to work either way. - New sysctl: "hw.snd.compat_linux_mmap" to allow PROT_EXEC page mapping, due to recent changes in linux compatibility layer which require it. All linux applications that using sound + mmap() (mostly games) require this to be enabled. Disabled by default. - Other goodies.. too many, that will increase releng7 shareholder value and make users of releng6 (and below) cry ;) * This commit should be atomic. If anything goes wrong (not counting problem originated from elsewhere), I will not hesitate to revert everything back within 12 hours. This substantial changes itself not a rocket science and the process has begun for almost 2 years, and lots of incremental changes are already in place during that period of time. * Some issues does occur in snd_emu10kx (note the 'x') due to various internal locking issues and it is currently being worked on by chibis. Tested by: chibis (Yuriy Tsibizov), joel, Alexandre Vieira, many innocent souls...
2007-06-16 03:37:28 +00:00
#define SND_DEV_DSPHW_CD 15 /* s16le/stereo 44100Hz CD */
#define SND_DEV_DSP_MMAP 16 /* OSSv4 compatible /dev/dsp_mmap */
#define SND_DEV_LAST SND_DEV_DSP_MMAP
#define SND_DEV_MAX PCMMAXDEV
#define DSP_DEFAULT_SPEED 8000
#define ON 1
#define OFF 0
extern int pcm_veto_load;
extern int snd_unit;
Welcome to Once-a-year Sound Mega-Commit. Enjoy numerous updates and fixes in every sense. General ------- - Multichannel safe, endian safe, format safe * Large part of critical pcm filters such as vchan.c, feeder_rate.c, feeder_volume.c, feeder_fmt.c and feeder.c has been rewritten so that using them does not cause the pcm data to be converted to 16bit little endian. * Macrosses for accessing pcm data safely are defined within sound.h in the form of PCM_READ_* / PCM_WRITE_* * Currently, most of them are probably limited for mono/stereo handling, but the future addition of true multichannel will be much easier. - Low latency operation * Well, this require lot more works to do not just within sound driver, but we're heading towards right direction. Buffer/block sizing within channel.c is rewritten to calculate precise allocation for various combination of sample/data/rate size. As a result, applying correct SNDCTL_DSP_POLICY value will achive expected latency behaviour simmilar to what commercial 4front driver do. * Signal handling fix. ctrl+c of "cat /dev/zero > /dev/dsp" does not result long delay. * Eliminate sound truncation if the sound data is too small. DIY: 1) Download / extract http://people.freebsd.org/~ariff/lowlatency/shortfiles.tar.gz 2) Do a comparison between "cat state*.au > /dev/dsp" and "for x in state*.au ; do cat $x > /dev/dsp ; done" - there should be no "perceivable" differences. Double close for PR kern/31445. CAVEAT: Low latency come with (unbearable) price especially for poorly written applications. Applications that trying to act smarter by requesting (wrong) blocksize/blockcount will suffer the most. Fixup samples/patches can be found at: http://people.freebsd.org/~ariff/ports/ - Switch minimum/maximum sampling rate limit to "1" and "2016000" (48k * 42) due to closer compatibility with 4front driver. Discussed with: marcus@ (long time ago?) - All driver specific sysctls in the form of "hw.snd.pcm%d.*" have been moved to their own dev sysctl nodes, notably: hw.snd.pcm%d.vchans -> dev.pcm.%d.vchans Bump __FreeBSD_version. Driver specific --------------- - Ditto for sysctls. - snd_atiixp, snd_es137x, snd_via8233, snd_hda * Numerous cleanups and fixes. * _EXPERIMENTAL_ polling mode support using simple callout_* mechanisme. This was intended for pure debugging and latency measurement, but proven good enough in few unexpected and rare cases (such as problematic shared IRQ with GIANT devices - USB). Polling can be enabled/disabled through dev.pcm.0.polling. Disabled by default. - snd_ich * Fix possible overflow during speed calibration. Delay final initialization (pcm_setstatus) after calibration finished. PR: kern/100169 Tested by: Kevin Overman <oberman@es.net> * Inverted EAPD for few Nec VersaPro. PR: kern/104715 Submitted by: KAWATA Masahiko <kawata@mta.biglobe.ne.jp> Thanks to various people, notably Joel Dahl, Yuriy Tsibizov, Kevin Oberman, those at #freebsd-azalia @ freenode and others for testing. Joel Dahl will do the manpage update.
