d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6ca048f825
freebsd-dev/sys/dev/sound/isa/sb16.c

854 lines
20 KiB
C
Raw Normal View History

Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
/*
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
* Copyright (c) 1999 Cameron Grant <gandalf@vilnya.demon.co.uk>
Update the pcm driver to the most recent version. This should add support for Vibra16X, OPTi925, and bring in several assorted fixes to the code and documentation. Also present here are apm hooks so that laptops can properly reconfigure the hardware after suspend (tested on the Libretto50). Reviewed by: jordan
1998-10-02 17:26:37 +00:00
* Copyright 1997,1998 Luigi Rizzo.
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
*
* Derived from files in the Voxware 3.5 distribution,
* Copyright by Hannu Savolainen 1994, under the same copyright
* conditions.
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
* All rights reserved.
*
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
This updates Luigi's sound code to the basic code in snd971023... changes: o rip the old select from his distribution to prevent extra pollution o the code now uses audio dma, helps reduce clicks o improved card support, should work in full duplex on sb16 cards o add better voxware ioctl support pointed out by Joao Carlos Mendes Luis <jonny@coppe.ufrj.br> o remove an unused file that I included for more complete history o and MANY other changes I have personally tested this code with a CS4237 based card and an AWE32 (non-PnP). Both cards worked fine in 8bit and 16bit mode.
1997-10-31 12:24:28 +00:00
* 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
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
This updates Luigi's sound code to the basic code in snd971023... changes: o rip the old select from his distribution to prevent extra pollution o the code now uses audio dma, helps reduce clicks o improved card support, should work in full duplex on sb16 cards o add better voxware ioctl support pointed out by Joao Carlos Mendes Luis <jonny@coppe.ufrj.br> o remove an unused file that I included for more complete history o and MANY other changes I have personally tested this code with a CS4237 based card and an AWE32 (non-PnP). Both cards worked fine in 8bit and 16bit mode.
1997-10-31 12:24:28 +00:00
*
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
* 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.
*
$Id$ -> $FreeBSD$
1999-09-05 15:21:05 +00:00
* $FreeBSD$
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
*/
repo-copied to make way for newmidi, this commit updates include paths
1999-11-20 16:50:33 +00:00
#include <dev/sound/pcm/sound.h>
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
repo-copied to make way for newmidi, this commit updates include paths
1999-11-20 16:50:33 +00:00
#include <dev/sound/isa/sb.h>
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
#include <dev/sound/chip.h>
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
#include <sys/mutex.h>
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
#include "mixer_if.h"
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
#define SB16_BUFFSIZE 4096
general tidyup
2000-01-10 03:22:28 +00:00
#define PLAIN_SB16(x) ((((x)->bd_flags) & (BD_F_SB16|BD_F_SB16X)) == BD_F_SB16)
make ess cards use a 64k buffer again, by implementing esschan_init()
1999-12-30 04:04:43 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
static u_int32_t sb16_fmt8[] = {
rework feeder sytem to allow feeders in klds modify driver capability reporting format to list every audio format seperately- required for above and because we could not previously indicate that mono was unsupported. there should be no functional impact.
2000-08-20 22:18:56 +00:00
AFMT_U8,
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
AFMT_STEREO | AFMT_U8,
rework feeder sytem to allow feeders in klds modify driver capability reporting format to list every audio format seperately- required for above and because we could not previously indicate that mono was unsupported. there should be no functional impact.
2000-08-20 22:18:56 +00:00
0
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
};
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
static struct pcmchan_caps sb16_caps8 = {5000, 45000, sb16_fmt8, 0};
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
static u_int32_t sb16_fmt16[] = {
rework feeder sytem to allow feeders in klds modify driver capability reporting format to list every audio format seperately- required for above and because we could not previously indicate that mono was unsupported. there should be no functional impact.
2000-08-20 22:18:56 +00:00
AFMT_S16_LE,
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
AFMT_STEREO | AFMT_S16_LE,
rework feeder sytem to allow feeders in klds modify driver capability reporting format to list every audio format seperately- required for above and because we could not previously indicate that mono was unsupported. there should be no functional impact.
2000-08-20 22:18:56 +00:00
0
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
};
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
static struct pcmchan_caps sb16_caps16 = {5000, 45000, sb16_fmt16, 0};
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
rework feeder sytem to allow feeders in klds modify driver capability reporting format to list every audio format seperately- required for above and because we could not previously indicate that mono was unsupported. there should be no functional impact.
