freebsd-skq/sys/dev/sound/pcm/feeder_rate.c

641 lines
20 KiB
C
Raw Normal View History

/*-
* Copyright (c) 1999 Cameron Grant <cg@FreeBSD.org>
* Copyright (c) 2003 Orion Hodson <orion@FreeBSD.org>
* Copyright (c) 2005 Ariff Abdullah <ariff@FreeBSD.org>
* 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.
*/
/*
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
* 2006-02-21:
* ==========
*
* Major cleanup and overhaul to remove much redundant codes.
* Highlights:
* 1) Support for signed / unsigned 16, 24 and 32 bit,
* big / little endian,
* 2) Unlimited channels.
*
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
* 2005-06-11:
* ==========
*
* *New* and rewritten soft sample rate converter supporting arbitrary sample
* rates, fine grained scaling/coefficients and a unified up/down stereo
* converter. Most of the disclaimers from orion's notes also applies
* here, regarding linear interpolation deficiencies and pre/post
* anti-aliasing filtering issues. This version comes with a much simpler and
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
* tighter interface, although it works almost exactly like the older one.
*
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* *
* This new implementation is fully dedicated in memory of Cameron Grant, *
* the creator of the magnificent, highly addictive feeder infrastructure. *
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
* *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
*
* Orion's notes:
* =============
*
* This rate conversion code uses linear interpolation without any
* pre- or post- interpolation filtering to combat aliasing. This
* greatly limits the sound quality and should be addressed at some
* stage in the future.
*
* Since this accuracy of interpolation is sensitive and examination
* of the algorithm output is harder from the kernel, the code is
* designed to be compiled in the kernel and in a userland test
* harness. This is done by selectively including and excluding code
* with several portions based on whether _KERNEL is defined. It's a
* little ugly, but exceedingly useful. The testsuite and its
* revisions can be found at:
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
* http://people.freebsd.org/~orion/files/feedrate/
*
* Special thanks to Ken Marx for exposing flaws in the code and for
* testing revisions.
*/
#include <dev/sound/pcm/sound.h>
#include "feeder_if.h"
SND_DECLARE_FILE("$FreeBSD$");
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
#define RATE_ASSERT(x, y) /* KASSERT(x,y) */
#define RATE_TEST(x, y) /* if (!(x)) printf y */
#define RATE_TRACE(x...) /* printf(x) */
MALLOC_DEFINE(M_RATEFEEDER, "ratefeed", "pcm rate feeder");
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
/*
* Don't overflow 32bit integer, since everything is done
* within 32bit arithmetic.
*/
#define RATE_FACTOR_MIN 1
#define RATE_FACTOR_MAX PCM_S24_MAX
#define RATE_FACTOR_SAFE(val) (!((val) < RATE_FACTOR_MIN || \
(val) > RATE_FACTOR_MAX))
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
struct feed_rate_info;
typedef uint32_t (*feed_rate_converter)(struct feed_rate_info *,
uint8_t *, uint32_t);
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
struct feed_rate_info {
uint32_t src, dst; /* rounded source / destination rates */
uint32_t rsrc, rdst; /* original source / destination rates */
uint32_t gx, gy; /* interpolation / decimation ratio */
uint32_t alpha; /* interpolation distance */
uint32_t pos, bpos; /* current sample / buffer positions */
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
uint32_t bufsz; /* total buffer size limit */
uint32_t bufsz_init; /* allocated buffer size */
uint32_t channels; /* total channels */
uint32_t bps; /* bytes-per-sample */
#ifdef FEEDRATE_STRAY
uint32_t stray; /* stray bytes */
#endif
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
uint8_t *buffer;
feed_rate_converter convert;
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
};
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
int feeder_rate_min = FEEDRATE_RATEMIN;
int feeder_rate_max = FEEDRATE_RATEMAX;
int feeder_rate_round = FEEDRATE_ROUNDHZ;
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