2006-11-26 12:24:06 +00:00
extern int snd_maxautovchans;
extern int snd_verbose;
extern devclass_t pcm_devclass;
MFp4 the sound Google Summer of Code project: The goal was to sync with the OSSv4 API 4Front Technologies uses in their proprietary OSS driver. This was successful as far as possible. The part of the API which is stable is implemented, for the rest there are some stubs already. New system ioctls: - SNDCTL_SYSINFO - obtain audio system info (version, # of audio/midi/ mixer devices, etc.) - SNDCTL_AUDIOINFO - fetch details about a specific audio device - SNDCTL_MIXERINFO - fetch details about a specific mixer device New audio ioctls: - Sync groups (SNDCTL_DSP_SYNCGROUP/SNDCTL_DSP_SYNCSTART) which allow triggered playback/recording on multiple devices (even across processes simultaneously). - Peak meters (SNDCTL_DSP_GETIPEAKS/SNDCTL_DSP_GETOPEAKS) - can query audio drivers for peak levels (needs driver support, disabled for now). - Per channel playback/recording levels - SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL. Note that these are still in name only, just wrapping around the AC97-style mixer at the moment. The next step is to push them down to the drivers. Audio ioctls still under development by 4Front (for which stubs may exist in this commit): - SNDCTL_GETNAME, SNDCTL_{GET,SET}{SONG,LABEL} - SNDCTL_DSP_{GET,SET}_CHNORDER - SNDCTL_MIX_ENUMINFO, SNDCTL_MIX_EXTINFO - (might be documented enough in the OSS releases to work on this. These ioctls cover the cool "twiddle any knob on your card" features.) Missing: - SNDCTL_DSP_COOKEDMODE -- this ioctl is used to give applications direct access to a card's buffers, bypassing the feeder architecture. It's a toughy -- "someone" needs to decide : (a) if this is desireable, and (b) if it's reasonably feasible. Updates for driver writers: So far, only two routines to the channel class (in channel_if.m) are added. One is for fetching a list of discrete supported playback/recording rates of a channel, and the other is for fetching peak level info (useful for drawing peak meters). Interested parties may want to help pushing down SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL into the drivers. To use the new stuff you need to rebuild the sound drivers or your kernel (depending on if you use modules or not) and to install soundcard.h (a buildworld/installworld handles this). Sponsored by: Google SoC 2006 Submitted by: ryanb Many thanks to: 4Front Technologies for their cooperation, explanations and the nice license of their soundcard.h.