2000-08-20 22:18:56 +00:00
static u_int32_t sb16x_fmt[] = {
AFMT_U8,
AFMT_STEREO | AFMT_U8,
AFMT_S16_LE,
AFMT_STEREO | AFMT_S16_LE,
0
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
};
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
static struct pcmchan_caps sb16x_caps = {5000, 49000, sb16x_fmt, 0};
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
struct sb_info;
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
struct sb_chinfo {
struct sb_info *parent;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
struct pcm_channel *channel;
struct snd_dbuf *buffer;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
int dir, run, dch;
u_int32_t fmt, spd, blksz;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
};
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
struct sb_info {
struct resource *io_base; /* I/O address for the board */
struct resource *irq;
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
struct resource *drq1;
struct resource *drq2;
detach support remove un-needed setdir functions add bus_teardown_intr calls where necessary destroy our dma tags where necessary destroy ac97 before releasing resources
2000-09-09 19:21:04 +00:00
void *ih;
general tidyup
2000-01-10 03:22:28 +00:00
bus_dma_tag_t parent_dmat;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
int bd_id;
u_long bd_flags; /* board-specific flags */
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
int prio, prio16;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
struct sb_chinfo pch, rch;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
device_t parent_dev;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
};
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
static void sb_lock(struct sb_info *sb);
static void sb_unlock(struct sb_info *sb);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int sb_rd(struct sb_info *sb, int reg);
static void sb_wr(struct sb_info *sb, int reg, u_int8_t val);
static int sb_cmd(struct sb_info *sb, u_char val);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/* static int sb_cmd1(struct sb_info *sb, u_char cmd, int val); */
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int sb_cmd2(struct sb_info *sb, u_char cmd, int val);
static u_int sb_get_byte(struct sb_info *sb);
static void sb_setmixer(struct sb_info *sb, u_int port, u_int value);
static int sb_getmixer(struct sb_info *sb, u_int port);
static int sb_reset_dsp(struct sb_info *sb);
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
static void sb_intr(void *arg);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
/*
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
* Common code for the midi and pcm functions
*
* sb_cmd write a single byte to the CMD port.
* sb_cmd1 write a CMD + 1 byte arg
* sb_cmd2 write a CMD + 2 byte arg
* sb_get_byte returns a single byte from the DSP data port
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
*/
static int
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
port_rd(struct resource *port, int off)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return bus_space_read_1(rman_get_bustag(port), rman_get_bushandle(port), off);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
}
static void
port_wr(struct resource *port, int off, u_int8_t data)
{
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return bus_space_write_1(rman_get_bustag(port), rman_get_bushandle(port), off, data);
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
}
static int
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb_rd(struct sb_info *sb, int reg)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
return port_rd(sb->io_base, reg);
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static void
sb_wr(struct sb_info *sb, int reg, u_int8_t val)
{
port_wr(sb->io_base, reg, val);
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
}
static int
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb_dspwr(struct sb_info *sb, u_char val)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
int i;
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
for (i = 0; i < 1000; i++) {
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
if ((sb_rd(sb, SBDSP_STATUS) & 0x80))
DELAY((i > 100)? 1000 : 10);
else {
sb_wr(sb, SBDSP_CMD, val);
return 1;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
}
}
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
if (curproc->p_intr_nesting_level == 0)
fix whitespace bogons
2001-03-25 21:43:24 +00:00
printf("sb_dspwr(0x%02x) timed out.\n", val);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
return 0;
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int
sb_cmd(struct sb_info *sb, u_char val)
{
#if 0
printf("sb_cmd: %x\n", val);
#endif
return sb_dspwr(sb, val);
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/*
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int
sb_cmd1(struct sb_info *sb, u_char cmd, int val)
{
#if 0
printf("sb_cmd1: %x, %x\n", cmd, val);
#endif
if (sb_dspwr(sb, cmd)) {
return sb_dspwr(sb, val & 0xff);
} else return 0;
}
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
*/
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int
sb_cmd2(struct sb_info *sb, u_char cmd, int val)
{
#if 0
printf("sb_cmd2: %x, %x\n", cmd, val);
#endif
if (sb_dspwr(sb, cmd)) {
return sb_dspwr(sb, val & 0xff) &&
sb_dspwr(sb, (val >> 8) & 0xff);
} else return 0;
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
/*
* in the SB, there is a set of indirect "mixer" registers with
* address at offset 4, data at offset 5
*/
static void
sb_setmixer(struct sb_info *sb, u_int port, u_int value)
{
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_lock(sb);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb_wr(sb, SB_MIX_ADDR, (u_char) (port & 0xff)); /* Select register */
DELAY(10);
sb_wr(sb, SB_MIX_DATA, (u_char) (value & 0xff));
DELAY(10);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_unlock(sb);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int
sb_getmixer(struct sb_info *sb, u_int port)
{
int val;
update of Luigi's sound drivers... this updates to 971117 plus a small sb change that was after that release..
