Thread: [Alsa-user] Cyrrus Logic Crystal CS4237B to configure
Brought to you by:
perex
Menu
▾
▴
alsa-user
|
From: Hulin T. <hul...@wa...> - 2006-05-23 22:03:01
|
Hello alsa users ! I rode the documentation of Alsa to configure my Cyrrus Logic Crystal=20 CS4237B (it's an ISA PnP soundcard) and I was install the last version=20 of alsa by this link http://www.tldp.org/HOWTO/Alsa-sound-7.html . However, my kernel Linux 2.6.12-10-k7 can't recognize my soundcard.=20 Indeed, when I type "setpnp 0f on", I obtain "lspnp: /proc/bus/pnp not=20 available", however I installed pnpbios-tools_3.2.8-8_i386. Modules of alsa are loaded, and modules of OSS are stoped. Cursiously, with OSS, it works. I'm using Ubuntu on a dell latitude CP MMX 233 ST. Same result with the=20 last GeexBox too. So, alsaconf had found my soundcard on mandrake 9.1,=20 but not in my last configuration. A "sudo modprobe snd-cs4232" produce : FATAL: Error inserting snd_cs4232=20 (/lib/modules/2.6.12-10-386/kernel/sound/isa/cs423x/snd-cs4232.ko): No=20 such device FATAL: Error running install command for snd_cs4232 When I launch xmms from a terminal, I have these messages : With OSS : ** WARNING ** oss_open(): Failed to open audio device (/dev/dsp): Aucun=20 p=E9riph=E9rique de ce type modprobe: FATAL Error running install command for sound_slot_0 With Alsa : ** WARNING **: ALSA_GET_MIXER(): Attaching to mixer hw:0 failed: Aucun=20 p=E9riph=E9rique de ce type ALSA lib pcm_dmix.c:802:(snd_pcm_dmix_open) unable to open slave ** WARNING **: alsa_setup(): Failed to open pcm device (default): Aucun=20 p=E9riph=E9rique de ce type What so strange, isn't it ? I post this message in Ubuntu Forum and I=20 searched an answer with Google (our big friend), but noone has worked.=20 Now, I don't know what doing ! I've seen that I don't alone with this problem. Can you help me ? An idea to continue ? Thanks you very much ! Regards, Thibaud Hulin. |
|
From: Rene H. <ren...@ke...> - 2006-05-24 01:05:03
|
Hulin Thibaud wrote: > I rode the documentation of Alsa Yeeehaaa! > to configure my Cyrrus Logic Crystal CS4237B (it's an ISA PnP > soundcard) and I was install the last version of alsa by this link > http://www.tldp.org/HOWTO/Alsa-sound-7.html . That's really wholly obsolete, I'm afraid. Ignore it. > However, my kernel Linux 2.6.12-10-k7 can't recognize my soundcard. > Indeed, when I type "setpnp 0f on", I obtain "lspnp: /proc/bus/pnp > not available", however I installed pnpbios-tools_3.2.8-8_i386. You shouldn't use any external PnP tools on 2.6; "modprobe snd-cs4236" should be all you need assuming you have ISA-PnP support enabled in your kernel. If this doesn't work (check dmesg, and start alsamixer after this to unmute channels and increase volume) please verify whether or not you have ISA-PnP support in the kernel, for example by seeing if the card shows up under /sys/devices/pnp. Example for a CS4236 card: $ cat /sys/devices/pnp1/01\:01/card_id CSCa836 Please do not post in HTML by the way. And certainly not in HTML with a black background... Rene. |
|
From: Rene H. <ren...@ke...> - 2006-05-25 14:55:54
|
Hulin Thibaud wrote: > I followed your councils. About the post in text, I believe... that's > work ! It does. It is also very much preferred if you keep the list you posted on in CC (using "reply to all" would've done that). That way, others can look in on the conversation, correct given advice, give more advice, learn from advice as well, and so on. > (/lib/modules/2.6.12-10-386/kernel/sound/isa/cs423x/snd-cs4236.ko): No > such device Okay... > I suppose that I have the IsaPNP support (how can I be sure ?), because > I have the folder /sys/devices/pnp and pnp1. > But I don't have the file card_id (searched with 'locate'). "locate" is not useful here. It uses a database, but /sys is a virtual filesystem (meaning the files don't really "exist" on disk -- it's just an illusion made up by the kernel, same as /proc). Anyways, you can just browse around in /sys to see what's there. From the console for example with "mc /sys", or using whatever graphical file manager you use. > I have only links bus and driver, and the directory power, and the files > id, options and resources. Okay, that means you do have a device there it seems, as the bus and driver links only exist in device directories. That is, in a directory such as: $ tree /sys/devices/pnp1/01:01/01:01.00 /sys/devices/pnp1/01:01/01:01.00 |-- bus -> ../../../../bus/pnp |-- driver -> ../../../../bus/pnp/drivers/cs4236_isapnp |-- id |-- options |-- power |-- resources `-- uevent Please show me the output of that same "tree" command (for the directory where your "bus" and "driver" and id file was). Also, what does that "id" file say in that directory? For me: $ cat /sys/devices/pnp1/01:01/01:01.00/id CSCa800 The card_id file I said should be one directory up. Is that CS4237 card by the way the only ISA-PnP card you have in the machine? > So, the command pnpdump conclue "No boards found"... pnpdump is another tool which you should not be using anymore on 2.6. It should still work though, and it bypasses the kernel. If it isn't finding anything, that's very strange seeing as how apparently the kernel code _did_ find something. Please also just make sure the board(s) is/are inserted properly. Rene. |
|
From: Rene H. <ren...@ke...> - 2006-05-25 15:29:23
|
Rene Herman wrote: >> I suppose that I have the IsaPNP support (how can I be sure ?), >> because I have the folder /sys/devices/pnp and pnp1. By the way, also please try a "zgrep ISAPNP /proc/config.gz". You should see a CONFIG_ISAPNP=y. If the /proc/config.gz file does not exist, go shoot your distribution's kernel maintainer and try "grep ISA_PNP /boot/config-$(uname -r)" instead. For all I know, your pnp1 is PnP-BIOS, with pnp0 being PNPACPI or something... Rene. |
|
From: Hulin T. <hul...@wa...> - 2006-05-26 08:42:59
|
Rene Herman a =E9crit :
> Hulin Thibaud wrote:
>=20
>> I followed your councils. About the post in text, I believe... that's
>> work !