2006-09-23 20:45:47 +00:00
extern struct unrhdr *pcmsg_unrhdr;
/*
* some macros for debugging purposes
* DDB/DEB to enable/disable debugging stuff
* BVDDB to enable debugging when bootverbose
*/
#define BVDDB(x) if (bootverbose) x
#ifndef DEB
#define DEB(x)
#endif
SYSCTL_DECL(_hw_snd);
struct pcm_channel *pcm_getfakechan(struct snddev_info *d);
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
int pcm_chnalloc(struct snddev_info *d, struct pcm_channel **ch, int direction, pid_t pid, int devunit);
int pcm_chnrelease(struct pcm_channel *c);
int pcm_chnref(struct pcm_channel *c, int ref);
int pcm_inprog(struct snddev_info *d, int delta);
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
struct pcm_channel *pcm_chn_create(struct snddev_info *d, struct pcm_channel *parent, kobj_class_t cls, int dir, int num, void *devinfo);
int pcm_chn_destroy(struct pcm_channel *ch);
int pcm_chn_add(struct snddev_info *d, struct pcm_channel *ch);
int pcm_chn_remove(struct snddev_info *d, struct pcm_channel *ch);
int pcm_addchan(device_t dev, int dir, kobj_class_t cls, void *devinfo);
unsigned int pcm_getbuffersize(device_t dev, unsigned int minbufsz, unsigned int deflt, unsigned int maxbufsz);
int pcm_register(device_t dev, void *devinfo, int numplay, int numrec);
int pcm_unregister(device_t dev);
int pcm_setstatus(device_t dev, char *str);
u_int32_t pcm_getflags(device_t dev);
void pcm_setflags(device_t dev, u_int32_t val);
void *pcm_getdevinfo(device_t dev);
int snd_setup_intr(device_t dev, struct resource *res, int flags,
driver_intr_t hand, void *param, void **cookiep);
void *snd_mtxcreate(const char *desc, const char *type);
void snd_mtxfree(void *m);
void snd_mtxassert(void *m);
#define snd_mtxlock(m) mtx_lock(m)
#define snd_mtxunlock(m) mtx_unlock(m)
int sysctl_hw_snd_vchans(SYSCTL_HANDLER_ARGS);
typedef int (*sndstat_handler)(struct sbuf *s, device_t dev, int verbose);
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
int sndstat_acquire(struct thread *td);
int sndstat_release(struct thread *td);
int sndstat_register(device_t dev, char *str, sndstat_handler handler);
int sndstat_registerfile(char *str);
int sndstat_unregister(device_t dev);
int sndstat_unregisterfile(char *str);
#define SND_DECLARE_FILE(version) \
_SND_DECLARE_FILE(__LINE__, version)
#define _SND_DECLARE_FILE(uniq, version) \
__SND_DECLARE_FILE(uniq, version)
#define __SND_DECLARE_FILE(uniq, version) \
static char sndstat_vinfo[] = version; \
SYSINIT(sdf_ ## uniq, SI_SUB_DRIVERS, SI_ORDER_MIDDLE, sndstat_registerfile, sndstat_vinfo); \
SYSUNINIT(sdf_ ## uniq, SI_SUB_DRIVERS, SI_ORDER_MIDDLE, sndstat_unregisterfile, sndstat_vinfo);
/* usage of flags in device config entry (config file) */
#define DV_F_DRQ_MASK 0x00000007 /* mask for secondary drq */
#define DV_F_DUAL_DMA 0x00000010 /* set to use secondary dma channel */
/* ought to be made obsolete but still used by mss */
#define DV_F_DEV_MASK 0x0000ff00 /* force device type/class */
#define DV_F_DEV_SHIFT 8 /* force device type/class */
#define PCM_DEBUG_MTX
/*
* this is rather kludgey- we need to duplicate these struct def'ns from sound.c
* so that the macro versions of pcm_{,un}lock can dereference them.
* we also have to do this now makedev() has gone away.