1997-11-23 07:03:21 +00:00
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_lock(sb);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb_wr(sb, SB_MIX_ADDR, (u_char) (port & 0xff)); /* Select register */
DELAY(10);
val = sb_rd(sb, SB_MIX_DATA);
DELAY(10);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_unlock(sb);
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
return val;
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
}
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static u_int
sb_get_byte(struct sb_info *sb)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
int i;
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
for (i = 1000; i > 0; i--) {
if (sb_rd(sb, DSP_DATA_AVAIL) & 0x80)
return sb_rd(sb, DSP_READ);
else
DELAY(20);
}
return 0xffff;
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
static int
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb_reset_dsp(struct sb_info *sb)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
u_char b;
sb_lock(sb);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb_wr(sb, SBDSP_RST, 3);
DELAY(100);
sb_wr(sb, SBDSP_RST, 0);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
b = sb_get_byte(sb);
sb_unlock(sb);
if (b != 0xAA) {
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
DEB(printf("sb_reset_dsp 0x%lx failed\n",
rman_get_start(d->io_base)));
return ENXIO; /* Sorry */
}
return 0;
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
}
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/************************************************************/
struct sb16_mixent {
int reg;
int bits;
int ofs;
int stereo;
};
static const struct sb16_mixent sb16_mixtab[32] = {
[SOUND_MIXER_VOLUME] = { 0x30, 5, 3, 1 },
[SOUND_MIXER_PCM] = { 0x32, 5, 3, 1 },
[SOUND_MIXER_SYNTH] = { 0x34, 5, 3, 1 },
[SOUND_MIXER_CD] = { 0x36, 5, 3, 1 },
[SOUND_MIXER_LINE] = { 0x38, 5, 3, 1 },
[SOUND_MIXER_MIC] = { 0x3a, 5, 3, 0 },
[SOUND_MIXER_SPEAKER] = { 0x3b, 5, 3, 0 },
[SOUND_MIXER_IGAIN] = { 0x3f, 2, 6, 1 },
[SOUND_MIXER_OGAIN] = { 0x41, 2, 6, 1 },
[SOUND_MIXER_TREBLE] = { 0x44, 4, 4, 1 },
[SOUND_MIXER_BASS] = { 0x46, 4, 4, 1 },
};
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb16mix_init(struct snd_mixer *m)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
{
struct sb_info *sb = mix_getdevinfo(m);
mix_setdevs(m, SOUND_MASK_SYNTH | SOUND_MASK_PCM | SOUND_MASK_SPEAKER |
SOUND_MASK_LINE | SOUND_MASK_MIC | SOUND_MASK_CD |
SOUND_MASK_IGAIN | SOUND_MASK_OGAIN |
SOUND_MASK_VOLUME | SOUND_MASK_BASS | SOUND_MASK_TREBLE);
mix_setrecdevs(m, SOUND_MASK_SYNTH | SOUND_MASK_LINE |
SOUND_MASK_MIC | SOUND_MASK_CD);
sb_setmixer(sb, 0x3c, 0x1f); /* make all output active */
sb_setmixer(sb, 0x3d, 0); /* make all inputs-l off */
sb_setmixer(sb, 0x3e, 0); /* make all inputs-r off */
return 0;
}
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb16mix_set(struct snd_mixer *m, unsigned dev, unsigned left, unsigned right)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
{
struct sb_info *sb = mix_getdevinfo(m);
const struct sb16_mixent *e;
int max;
e = &sb16_mixtab[dev];
max = (1 << e->bits) - 1;
left = (left * max) / 100;
right = (right * max) / 100;
sb_setmixer(sb, e->reg, left << e->ofs);
if (e->stereo)
fix paste-o in mixer code - actually set right channel volume instead of doing the left channel twice.
2000-11-06 02:47:43 +00:00
sb_setmixer(sb, e->reg + 1, right << e->ofs);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
else
right = left;
left = (left * 100) / max;
right = (right * 100) / max;
return left | (right << 8);
}
static int
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb16mix_setrecsrc(struct snd_mixer *m, u_int32_t src)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
{
struct sb_info *sb = mix_getdevinfo(m);
u_char recdev;
recdev = 0;
if (src & SOUND_MASK_MIC)
recdev |= 0x01; /* mono mic */
if (src & SOUND_MASK_CD)
recdev |= 0x06; /* l+r cd */
if (src & SOUND_MASK_LINE)
recdev |= 0x18; /* l+r line */
if (src & SOUND_MASK_SYNTH)
recdev |= 0x60; /* l+r midi */
sb_setmixer(sb, SB16_IMASK_L, recdev);
sb_setmixer(sb, SB16_IMASK_R, recdev);
/*
* since the same volume controls apply to the input and
* output sections, the best approach to have a consistent
* behaviour among cards would be to disable the output path
* on devices which are used to record.
* However, since users like to have feedback, we only disable
* the mic -- permanently.