TUNABLE_INT("hw.snd.feeder_rate_min", &feeder_rate_min);
TUNABLE_INT("hw.snd.feeder_rate_max", &feeder_rate_max);
TUNABLE_INT("hw.snd.feeder_rate_round", &feeder_rate_round);
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
static int
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
sysctl_hw_snd_feeder_rate_min(SYSCTL_HANDLER_ARGS)
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
{
int err, val;
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
val = feeder_rate_min;
err = sysctl_handle_int(oidp, &val, 0, req);
if (err != 0 || req->newptr == NULL)
return (err);
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
if (RATE_FACTOR_SAFE(val) && val < feeder_rate_max)
feeder_rate_min = val;
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
else
err = EINVAL;
return (err);
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
}
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
SYSCTL_PROC(_hw_snd, OID_AUTO, feeder_rate_min, CTLTYPE_INT | CTLFLAG_RW,
0, sizeof(int), sysctl_hw_snd_feeder_rate_min, "I",
"minimum allowable rate");
static int
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
sysctl_hw_snd_feeder_rate_max(SYSCTL_HANDLER_ARGS)
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
{
int err, val;
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
val = feeder_rate_max;
err = sysctl_handle_int(oidp, &val, 0, req);
if (err != 0 || req->newptr == NULL)
return (err);
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
if (RATE_FACTOR_SAFE(val) && val > feeder_rate_min)
feeder_rate_max = val;
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
else
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
err = EINVAL;
return (err);
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
}
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
SYSCTL_PROC(_hw_snd, OID_AUTO, feeder_rate_max, CTLTYPE_INT | CTLFLAG_RW,
0, sizeof(int), sysctl_hw_snd_feeder_rate_max, "I",
"maximum allowable rate");
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
static int
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
sysctl_hw_snd_feeder_rate_round(SYSCTL_HANDLER_ARGS)
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
{
int err, val;
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
val = feeder_rate_round;
err = sysctl_handle_int(oidp, &val, 0, req);
if (err != 0 || req->newptr == NULL)
return (err);
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
if (val < FEEDRATE_ROUNDHZ_MIN || val > FEEDRATE_ROUNDHZ_MAX)
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
err = EINVAL;
else
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
feeder_rate_round = val - (val % FEEDRATE_ROUNDHZ);
return (err);
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
}
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
SYSCTL_PROC(_hw_snd, OID_AUTO, feeder_rate_round, CTLTYPE_INT | CTLFLAG_RW,
0, sizeof(int), sysctl_hw_snd_feeder_rate_round, "I",
"sample rate converter rounding threshold");
#define FEEDER_RATE_CONVERT(FMTBIT, RATE_INTCAST, SIGN, SIGNS, ENDIAN, ENDIANS) \
static uint32_t \
feed_convert_##SIGNS##FMTBIT##ENDIANS(struct feed_rate_info *info, \
uint8_t *dst, uint32_t max) \
{ \
uint32_t ret, smpsz, ch, pos, bpos, gx, gy, alpha, d1, d2; \
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
int32_t x, y; \
int i; \
uint8_t *src, *sx, *sy; \
\
ret = 0; \
alpha = info->alpha; \
gx = info->gx; \
gy = info->gy; \
pos = info->pos; \
bpos = info->bpos; \
src = info->buffer + pos; \
ch = info->channels; \
smpsz = PCM_##FMTBIT##_BPS * ch; \
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
for (;;) { \
if (alpha < gx) { \
alpha += gy; \
pos += smpsz; \
if (pos == bpos) \
break; \
src += smpsz; \
} else { \
alpha -= gx; \
d1 = (alpha << PCM_FXSHIFT) / gy; \
d2 = (1U << PCM_FXSHIFT) - d1; \
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
sx = src - smpsz; \
sy = src; \
i = ch; \
do { \
x = PCM_READ_##SIGN##FMTBIT##_##ENDIAN(sx); \
y = PCM_READ_##SIGN##FMTBIT##_##ENDIAN(sy); \
x = (((RATE_INTCAST)x * d1) + \
((RATE_INTCAST)y * d2)) >> PCM_FXSHIFT; \
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