*/
struct snddev_info {
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
struct {
struct {
SLIST_HEAD(, pcm_channel) head;
struct {
SLIST_HEAD(, pcm_channel) head;
} busy;
} pcm;
} channels;
Last (again ?!?) major commit for RELENG_7, featuring total Giant eradication in/from userland path, countless locking fixes, etc. - General sleep call through msleep(9) has been converted to condvar(9) with better consistencies. - Heavily guard every possible "slow path" entries (open(), close(), few ioctl()s, sysctls), but once it entering "fast path" (io, interrupt started), they are free to fly on their own. - Rearrange locking sequences, resulting better concurrency and serialization. Large part doesn't even need locking at all, and will be removed in future. Less clutter, except in few places due to lock ordering. - Anonymous mixer object creation/deletion to simplify mixer handling beyond typical mixer ioctls. Submitted by: chibis (with modifications) - Add few mix_[get|set|..] functions to avoid calling mixer_ioctl() directly using cryptic arguments. - Locking fixes to avoid possible deadlock with (still under Giant) USB. - Better simplex/duplex device handling. - Recover mmap() functionality for recording, which has been lost since 2.2.x - 3.x (the introduction of newpcm). Full-duplex mmap still doesn't work (due to VM/page design), but people still can mmap both by opening each direction separately. mmaped playback is guarantee to work either way. - New sysctl: "hw.snd.compat_linux_mmap" to allow PROT_EXEC page mapping, due to recent changes in linux compatibility layer which require it. All linux applications that using sound + mmap() (mostly games) require this to be enabled. Disabled by default. - Other goodies.. too many, that will increase releng7 shareholder value and make users of releng6 (and below) cry ;) * This commit should be atomic. If anything goes wrong (not counting problem originated from elsewhere), I will not hesitate to revert everything back within 12 hours. This substantial changes itself not a rocket science and the process has begun for almost 2 years, and lots of incremental changes are already in place during that period of time. * Some issues does occur in snd_emu10kx (note the 'x') due to various internal locking issues and it is currently being worked on by chibis. Tested by: chibis (Yuriy Tsibizov), joel, Alexandre Vieira, many innocent souls...
2007-06-16 03:37:28 +00:00
TAILQ_HEAD(dsp_cdevinfo_linkhead, dsp_cdevinfo) dsp_cdevinfo_pool;
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
struct snd_clone *clones;
struct pcm_channel *fakechan;
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
unsigned devcount, playcount, reccount, pvchancount, rvchancount ;
unsigned flags;
int inprog;
unsigned int bufsz;
void *devinfo;
device_t dev;
char status[SND_STATUSLEN];
struct mtx *lock;
struct cdev *mixer_dev;
Last major commit and updates for RELENG_7: - Rework the entire pcm_channel structure: * Remove rarely used link placeholder, instead, make each pcm_channel as head/link of each own/each other. Unlock - Lock sequence due to sleep malloc has been reduced. * Implement "busy" queue which will contain list of busy/active channels. This greatly reduce locking contention for example while servicing interrupt for hardware with many channels or when virtual channels reach its 256 peak channels. - So I heard you like v chan ... O RLY? Welcome to Virtual **Record** Channels (vrec, rec vchans, vchans for recording, Rec-Chan, you decide), the ultimate solutions for your nagging O_RDWR full-duplex wannabe (note: flash plugins) monopolizing single record channel causing EBUSY. Vrec works exactly like Vchans (or, should I rename it to "Vplay" :) , except that it operates on the opposite direction (recording). Up to 256 vrecs (like vchans) are possible. Notes: * Relocate dev.pcm.%d.{vchans,vchanformat,vchanrate} to each of its respective node/direction: dev.pcm.%d.play.* for "play" (cdev = dsp%d.vp%d) dev.pcm.%d.rec.* for "record" (cdev = dsp%d.vr%d) * Don't expect that it will magically give you ability to split "recording source" (eg: 1 channel for cdrom, 1 channel for mic, etc). Just admit that you only have a *single* recording source / channel. Please bug your hardware vendor instead :) - Bump maxautovchans from 4 to 16. For a full-fledged multimedia desktop/workstation with too many soundservers installed (esound, artsd, jackd, pulse/polypaudio, ding-dong pling plong mudkip fuh fuh, etc), 4 seems inadequate. There will be no memory penalty here, since virtual channels are allocate only by demand. - Nuke/Rework the entire statically created cdev entries. Everything is clonable through snd own clone manager which designed to withstand many kind of abusive devfs droids such as: * while : ; do /bin/test -e /dev/dsp ; done * jot 16777216 0 | while read x ; do ls /dev/dsp0.$x ; done * hundreds (could be thousands) concurrent threads/process opening "/dev/dsp" (previously, this might result EBUSY even with just 3 contesting threads/procs). o Reusable clone objects (instead of creating new one like there's no tomorrow) after certain expiration deadline. The clone allocator will decide whether to reuse, share, or creating new clone. o Automatic garbage collector. - Dynamic unit magic allocator. Maximum attached soundcards can be tuned using tunable "hw.snd.maxunit" (Default to 512). Minimum is 16, and maximum is 2048. - ..other fixes, mostly related to concurrency issues. joel@ will do the manpage updates on sound(4). Have fun.