*/
sb_setmixer(sb, SB16_OMASK, 0x1f & ~1);
return src;
}
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
static kobj_method_t sb16mix_mixer_methods[] = {
KOBJMETHOD(mixer_init, sb16mix_init),
KOBJMETHOD(mixer_set, sb16mix_set),
KOBJMETHOD(mixer_setrecsrc, sb16mix_setrecsrc),
{ 0, 0 }
};
MIXER_DECLARE(sb16mix_mixer);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/************************************************************/
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
static void
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb16_release_resources(struct sb_info *sb, device_t dev)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
if (sb->irq) {
detach support remove un-needed setdir functions add bus_teardown_intr calls where necessary destroy our dma tags where necessary destroy ac97 before releasing resources
2000-09-09 19:21:04 +00:00
if (sb->ih)
bus_teardown_intr(dev, sb->irq, sb->ih);
bus_release_resource(dev, SYS_RES_IRQ, 0, sb->irq);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb->irq = 0;
}
fix a typo preventing the second dma channel being released use isa_dma_release when releasing dma channels
2001-03-16 21:04:14 +00:00
if (sb->drq2 && (sb->drq2 != sb->drq1)) {
isa_dma_release(rman_get_start(sb->drq2));
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
bus_release_resource(dev, SYS_RES_DRQ, 1, sb->drq2);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb->drq2 = 0;
}
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sb->drq1) {
fix a typo preventing the second dma channel being released use isa_dma_release when releasing dma channels
2001-03-16 21:04:14 +00:00
isa_dma_release(rman_get_start(sb->drq1));
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
bus_release_resource(dev, SYS_RES_DRQ, 0, sb->drq1);
sb->drq1 = 0;
}
if (sb->io_base) {
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
bus_release_resource(dev, SYS_RES_IOPORT, 0, sb->io_base);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
sb->io_base = 0;
}
detach support remove un-needed setdir functions add bus_teardown_intr calls where necessary destroy our dma tags where necessary destroy ac97 before releasing resources
2000-09-09 19:21:04 +00:00
if (sb->parent_dmat) {
bus_dma_tag_destroy(sb->parent_dmat);
sb->parent_dmat = 0;
}
free(sb, M_DEVBUF);
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
}
static int
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb16_alloc_resources(struct sb_info *sb, device_t dev)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
int rid;
rid = 0;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
if (!sb->io_base)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->io_base = bus_alloc_resource(dev, SYS_RES_IOPORT, &rid, 0, ~0, 1, RF_ACTIVE);
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
rid = 0;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
if (!sb->irq)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->irq = bus_alloc_resource(dev, SYS_RES_IRQ, &rid, 0, ~0, 1, RF_ACTIVE);
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
rid = 0;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
if (!sb->drq1)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->drq1 = bus_alloc_resource(dev, SYS_RES_DRQ, &rid, 0, ~0, 1, RF_ACTIVE);
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
rid = 1;
split up ess and sb code rewrite ess mixer to use native registers rewrite play/rec code to use more accurate timer when available add code to use audio2 for playback, but disable it as no irqs are generated
2000-03-28 18:31:01 +00:00
if (!sb->drq2)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->drq2 = bus_alloc_resource(dev, SYS_RES_DRQ, &rid, 0, ~0, 1, RF_ACTIVE);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
if (sb->io_base && sb->drq1 && sb->irq) {
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
int bs = SB16_BUFFSIZE;
allocate isa bounce buffers of the right size for ess/mss cards, fixes panics reported
2000-01-05 00:39:20 +00:00
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
isa_dma_acquire(rman_get_start(sb->drq1));
allocate isa bounce buffers of the right size for ess/mss cards, fixes panics reported
2000-01-05 00:39:20 +00:00
isa_dmainit(rman_get_start(sb->drq1), bs);
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
if (sb->drq2) {
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
isa_dma_acquire(rman_get_start(sb->drq2));
allocate isa bounce buffers of the right size for ess/mss cards, fixes panics reported
2000-01-05 00:39:20 +00:00
isa_dmainit(rman_get_start(sb->drq2), bs);
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
} else {
sb->drq2 = sb->drq1;
pcm_setflags(dev, pcm_getflags(dev) | SD_F_SIMPLEX);
update of Luigi's sound drivers... this updates to 971117 plus a small sb change that was after that release..
1997-11-23 07:03:21 +00:00
}
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
return 0;
} else return ENXIO;
}
Support ESS1868 (and probably ESS688 & ESS1668). Submitted by: Max Khon <fjoe@husky.iclub.nsu.ru>
1998-12-10 18:36:10 +00:00
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
/* sbc does locking for us */
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static void
sb_intr(void *arg)
{
struct sb_info *sb = (struct sb_info *)arg;
int reason = 3, c;
/*
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
* The Vibra16X has separate flags for 8 and 16 bit transfers, but
* I have no idea how to tell capture from playback interrupts...
*/
reason = 0;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_lock(sb);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
c = sb_getmixer(sb, IRQ_STAT);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
if (c & 1)
sb_rd(sb, DSP_DATA_AVAIL); /* 8-bit int ack */
if (c & 2)
sb_rd(sb, DSP_DATA_AVL16); /* 16-bit int ack */
sb_unlock(sb);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/*
* this tells us if the source is 8-bit or 16-bit dma. We
* have to check the io channel to map it to read or write...