PCM_WRITE_##SIGN##FMTBIT##_##ENDIAN(dst, x); \
dst += PCM_##FMTBIT##_BPS; \
sx += PCM_##FMTBIT##_BPS; \
sy += PCM_##FMTBIT##_BPS; \
ret += PCM_##FMTBIT##_BPS; \
} while (--i != 0); \
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
if (ret == max) \
break; \
} \
} \
info->alpha = alpha; \
info->pos = pos; \
return (ret); \
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
}
FEEDER_RATE_CONVERT(8, int32_t, S, s, NE, ne)
FEEDER_RATE_CONVERT(16, int32_t, S, s, LE, le)
FEEDER_RATE_CONVERT(24, int32_t, S, s, LE, le)
FEEDER_RATE_CONVERT(32, intpcm_t, S, s, LE, le)
FEEDER_RATE_CONVERT(16, int32_t, S, s, BE, be)
FEEDER_RATE_CONVERT(24, int32_t, S, s, BE, be)
FEEDER_RATE_CONVERT(32, intpcm_t, S, s, BE, be)
FEEDER_RATE_CONVERT(8, int32_t, U, u, NE, ne)
FEEDER_RATE_CONVERT(16, int32_t, U, u, LE, le)
FEEDER_RATE_CONVERT(24, int32_t, U, u, LE, le)
FEEDER_RATE_CONVERT(32, intpcm_t, U, u, LE, le)
FEEDER_RATE_CONVERT(16, int32_t, U, u, BE, be)
FEEDER_RATE_CONVERT(24, int32_t, U, u, BE, be)
FEEDER_RATE_CONVERT(32, intpcm_t, U, u, BE, 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
static void
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
feed_speed_ratio(uint32_t src, uint32_t dst, uint32_t *gx, uint32_t *gy)
{
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
uint32_t w, x = src, y = dst;
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
while (y != 0) {
w = x % y;
x = y;
y = w;
}
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
*gx = src / x;
*gy = dst / x;
}
static void
feed_rate_reset(struct feed_rate_info *info)
{
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
info->src = info->rsrc - (info->rsrc %
((feeder_rate_round > 0) ? feeder_rate_round : 1));
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
info->dst = info->rdst - (info->rdst %
((feeder_rate_round > 0) ? feeder_rate_round : 1));
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
info->gx = 1;
info->gy = 1;
info->alpha = 0;
info->channels = 1;
info->bps = PCM_8_BPS;
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
info->convert = NULL;
info->bufsz = info->bufsz_init;
info->pos = 1;
info->bpos = 2;
#ifdef FEEDRATE_STRAY
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
info->stray = 0;
#endif
}
static int
feed_rate_setup(struct pcm_feeder *f)
{
struct feed_rate_info *info = f->data;
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
static const struct {
uint32_t format; /* pcm / audio format */
uint32_t bps; /* bytes-per-sample, regardless of
total channels */
feed_rate_converter convert;
} convtbl[] = {
{ AFMT_S8, PCM_8_BPS, feed_convert_s8ne },
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
{ AFMT_S16_LE, PCM_16_BPS, feed_convert_s16le },
{ AFMT_S24_LE, PCM_24_BPS, feed_convert_s24le },
{ AFMT_S32_LE, PCM_32_BPS, feed_convert_s32le },
{ AFMT_S16_BE, PCM_16_BPS, feed_convert_s16be },
{ AFMT_S24_BE, PCM_24_BPS, feed_convert_s24be },
{ AFMT_S32_BE, PCM_32_BPS, feed_convert_s32be },
{ AFMT_U8, PCM_8_BPS, feed_convert_u8ne },
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
{ AFMT_U16_LE, PCM_16_BPS, feed_convert_u16le },
{ AFMT_U24_LE, PCM_24_BPS, feed_convert_u24le },
{ AFMT_U32_LE, PCM_32_BPS, feed_convert_u32le },
{ AFMT_U16_BE, PCM_16_BPS, feed_convert_u16be },
{ AFMT_U24_BE, PCM_24_BPS, feed_convert_u24be },
{ AFMT_U32_BE, PCM_32_BPS, feed_convert_u32be },
{ 0, 0, NULL },
};
uint32_t i;
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
feed_rate_reset(info);
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
if (info->src != info->dst)
feed_speed_ratio(info->src, info->dst, &info->gx, &info->gy);
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
if (!(RATE_FACTOR_SAFE(info->gx) && RATE_FACTOR_SAFE(info->gy)))
return (-1);
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
for (i = 0; i < sizeof(convtbl) / sizeof(*convtbl); i++) {
if (convtbl[i].format == 0)
return (-1);
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
if ((f->desc->out & ~AFMT_STEREO) == convtbl[i].format) {
info->bps = convtbl[i].bps;
info->convert = convtbl[i].convert;
break;
}
}
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
/*
* No need to interpolate/decimate, just do plain copy.