2007-05-31 18:43:33 +00:00
uint32_t pvchanrate, pvchanformat;
uint32_t rvchanrate, rvchanformat;
struct sysctl_ctx_list play_sysctl_ctx, rec_sysctl_ctx;
struct sysctl_oid *play_sysctl_tree, *rec_sysctl_tree;
Last (again ?!?) major commit for RELENG_7, featuring total Giant eradication in/from userland path, countless locking fixes, etc. - General sleep call through msleep(9) has been converted to condvar(9) with better consistencies. - Heavily guard every possible "slow path" entries (open(), close(), few ioctl()s, sysctls), but once it entering "fast path" (io, interrupt started), they are free to fly on their own. - Rearrange locking sequences, resulting better concurrency and serialization. Large part doesn't even need locking at all, and will be removed in future. Less clutter, except in few places due to lock ordering. - Anonymous mixer object creation/deletion to simplify mixer handling beyond typical mixer ioctls. Submitted by: chibis (with modifications) - Add few mix_[get|set|..] functions to avoid calling mixer_ioctl() directly using cryptic arguments. - Locking fixes to avoid possible deadlock with (still under Giant) USB. - Better simplex/duplex device handling. - Recover mmap() functionality for recording, which has been lost since 2.2.x - 3.x (the introduction of newpcm). Full-duplex mmap still doesn't work (due to VM/page design), but people still can mmap both by opening each direction separately. mmaped playback is guarantee to work either way. - New sysctl: "hw.snd.compat_linux_mmap" to allow PROT_EXEC page mapping, due to recent changes in linux compatibility layer which require it. All linux applications that using sound + mmap() (mostly games) require this to be enabled. Disabled by default. - Other goodies.. too many, that will increase releng7 shareholder value and make users of releng6 (and below) cry ;) * This commit should be atomic. If anything goes wrong (not counting problem originated from elsewhere), I will not hesitate to revert everything back within 12 hours. This substantial changes itself not a rocket science and the process has begun for almost 2 years, and lots of incremental changes are already in place during that period of time. * Some issues does occur in snd_emu10kx (note the 'x') due to various internal locking issues and it is currently being worked on by chibis. Tested by: chibis (Yuriy Tsibizov), joel, Alexandre Vieira, many innocent souls...
2007-06-16 03:37:28 +00:00
struct cv cv;
};
MFp4 the sound Google Summer of Code project: The goal was to sync with the OSSv4 API 4Front Technologies uses in their proprietary OSS driver. This was successful as far as possible. The part of the API which is stable is implemented, for the rest there are some stubs already. New system ioctls: - SNDCTL_SYSINFO - obtain audio system info (version, # of audio/midi/ mixer devices, etc.) - SNDCTL_AUDIOINFO - fetch details about a specific audio device - SNDCTL_MIXERINFO - fetch details about a specific mixer device New audio ioctls: - Sync groups (SNDCTL_DSP_SYNCGROUP/SNDCTL_DSP_SYNCSTART) which allow triggered playback/recording on multiple devices (even across processes simultaneously). - Peak meters (SNDCTL_DSP_GETIPEAKS/SNDCTL_DSP_GETOPEAKS) - can query audio drivers for peak levels (needs driver support, disabled for now). - Per channel playback/recording levels - SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL. Note that these are still in name only, just wrapping around the AC97-style mixer at the moment. The next step is to push them down to the drivers. Audio ioctls still under development by 4Front (for which stubs may exist in this commit): - SNDCTL_GETNAME, SNDCTL_{GET,SET}{SONG,LABEL} - SNDCTL_DSP_{GET,SET}_CHNORDER - SNDCTL_MIX_ENUMINFO, SNDCTL_MIX_EXTINFO - (might be documented enough in the OSS releases to work on this. These ioctls cover the cool "twiddle any knob on your card" features.) Missing: - SNDCTL_DSP_COOKEDMODE -- this ioctl is used to give applications direct access to a card's buffers, bypassing the feeder architecture. It's a toughy -- "someone" needs to decide : (a) if this is desireable, and (b) if it's reasonably feasible. Updates for driver writers: So far, only two routines to the channel class (in channel_if.m) are added. One is for fetching a list of discrete supported playback/recording rates of a channel, and the other is for fetching peak level info (useful for drawing peak meters). Interested parties may want to help pushing down SNDCTL_DSP_{GET,SET}{PLAY,REC}VOL into the drivers. To use the new stuff you need to rebuild the sound drivers or your kernel (depending on if you use modules or not) and to install soundcard.h (a buildworld/installworld handles this). Sponsored by: Google SoC 2006 Submitted by: ryanb Many thanks to: 4Front Technologies for their cooperation, explanations and the nice license of their soundcard.h.