*/
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
if (sb->bd_flags & BD_F_SB16X) {
if (c & 1) { /* 8-bit format */
if (sb->pch.fmt & AFMT_8BIT)
reason |= 1;
if (sb->rch.fmt & AFMT_8BIT)
reason |= 2;
}
if (c & 2) { /* 16-bit format */
if (sb->pch.fmt & AFMT_16BIT)
reason |= 1;
if (sb->rch.fmt & AFMT_16BIT)
reason |= 2;
}
} else {
if (c & 1) { /* 8-bit dma */
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sb->pch.dch == 1)
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
reason |= 1;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sb->rch.dch == 1)
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
reason |= 2;
}
if (c & 2) { /* 16-bit dma */
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sb->pch.dch == 2)
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
reason |= 1;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sb->rch.dch == 2)
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
reason |= 2;
}
}
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
#if 0
printf("sb_intr: reason=%d c=0x%x\n", reason, c);
#endif
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if ((reason & 1) && (sb->pch.run))
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
chn_intr(sb->pch.channel);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if ((reason & 2) && (sb->rch.run))
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
chn_intr(sb->rch.channel);
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
}
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb_setup(struct sb_info *sb)
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
{
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
struct sb_chinfo *ch;
u_int8_t v;
int l, pprio;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_lock(sb);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if (sb->bd_flags & BD_F_DMARUN)
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadma(sb->pch.buffer, PCMTRIG_STOP);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if (sb->bd_flags & BD_F_DMARUN2)
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadma(sb->rch.buffer, PCMTRIG_STOP);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->bd_flags &= ~(BD_F_DMARUN | BD_F_DMARUN2);
sb_reset_dsp(sb);
if (sb->bd_flags & BD_F_SB16X) {
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
pprio = sb->pch.run? 1 : 0;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadmasetup(sb->pch.buffer, pprio? sb->drq1 : NULL);
sb->pch.dch = pprio? 1 : 0;
sndbuf_isadmasetup(sb->rch.buffer, pprio? sb->drq2 : sb->drq1);
sb->rch.dch = pprio? 2 : 1;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
} else {
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if (sb->pch.run && sb->rch.run) {
pprio = (sb->rch.fmt & AFMT_16BIT)? 0 : 1;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadmasetup(sb->pch.buffer, pprio? sb->drq2 : sb->drq1);
sb->pch.dch = pprio? 2 : 1;
sndbuf_isadmasetup(sb->rch.buffer, pprio? sb->drq1 : sb->drq2);
sb->rch.dch = pprio? 1 : 2;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
} else {
if (sb->pch.run) {
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadmasetup(sb->pch.buffer, (sb->pch.fmt & AFMT_16BIT)? sb->drq2 : sb->drq1);
sb->pch.dch = (sb->pch.fmt & AFMT_16BIT)? 2 : 1;
sndbuf_isadmasetup(sb->rch.buffer, (sb->pch.fmt & AFMT_16BIT)? sb->drq1 : sb->drq2);
sb->rch.dch = (sb->pch.fmt & AFMT_16BIT)? 1 : 2;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
} else if (sb->rch.run) {
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadmasetup(sb->pch.buffer, (sb->rch.fmt & AFMT_16BIT)? sb->drq1 : sb->drq2);
sb->pch.dch = (sb->rch.fmt & AFMT_16BIT)? 1 : 2;
sndbuf_isadmasetup(sb->rch.buffer, (sb->rch.fmt & AFMT_16BIT)? sb->drq2 : sb->drq1);
sb->rch.dch = (sb->rch.fmt & AFMT_16BIT)? 2 : 1;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
}
}
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadmasetdir(sb->pch.buffer, PCMDIR_PLAY);
sndbuf_isadmasetdir(sb->rch.buffer, PCMDIR_REC);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/*
adjust dma channels for vibra16x; recording should now work. full duplex does not work on vibra16x, so is disabled.
2000-11-07 00:38:59 +00:00
printf("setup: [pch = %d, pfmt = %d, pgo = %d] [rch = %d, rfmt = %d, rgo = %d]\n",
sb->pch.dch, sb->pch.fmt, sb->pch.run, sb->rch.dch, sb->rch.fmt, sb->rch.run);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
*/
ch = &sb->pch;
if (ch->run) {
l = ch->blksz;
if (ch->fmt & AFMT_16BIT)
l >>= 1;
l--;
/* play speed */
RANGE(ch->spd, 5000, 45000);
sb_cmd(sb, DSP_CMD_OUT16);
sb_cmd(sb, ch->spd >> 8);
sb_cmd(sb, ch->spd & 0xff);
/* play format, length */
v = DSP_F16_AUTO | DSP_F16_FIFO_ON | DSP_F16_DAC;
v |= (ch->fmt & AFMT_16BIT)? DSP_DMA16 : DSP_DMA8;
sb_cmd(sb, v);
v = (ch->fmt & AFMT_STEREO)? DSP_F16_STEREO : 0;
v |= (ch->fmt & AFMT_SIGNED)? DSP_F16_SIGNED : 0;
sb_cmd2(sb, v, l);
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadma(ch->buffer, PCMTRIG_START);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->bd_flags |= BD_F_DMARUN;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
}
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
ch = &sb->rch;
if (ch->run) {
l = ch->blksz;
if (ch->fmt & AFMT_16BIT)
l >>= 1;
l--;
/* record speed */
RANGE(ch->spd, 5000, 45000);
sb_cmd(sb, DSP_CMD_IN16);
sb_cmd(sb, ch->spd >> 8);
sb_cmd(sb, ch->spd & 0xff);
/* record format, length */
v = DSP_F16_AUTO | DSP_F16_FIFO_ON | DSP_F16_ADC;
v |= (ch->fmt & AFMT_16BIT)? DSP_DMA16 : DSP_DMA8;
sb_cmd(sb, v);
v = (ch->fmt & AFMT_STEREO)? DSP_F16_STEREO : 0;
v |= (ch->fmt & AFMT_SIGNED)? DSP_F16_SIGNED : 0;
sb_cmd2(sb, v, l);
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
sndbuf_isadma(ch->buffer, PCMTRIG_START);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->bd_flags |= BD_F_DMARUN2;
}
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb_unlock(sb);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return 0;
Upgrade Luigi's audio driver to Jan/23 version... will bring in Feb/15th version as soon as I've tested it... Installed/tested on my home machine...any problems, please report directly to me.