*/
if (info->gx == info->gy)
info->convert = NULL;
info->channels = (f->desc->out & AFMT_STEREO) ? 2 : 1;
info->pos = info->bps * info->channels;
info->bpos = info->pos << 1;
info->bufsz -= info->bufsz % info->pos;
memset(info->buffer, sndbuf_zerodata(f->desc->out), info->bpos);
RATE_TRACE("%s: %u (%u) -> %u (%u) [%u/%u] , "
"format=0x%08x, channels=%u, bufsz=%u\n",
__func__, info->src, info->rsrc, info->dst, info->rdst,
info->gx, info->gy, f->desc->out, info->channels,
info->bufsz - info->pos);
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
return (0);
}
static int
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
feed_rate_set(struct pcm_feeder *f, int what, int32_t value)
{
struct feed_rate_info *info = f->data;
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
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
if (value < feeder_rate_min || value > feeder_rate_max)
return (-1);
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
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
switch (what) {
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
case FEEDRATE_SRC:
info->rsrc = value;
break;
case FEEDRATE_DST:
info->rdst = value;
break;
default:
return (-1);
}
return (feed_rate_setup(f));
}
static int
feed_rate_get(struct pcm_feeder *f, int what)
{
struct feed_rate_info *info = f->data;
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
switch (what) {
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
case FEEDRATE_SRC:
return (info->rsrc);
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
case FEEDRATE_DST:
return (info->rdst);
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
default:
return (-1);
}
return (-1);
}
static int
feed_rate_init(struct pcm_feeder *f)
{
struct feed_rate_info *info;
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
if (f->desc->out != f->desc->in)
return (EINVAL);
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
info = malloc(sizeof(*info), M_RATEFEEDER, M_NOWAIT | M_ZERO);
2003-04-20 17:08:56 +00:00
if (info == NULL)
return (ENOMEM);
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
/*
* bufsz = sample from last cycle + conversion space
*/
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
info->bufsz_init = 8 + feeder_buffersize;
info->buffer = malloc(sizeof(*info->buffer) * info->bufsz_init,
M_RATEFEEDER, M_NOWAIT | M_ZERO);
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
if (info->buffer == NULL) {
free(info, M_RATEFEEDER);
return (ENOMEM);
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
}
info->rsrc = DSP_DEFAULT_SPEED;
info->rdst = DSP_DEFAULT_SPEED;
f->data = info;
return (feed_rate_setup(f));
}
static int
feed_rate_free(struct pcm_feeder *f)
{
struct feed_rate_info *info = f->data;
if (info != NULL) {
if (info->buffer != NULL)
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
free(info->buffer, M_RATEFEEDER);
free(info, M_RATEFEEDER);
}
f->data = NULL;
return (0);
}
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
static int
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
feed_rate(struct pcm_feeder *f, struct pcm_channel *c, uint8_t *b,
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
uint32_t count, void *source)
{
struct feed_rate_info *info = f->data;
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
uint32_t i, smpsz;
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
int32_t fetch, slot;
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
if (info->convert == NULL)
return (FEEDER_FEED(f->source, c, b, count, source));
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
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
/*
* This loop has been optimized to generalize both up / down
* sampling without causing missing samples or excessive buffer
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
* feeding. The tricky part is to calculate *precise* (slot) value
* needed for the entire conversion space since we are bound to
* return and fill up the buffer according to the requested 'count'.
* Too much feeding will cause the extra buffer stay within temporary
* circular buffer forever and always manifest itself as a truncated
* sound during end of playback / recording. Too few, and we end up
* with possible underruns and waste of cpu cycles.