2006-09-23 20:45:47 +00:00
void sound_oss_sysinfo(oss_sysinfo *);
#ifdef PCM_DEBUG_MTX
#define pcm_lock(d) mtx_lock(((struct snddev_info *)(d))->lock)
#define pcm_unlock(d) mtx_unlock(((struct snddev_info *)(d))->lock)
#else
void pcm_lock(struct snddev_info *d);
void pcm_unlock(struct snddev_info *d);
#endif
Last (again ?!?) major commit for RELENG_7, featuring total Giant eradication in/from userland path, countless locking fixes, etc. - General sleep call through msleep(9) has been converted to condvar(9) with better consistencies. - Heavily guard every possible "slow path" entries (open(), close(), few ioctl()s, sysctls), but once it entering "fast path" (io, interrupt started), they are free to fly on their own. - Rearrange locking sequences, resulting better concurrency and serialization. Large part doesn't even need locking at all, and will be removed in future. Less clutter, except in few places due to lock ordering. - Anonymous mixer object creation/deletion to simplify mixer handling beyond typical mixer ioctls. Submitted by: chibis (with modifications) - Add few mix_[get|set|..] functions to avoid calling mixer_ioctl() directly using cryptic arguments. - Locking fixes to avoid possible deadlock with (still under Giant) USB. - Better simplex/duplex device handling. - Recover mmap() functionality for recording, which has been lost since 2.2.x - 3.x (the introduction of newpcm). Full-duplex mmap still doesn't work (due to VM/page design), but people still can mmap both by opening each direction separately. mmaped playback is guarantee to work either way. - New sysctl: "hw.snd.compat_linux_mmap" to allow PROT_EXEC page mapping, due to recent changes in linux compatibility layer which require it. All linux applications that using sound + mmap() (mostly games) require this to be enabled. Disabled by default. - Other goodies.. too many, that will increase releng7 shareholder value and make users of releng6 (and below) cry ;) * This commit should be atomic. If anything goes wrong (not counting problem originated from elsewhere), I will not hesitate to revert everything back within 12 hours. This substantial changes itself not a rocket science and the process has begun for almost 2 years, and lots of incremental changes are already in place during that period of time. * Some issues does occur in snd_emu10kx (note the 'x') due to various internal locking issues and it is currently being worked on by chibis. Tested by: chibis (Yuriy Tsibizov), joel, Alexandre Vieira, many innocent souls...
2007-06-16 03:37:28 +00:00
/*
* For PCM_CV_[WAIT | ACQUIRE | RELEASE], be sure to surround these
* with pcm_lock/unlock() sequence, or I'll come to gnaw upon you!