1998-02-17 14:17:13 +00:00
}
general tidyup
2000-01-10 03:22:28 +00:00
/* channel interface */
static void *
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb16chan_init(kobj_t obj, void *devinfo, struct snd_dbuf *b, struct pcm_channel *c, int dir)
general tidyup
2000-01-10 03:22:28 +00:00
{
struct sb_info *sb = devinfo;
struct sb_chinfo *ch = (dir == PCMDIR_PLAY)? &sb->pch : &sb->rch;
ch->parent = sb;
ch->channel = c;
ch->buffer = b;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
ch->dir = dir;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sndbuf_alloc(ch->buffer, sb->parent_dmat, SB16_BUFFSIZE) == -1)
general tidyup
2000-01-10 03:22:28 +00:00
return NULL;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return ch;
general tidyup
2000-01-10 03:22:28 +00:00
}
static int
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_setformat(kobj_t obj, void *data, u_int32_t format)
general tidyup
2000-01-10 03:22:28 +00:00
{
struct sb_chinfo *ch = data;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
struct sb_info *sb = ch->parent;
general tidyup
2000-01-10 03:22:28 +00:00
ch->fmt = format;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->prio = ch->dir;
sb->prio16 = (ch->fmt & AFMT_16BIT)? 1 : 0;
general tidyup
2000-01-10 03:22:28 +00:00
return 0;
}
static int
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_setspeed(kobj_t obj, void *data, u_int32_t speed)
general tidyup
2000-01-10 03:22:28 +00:00
{
struct sb_chinfo *ch = data;
ch->spd = speed;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return speed;
general tidyup
2000-01-10 03:22:28 +00:00
}
static int
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_setblocksize(kobj_t obj, void *data, u_int32_t blocksize)
general tidyup
2000-01-10 03:22:28 +00:00
{
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
struct sb_chinfo *ch = data;
ch->blksz = blocksize;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
return ch->blksz;
general tidyup
2000-01-10 03:22:28 +00:00
}
static int
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_trigger(kobj_t obj, void *data, int go)
general tidyup
2000-01-10 03:22:28 +00:00
{
struct sb_chinfo *ch = data;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
struct sb_info *sb = ch->parent;
general tidyup
2000-01-10 03:22:28 +00:00
handle emulated dma reads don't try to get sample size from snd_dbuf
2000-05-26 21:15:47 +00:00
if (go == PCMTRIG_EMLDMAWR || go == PCMTRIG_EMLDMARD)
general tidyup
2000-01-10 03:22:28 +00:00
return 0;
if (go == PCMTRIG_START)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
ch->run = 1;
general tidyup
2000-01-10 03:22:28 +00:00
else
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
ch->run = 0;
fix paste-o in mixer code - actually set right channel volume instead of doing the left channel twice.
2000-11-06 02:47:43 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb_setup(sb);
general tidyup
2000-01-10 03:22:28 +00:00
return 0;
}
static int
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_getptr(kobj_t obj, void *data)
general tidyup
2000-01-10 03:22:28 +00:00
{
struct sb_chinfo *ch = data;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
return sndbuf_isadmaptr(ch->buffer);
general tidyup
2000-01-10 03:22:28 +00:00
}
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
static struct pcmchan_caps *
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_getcaps(kobj_t obj, void *data)
general tidyup
2000-01-10 03:22:28 +00:00
{
struct sb_chinfo *ch = data;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
struct sb_info *sb = ch->parent;
general tidyup
2000-01-10 03:22:28 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if ((sb->prio == 0) || (sb->prio == ch->dir))
general tidyup
2000-01-10 03:22:28 +00:00
return &sb16x_caps;
else
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return sb->prio16? &sb16_caps8 : &sb16_caps16;
general tidyup
2000-01-10 03:22:28 +00:00
}
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
static int
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
sb16chan_resetdone(kobj_t obj, void *data)
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
{
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
struct sb_chinfo *ch = data;
struct sb_info *sb = ch->parent;
* Ignore the wierd fakechan workarounds for simplex in dsp_ioctl(). Without this, ioctl commands for setting formats and speeds were essentially ignored for simplex devices until the application actually performed a read or write. * Make sure that both channels are set in the SB mixer code and provide a mixer table specifically for the ess18xx which supports the extended accuracy available on this part. * Fix a stupid bug in ess_format() which ignored the passed-in format and changed the hardware based on the value which was set last time. This meant that the hardware setting was often not set correctly at all. * Add a custom identify driver for the ESS1888 which automagically detects and adds the device in a pseudo-PnP way. This driver also emits the magic sequence which enables the sound hardware after a hard reset, allowing it to work correctly for the sound hardware of a PWS 433au (and probably all other PWS class alpha machines). With these changes, I was able to play back simple sounds on my 433au. I have not tested recording or any other formats other than 8bit ulaw and 16bit stereo.