*
* 'Stray' management exist to combat with possible unaligned
* buffering by the caller.
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
*/
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
smpsz = info->bps * info->channels;
RATE_TEST(count >= smpsz && (count % smpsz) == 0,
("%s: Count size not sample integral (%d)\n", __func__, count));
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
if (count < smpsz)
return (0);
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
count -= count % smpsz;
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
/*
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
* This slot count formula will stay here for the next million years
* to come. This is the key of our circular buffering precision.
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
*/
slot = (((info->gx * (count / smpsz)) + info->gy - info->alpha - 1) /
info->gy) * smpsz;
RATE_TEST((slot % smpsz) == 0,
("%s: Slot count not sample integral (%d)\n", __func__, slot));
#ifdef FEEDRATE_STRAY
RATE_TEST(info->stray == 0, ("%s: [1] Stray bytes: %u\n", __func__,
info->stray));
#endif
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
if (info->pos != smpsz && info->bpos - info->pos == smpsz &&
info->bpos + slot > info->bufsz) {
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
/*
* Copy last unit sample and its previous to
* beginning of buffer.
*/
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
bcopy(info->buffer + info->pos - smpsz, info->buffer,
sizeof(*info->buffer) * (smpsz << 1));
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
info->pos = smpsz;
info->bpos = smpsz << 1;
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
}
RATE_ASSERT(slot >= 0, ("%s: Negative Slot: %d\n", __func__, slot));
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
i = 0;
for (;;) {
for (;;) {
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
fetch = info->bufsz - info->bpos;
#ifdef FEEDRATE_STRAY
fetch -= info->stray;
#endif
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
RATE_ASSERT(fetch >= 0,
("%s: [1] Buffer overrun: %d > %d\n", __func__,
info->bpos, info->bufsz));
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
if (slot < fetch)
fetch = slot;
#ifdef FEEDRATE_STRAY
if (fetch < 1)
#else
if (fetch < smpsz)
#endif
break;
RATE_ASSERT((int)(info->bpos
#ifdef FEEDRATE_STRAY
- info->stray
#endif
) >= 0 &&
(info->bpos - info->stray) < info->bufsz,
("%s: DANGER - BUFFER OVERRUN! bufsz=%d, pos=%d\n",
__func__, info->bufsz, info->bpos
#ifdef FEEDRATE_STRAY
- info->stray
#endif
));
fetch = FEEDER_FEED(f->source, c,
info->buffer + info->bpos
#ifdef FEEDRATE_STRAY
- info->stray
#endif
, fetch, source);
#ifdef FEEDRATE_STRAY
info->stray = 0;
if (fetch == 0)
#else
if (fetch < smpsz)
#endif
break;
RATE_TEST((fetch % smpsz) == 0,
("%s: Fetch size not sample integral (%d)\n",
__func__, fetch));
#ifdef FEEDRATE_STRAY
info->stray += fetch % smpsz;
RATE_TEST(info->stray == 0,
("%s: Stray bytes detected (%d)\n", __func__,
info->stray));
#endif
fetch -= fetch % smpsz;
info->bpos += fetch;
slot -= fetch;
RATE_ASSERT(slot >= 0, ("%s: Negative Slot: %d\n",
__func__, slot));
if (slot == 0 || info->bpos == info->bufsz)
break;
}
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
if (info->pos == info->bpos) {
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
RATE_TEST(info->pos == smpsz,
("%s: EOF while in progress\n", __func__));
break;
}
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
RATE_ASSERT(info->pos <= info->bpos,
("%s: [2] Buffer overrun: %d > %d\n", __func__, info->pos,
info->bpos));
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
RATE_ASSERT(info->pos < info->bpos,
("%s: Zero buffer!\n", __func__));
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
RATE_ASSERT(((info->bpos - info->pos) % smpsz) == 0,
("%s: Buffer not sample integral (%d)\n", __func__,
info->bpos - info->pos));
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
i += info->convert(info, b + i, count - i);
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
RATE_ASSERT(info->pos <= info->bpos,
("%s: [3] Buffer overrun: %d > %d\n", __func__, info->pos,
info->bpos));
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
if (info->pos == info->bpos) {
/*
* End of buffer cycle. Copy last unit sample
* to beginning of buffer so next cycle can
* interpolate using it.