*/
#ifdef SND_DIAGNOSTIC
#define PCM_WAIT(x) do { \
if (mtx_owned((x)->lock) == 0) \
panic("%s(%d): [PCM WAIT] Mutex not owned!", \
__func__, __LINE__); \
while ((x)->flags & SD_F_BUSY) { \
if (snd_verbose > 3) \
device_printf((x)->dev, \
"%s(%d): [PCM WAIT] calling cv_wait().\n", \
__func__, __LINE__); \
cv_wait(&(x)->cv, (x)->lock); \
} \
} while(0)
#define PCM_ACQUIRE(x) do { \
if (mtx_owned((x)->lock) == 0) \
panic("%s(%d): [PCM ACQUIRE] Mutex not owned!", \
__func__, __LINE__); \
if ((x)->flags & SD_F_BUSY) \
panic("%s(%d): [PCM ACQUIRE] " \
"Trying to acquire BUSY cv!", __func__, __LINE__); \
(x)->flags |= SD_F_BUSY; \
} while(0)
#define PCM_RELEASE(x) do { \
if (mtx_owned((x)->lock) == 0) \
panic("%s(%d): [PCM RELEASE] Mutex not owned!", \
__func__, __LINE__); \
if ((x)->flags & SD_F_BUSY) { \
(x)->flags &= ~SD_F_BUSY; \
if ((x)->cv.cv_waiters != 0) { \
if ((x)->cv.cv_waiters > 1 && snd_verbose > 3) \
device_printf((x)->dev, \
"%s(%d): [PCM RELEASE] " \
"cv_waiters=%d > 1!\n", \
__func__, __LINE__, \
(x)->cv.cv_waiters); \
cv_broadcast(&(x)->cv); \
} \
} else \
panic("%s(%d): [PCM RELEASE] Releasing non-BUSY cv!", \
__func__, __LINE__); \
} while(0)
/* Quick version, for shorter path. */
#define PCM_ACQUIRE_QUICK(x) do { \
if (mtx_owned((x)->lock) != 0) \
panic("%s(%d): [PCM ACQUIRE QUICK] Mutex owned!", \
__func__, __LINE__); \
pcm_lock(x); \
PCM_WAIT(x); \
PCM_ACQUIRE(x); \
pcm_unlock(x); \
} while(0)
#define PCM_RELEASE_QUICK(x) do { \
if (mtx_owned((x)->lock) != 0) \
panic("%s(%d): [PCM RELEASE QUICK] Mutex owned!", \
__func__, __LINE__); \
pcm_lock(x); \
PCM_RELEASE(x); \
pcm_unlock(x); \
} while(0)
#define PCM_BUSYASSERT(x) do { \
if (!((x) != NULL && ((x)->flags & SD_F_BUSY))) \
panic("%s(%d): [PCM BUSYASSERT] " \
"Failed, snddev_info=%p", __func__, __LINE__, x); \
} while(0)
#define PCM_GIANT_ENTER(x) do { \
int _pcm_giant = 0; \
if (mtx_owned((x)->lock) != 0) \
panic("%s(%d): [GIANT ENTER] PCM lock owned!", \
__func__, __LINE__); \
if (mtx_owned(&Giant) != 0 && snd_verbose > 3) \
device_printf((x)->dev, \
"%s(%d): [GIANT ENTER] Giant owned!\n", \
__func__, __LINE__); \
if (!((x)->flags & SD_F_MPSAFE) && mtx_owned(&Giant) == 0) \
do { \
mtx_lock(&Giant); \
_pcm_giant = 1; \
} while(0)
#define PCM_GIANT_EXIT(x) do { \
if (mtx_owned((x)->lock) != 0) \
panic("%s(%d): [GIANT EXIT] PCM lock owned!", \
__func__, __LINE__); \
if (!(_pcm_giant == 0 || _pcm_giant == 1)) \
panic("%s(%d): [GIANT EXIT] _pcm_giant screwed!", \