1999-11-13 18:31:31 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb->prio = 0;
* Ignore the wierd fakechan workarounds for simplex in dsp_ioctl(). Without this, ioctl commands for setting formats and speeds were essentially ignored for simplex devices until the application actually performed a read or write. * Make sure that both channels are set in the SB mixer code and provide a mixer table specifically for the ess18xx which supports the extended accuracy available on this part. * Fix a stupid bug in ess_format() which ignored the passed-in format and changed the hardware based on the value which was set last time. This meant that the hardware setting was often not set correctly at all. * Add a custom identify driver for the ESS1888 which automagically detects and adds the device in a pseudo-PnP way. This driver also emits the magic sequence which enables the sound hardware after a hard reset, allowing it to work correctly for the sound hardware of a PWS 433au (and probably all other PWS class alpha machines). With these changes, I was able to play back simple sounds on my 433au. I have not tested recording or any other formats other than 8bit ulaw and 16bit stereo.
1999-11-13 18:31:31 +00:00
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
return 0;
mss and sb drivers updated for newpcm Reviewed by: peter, dfr
1999-09-01 04:01:57 +00:00
}
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
static kobj_method_t sb16chan_methods[] = {
KOBJMETHOD(channel_init, sb16chan_init),
KOBJMETHOD(channel_resetdone, sb16chan_resetdone),
KOBJMETHOD(channel_setformat, sb16chan_setformat),
KOBJMETHOD(channel_setspeed, sb16chan_setspeed),
KOBJMETHOD(channel_setblocksize, sb16chan_setblocksize),
KOBJMETHOD(channel_trigger, sb16chan_trigger),
KOBJMETHOD(channel_getptr, sb16chan_getptr),
KOBJMETHOD(channel_getcaps, sb16chan_getcaps),
{ 0, 0 }
};
CHANNEL_DECLARE(sb16chan);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
/************************************************************/
Import of Luigi Rizzo's sound code. For more information about the driver check out the README that is included. Submitted by: Luigi Rizzo <luigi@labinfo.iet.unipi.it>
1997-09-14 21:42:12 +00:00
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
static int
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb16_probe(device_t dev)
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
{
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
char buf[64];
Don't pass u_int32_t pointers to BUS_READ_IVAR since it tends to make alphas panic.
1999-12-24 16:22:45 +00:00
uintptr_t func, ver, r, f;
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
/* The parent device has already been probed. */
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
r = BUS_READ_IVAR(device_get_parent(dev), dev, 0, &func);
if (func != SCF_PCM)
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
return (ENXIO);
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
r = BUS_READ_IVAR(device_get_parent(dev), dev, 1, &ver);
make the vibra16x work for playback at least, record is untested
1999-12-18 22:21:47 +00:00
f = (ver & 0xffff0000) >> 16;
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
ver &= 0x0000ffff;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if (f & BD_F_SB16) {
snprintf(buf, sizeof buf, "SB16 DSP %d.%02d%s", (int) ver >> 8, (int) ver & 0xff,
(f & BD_F_SB16X)? " (ViBRA16X)" : "");
device_set_desc_copy(dev, buf);
return 0;
} else
split up ess and sb code rewrite ess mixer to use native registers rewrite play/rec code to use more accurate timer when available add code to use audio2 for playback, but disable it as no irqs are generated
2000-03-28 18:31:01 +00:00
return (ENXIO);
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
}
static int
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb16_attach(device_t dev)
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
{
struct sb_info *sb;
Don't pass u_int32_t pointers to BUS_READ_IVAR since it tends to make alphas panic.