*/
#ifdef FEEDRATE_STRAY
RATE_TEST(info->stray == 0,
("%s: [2] Stray bytes: %u\n", __func__,
info->stray));
#endif
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
bcopy(info->buffer + info->pos - smpsz, info->buffer,
sizeof(*info->buffer) * smpsz);
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
info->bpos = smpsz;
info->pos = smpsz;
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
}
if (i == count)
break;
}
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
RATE_TEST((slot == 0 && count == i) || (slot > 0 && count > i &&
info->pos == info->bpos && info->pos == smpsz),
("%s: Inconsistent slot/count! "
"Count Expect: %u , Got: %u, Slot Left: %d\n", __func__, count, i,
slot));
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
#ifdef FEEDRATE_STRAY
RATE_TEST(info->stray == 0, ("%s: [3] Stray bytes: %u\n", __func__,
info->stray));
#endif
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
return (i);
}
static struct pcm_feederdesc feeder_rate_desc[] = {
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
{FEEDER_RATE, AFMT_S8, AFMT_S8, 0},
{FEEDER_RATE, AFMT_S16_LE, AFMT_S16_LE, 0},
{FEEDER_RATE, AFMT_S24_LE, AFMT_S24_LE, 0},
{FEEDER_RATE, AFMT_S32_LE, AFMT_S32_LE, 0},
{FEEDER_RATE, AFMT_S16_BE, AFMT_S16_BE, 0},
{FEEDER_RATE, AFMT_S24_BE, AFMT_S24_BE, 0},
{FEEDER_RATE, AFMT_S32_BE, AFMT_S32_BE, 0},
{FEEDER_RATE, AFMT_S8 | AFMT_STEREO, AFMT_S8 | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_S16_LE | AFMT_STEREO, AFMT_S16_LE | AFMT_STEREO, 0},
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
{FEEDER_RATE, AFMT_S24_LE | AFMT_STEREO, AFMT_S24_LE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_S32_LE | AFMT_STEREO, AFMT_S32_LE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_S16_BE | AFMT_STEREO, AFMT_S16_BE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_S24_BE | AFMT_STEREO, AFMT_S24_BE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_S32_BE | AFMT_STEREO, AFMT_S32_BE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U8, AFMT_U8, 0},
{FEEDER_RATE, AFMT_U16_LE, AFMT_U16_LE, 0},
{FEEDER_RATE, AFMT_U24_LE, AFMT_U24_LE, 0},
{FEEDER_RATE, AFMT_U32_LE, AFMT_U32_LE, 0},
{FEEDER_RATE, AFMT_U16_BE, AFMT_U16_BE, 0},
{FEEDER_RATE, AFMT_U24_BE, AFMT_U24_BE, 0},
{FEEDER_RATE, AFMT_U32_BE, AFMT_U32_BE, 0},
{FEEDER_RATE, AFMT_U8 | AFMT_STEREO, AFMT_U8 | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U16_LE | AFMT_STEREO, AFMT_U16_LE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U24_LE | AFMT_STEREO, AFMT_U24_LE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U32_LE | AFMT_STEREO, AFMT_U32_LE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U16_BE | AFMT_STEREO, AFMT_U16_BE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U24_BE | AFMT_STEREO, AFMT_U24_BE | AFMT_STEREO, 0},
{FEEDER_RATE, AFMT_U32_BE | AFMT_STEREO, AFMT_U32_BE | AFMT_STEREO, 0},
{0, 0, 0, 0},
};
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
static kobj_method_t feeder_rate_methods[] = {
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
KOBJMETHOD(feeder_init, feed_rate_init),
KOBJMETHOD(feeder_free, feed_rate_free),
KOBJMETHOD(feeder_set, feed_rate_set),
KOBJMETHOD(feeder_get, feed_rate_get),
KOBJMETHOD(feeder_feed, feed_rate),
{0, 0}
};
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
FEEDER_DECLARE(feeder_rate, 2, NULL);