__func__, __LINE__); \
if ((x)->flags & SD_F_MPSAFE) { \
if (_pcm_giant == 1) \
panic("%s(%d): [GIANT EXIT] MPSAFE Giant?", \
__func__, __LINE__); \
if (mtx_owned(&Giant) != 0 && snd_verbose > 3) \
device_printf((x)->dev, \
"%s(%d): [GIANT EXIT] Giant owned!\n", \
__func__, __LINE__); \
} \
if (_pcm_giant != 0) { \
if (mtx_owned(&Giant) == 0) \
panic("%s(%d): [GIANT EXIT] Giant not owned!", \
__func__, __LINE__); \
_pcm_giant = 0; \
mtx_unlock(&Giant); \
} \
} while(0)
#else /* SND_DIAGNOSTIC */
#define PCM_WAIT(x) do { \
mtx_assert((x)->lock, MA_OWNED); \
while ((x)->flags & SD_F_BUSY) \
cv_wait(&(x)->cv, (x)->lock); \
} while(0)
#define PCM_ACQUIRE(x) do { \
mtx_assert((x)->lock, MA_OWNED); \
KASSERT(!((x)->flags & SD_F_BUSY), \
("%s(%d): [PCM ACQUIRE] Trying to acquire BUSY cv!", \
__func__, __LINE__)); \
(x)->flags |= SD_F_BUSY; \
} while(0)
#define PCM_RELEASE(x) do { \
mtx_assert((x)->lock, MA_OWNED); \
KASSERT((x)->flags & SD_F_BUSY, \
("%s(%d): [PCM RELEASE] Releasing non-BUSY cv!", \
__func__, __LINE__)); \
(x)->flags &= ~SD_F_BUSY; \
if ((x)->cv.cv_waiters != 0) \
cv_broadcast(&(x)->cv); \
} while(0)
/* Quick version, for shorter path. */
#define PCM_ACQUIRE_QUICK(x) do { \
mtx_assert((x)->lock, MA_NOTOWNED); \
pcm_lock(x); \
PCM_WAIT(x); \
PCM_ACQUIRE(x); \
pcm_unlock(x); \
} while(0)
#define PCM_RELEASE_QUICK(x) do { \
mtx_assert((x)->lock, MA_NOTOWNED); \
pcm_lock(x); \
PCM_RELEASE(x); \
pcm_unlock(x); \
} while(0)
#define PCM_BUSYASSERT(x) KASSERT(x != NULL && \
((x)->flags & SD_F_BUSY), \
("%s(%d): [PCM BUSYASSERT] " \
"Failed, snddev_info=%p", \
__func__, __LINE__, x))
#define PCM_GIANT_ENTER(x) do { \
int _pcm_giant = 0; \
mtx_assert((x)->lock, MA_NOTOWNED); \
if (!((x)->flags & SD_F_MPSAFE) && mtx_owned(&Giant) == 0) \
do { \
mtx_lock(&Giant); \
_pcm_giant = 1; \
} while(0)
#define PCM_GIANT_EXIT(x) do { \
mtx_assert((x)->lock, MA_NOTOWNED); \
KASSERT(_pcm_giant == 0 || _pcm_giant == 1, \
("%s(%d): [GIANT EXIT] _pcm_giant screwed!", \
__func__, __LINE__)); \
KASSERT(!((x)->flags & SD_F_MPSAFE) || \
(((x)->flags & SD_F_MPSAFE) && _pcm_giant == 0), \
("%s(%d): [GIANT EXIT] MPSAFE Giant?", \
__func__, __LINE__)); \
if (_pcm_giant != 0) { \
mtx_assert(&Giant, MA_OWNED); \
_pcm_giant = 0; \
mtx_unlock(&Giant); \
} \
} while(0)
#endif /* !SND_DIAGNOSTIC */
#define PCM_GIANT_LEAVE(x) \
PCM_GIANT_EXIT(x); \
} while(0)
#ifdef KLD_MODULE
#define PCM_KLDSTRING(a) ("kld " # a)
#else
#define PCM_KLDSTRING(a) ""
#endif
#endif /* _KERNEL */
#endif /* _OS_H_ */