1999-12-24 16:22:45 +00:00
uintptr_t ver;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
char status[SND_STATUSLEN];
int bs = SB16_BUFFSIZE;
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
Use the M_ZERO flag to malloc(9) Reviewed by: cg MFC after: 1 week
2001-06-21 19:45:59 +00:00
sb = (struct sb_info *)malloc(sizeof *sb, M_DEVBUF, M_NOWAIT | M_ZERO);
general tidyup
2000-01-10 03:22:28 +00:00
if (!sb)
return ENXIO;
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sb->parent_dev = device_get_parent(dev);
BUS_READ_IVAR(sb->parent_dev, dev, 1, &ver);
make sb dependant on sbc add support for non-pnp cards to sbc move card identification to sbc channel-swapping code is in sb now instead of dsp vibra16x support is still broken, but will be fixed soon note: sbc is now compulsory for sb cards for pnp cards use: device sbc0 for non-pnp cards eg: device sbc0 at isa? port 0x240 irq 5 drq 3 flags 0x15 (hints as oldpcm) both in addition to: device pcm0 Reviewed by: tanimura,dfr Said he liked it: peter
1999-12-12 02:30:19 +00:00
sb->bd_id = ver & 0x0000ffff;
sb->bd_flags = (ver & 0xffff0000) >> 16;
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
if (sb16_alloc_resources(sb, dev))
goto no;
if (sb_reset_dsp(sb))
goto no;
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
if (mixer_init(dev, &sb16mix_mixer_class, sb))
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
goto no;
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
if (snd_setup_intr(dev, sb->irq, INTR_MPSAFE, sb_intr, sb, &sb->ih))
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
goto no;
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (sb->bd_flags & BD_F_SB16X)
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
pcm_setflags(dev, pcm_getflags(dev) | SD_F_SIMPLEX);
sb->prio = 0;
if (bus_dma_tag_create(/*parent*/NULL, /*alignment*/2, /*boundary*/0,
/*lowaddr*/BUS_SPACE_MAXADDR_24BIT,
/*highaddr*/BUS_SPACE_MAXADDR,
/*filter*/NULL, /*filterarg*/NULL,
/*maxsize*/bs, /*nsegments*/1,
/*maxsegz*/0x3ffff,
/*flags*/0, &sb->parent_dmat) != 0) {
device_printf(dev, "unable to create dma tag\n");
goto no;
}
snprintf(status, SND_STATUSLEN, "at io 0x%lx irq %ld drq %ld",
rman_get_start(sb->io_base), rman_get_start(sb->irq),
rman_get_start(sb->drq1));
update code dealing with snd_dbuf objects to do so using a functional interface modify chn_setblocksize() to pick a default soft-blocksize appropriate to the sample rate and format in use. it will aim for a power of two size small enough to generate block sizes of at most 20ms. it will also set the hard-blocksize taking into account rate/format conversions in use. update drivers to implement setblocksize correctly: updated, tested: sb16, emu10k1, maestro, solo updated, untested: ad1816, ess, mss, sb8, csa not updated: ds1, es137x, fm801, neomagic, t4dwave, via82c686 i lack hardware to test: ad1816, csa, fm801, neomagic others will be updated/tested in the next few days.
2000-12-23 03:16:13 +00:00
if (!(pcm_getflags(dev) & SD_F_SIMPLEX))
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
snprintf(status + strlen(status), SND_STATUSLEN - strlen(status),
":%ld", rman_get_start(sb->drq2));
if (pcm_register(dev, sb, 1, 1))
goto no;
kobjify. this gives us several benefits, including: * easier extensibility- new optional methods can be added to ac97/mixer/channel classes without having to fixup every driver. * forward compatibility for drivers, provided no new mandatory methods are added.
2000-12-18 01:36:41 +00:00
pcm_addchan(dev, PCMDIR_REC, &sb16chan_class, sb);
pcm_addchan(dev, PCMDIR_PLAY, &sb16chan_class, sb);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
pcm_setstatus(dev, status);
return 0;
no:
sb16_release_resources(sb, dev);
return ENXIO;
}
static int
sb16_detach(device_t dev)
{
int r;
struct sb_info *sb;
r = pcm_unregister(dev);
if (r)
return r;
sb = pcm_getdevinfo(dev);
sb16_release_resources(sb, dev);
return 0;
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
}
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
static void
sb_lock(struct sb_info *sb) {
sbc_lock(device_get_softc(sb->parent_dev));
}
static void
sb_unlock(struct sb_info *sb) {
sbc_unlock(device_get_softc(sb->parent_dev));
}
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
static device_method_t sb16_methods[] = {
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
/* Device interface */
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
DEVMETHOD(device_probe, sb16_probe),
DEVMETHOD(device_attach, sb16_attach),
DEVMETHOD(device_detach, sb16_detach),
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
{ 0, 0 }
};
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
static driver_t sb16_driver = {
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
"pcm",
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
sb16_methods,
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
sizeof(struct snddev_info),
- Introduce the bridge drivers for Sound Blaser, GUS and Crystal Semiconductor CS461x/428x. - Add support for GUS and CS461x/428x pcm. Bridges reviewed by: dfr, cg GUS non-PnP support submitted by: Ville-Pertti Keinonen <will@iki.fi> GUS PnP support tested by: Michiru Saito <mich@mtci.ne.jp>
1999-11-22 06:07:49 +00:00
};
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
DRIVER_MODULE(snd_sb16, sbc, sb16_driver, pcm_devclass, 0, 0);
MODULE_DEPEND(snd_sb16, snd_pcm, PCM_MINVER, PCM_PREFVER, PCM_MAXVER);
mega-commit. this introduces a new buffering mechanism which results in dramatic simplification of the channel manager. as several structures have changed, we take the opportunity to move their definitions into the source files where they are used, make them private and de-typedef them. the sound drivers are updated to use snd_setup_intr instead of bus_setup_intr, and to comply with the de-typedefed structures. the ac97, mixer and channel layers have been updated with finegrained locking, as have some drivers- not all though. the rest will follow soon.
2001-03-24 23:10:29 +00:00
MODULE_DEPEND(snd_sb16, snd_sbc, 1, 1, 1);
split up sb16 and sb/sbpro drivers we do not support sb versions <2.00, and the sb8 driver has not been tested yet. these drivers are not yet enabled by default.
2000-10-28 19:20:03 +00:00
MODULE_VERSION(snd_sb16, 1);
Reference in New Issue Copy Permalink