>=20
> It does. It is also very much preferred if you keep the list you posted=
=20
> on in CC (using "reply to all" would've done that). That way, others ca=
n=20
> look in on the conversation, correct given advice, give more advice,=20
> learn from advice as well, and so on.
>=20
Great council ! Just this : I receive two messages, on from Rene Herman,=20
one other from Alsa user...
>> (/lib/modules/2.6.12-10-386/kernel/sound/isa/cs423x/snd-cs4236.ko): No=
=20
>> such device
>=20
> Okay...
>=20
>> I suppose that I have the IsaPNP support (how can I be sure ?),=20
>> because I have the folder /sys/devices/pnp and pnp1.
>> But I don't have the file card_id (searched with 'locate').
>=20
> "locate" is not useful here. It uses a database, but /sys is a virtual=
=20
> filesystem (meaning the files don't really "exist" on disk -- it's just=
=20
> an illusion made up by the kernel, same as /proc). Anyways, you can jus=
t=20
> browse around in /sys to see what's there. From the console for example=
=20
> with "mc /sys", or using whatever graphical file manager you use.
>=20
>> I have only links bus and driver, and the directory power, and the=20
>> files id, options and resources.
>=20
> Okay, that means you do have a device there it seems, as the bus and=20
> driver links only exist in device directories. That is, in a directory=20
> such as:
>=20
> $ tree /sys/devices/pnp1/01:01/01:01.00
> /sys/devices/pnp1/01:01/01:01.00
> |-- bus -> ../../../../bus/pnp
> |-- driver -> ../../../../bus/pnp/drivers/cs4236_isapnp
> |-- id
> |-- options
> |-- power
> |-- resources
> `-- uevent
>=20
> Please show me the output of that same "tree" command (for the director=
y=20
> where your "bus" and "driver" and id file was). Also, what does that=20
> "id" file say in that directory? For me:
>=20
> $ cat /sys/devices/pnp1/01:01/01:01.00/id
> CSCa800
In the folder pnp1, there is only on file, /sys/devices/pnp1/power/state
So, after installing tree with an apt-get install, I obtain with tree=20
/sys/devices/pnp0 :
/sys/devices/pnp0/
|-- 00:00
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:01
| |-- bus -> ../../../bus/pnp
| |-- driver -> ../../../bus/pnp/drivers/system
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:02
| |-- bus -> ../../../bus/pnp
| |-- driver -> ../../../bus/pnp/drivers/system
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:03
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:04
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:05
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:06
| |-- bus -> ../../../bus/pnp
| |-- driver -> ../../../bus/pnp/drivers/i8042 kbd
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:07
| |-- bus -> ../../../bus/pnp
| |-- driver -> ../../../bus/pnp/drivers/i8042 aux
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:08
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:09
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:0a
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:0b
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:0c
| |-- bus -> ../../../bus/pnp
| |-- driver -> ../../../bus/pnp/drivers/serial
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:0e
| |-- bus -> ../../../bus/pnp
| |-- driver -> ../../../bus/pnp/drivers/parport_pc
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:0f
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:10
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
|-- 00:11
| |-- bus -> ../../../bus/pnp
| |-- id
| |-- options
| |-- power
| | `-- state
| `-- resources
`-- power
`-- state
58 directories, 69 files
> The card_id file I said should be one directory up. Is that CS4237 card=
=20
> by the way the only ISA-PnP card you have in the machine?
I'm not perfectly sure. From the documentation of my computer, there not=20
an other device indicated with the bus ISA as I can rode here :
http://docs.us.dell.com/support/edocs/systems/pmojav/54723.pdf
With a lspci, I obtain :
0000:00:00.0 Host bridge: Intel Corp. 430TX - 82439TX MTXC (rev 01)
0000:00:01.0 Bridge: Intel Corp. 82371AB/EB/MB PIIX4 ISA (rev 01)
0000:00:01.1 IDE interface: Intel Corp. 82371AB/EB/MB PIIX4 IDE (rev 01)
0000:00:01.2 USB Controller: Intel Corp. 82371AB/EB/MB PIIX4 USB (rev 01)
0000:00:01.3 Bridge: Intel Corp. 82371AB/EB/MB PIIX4 ACPI (rev 01)
0000:00:02.0 VGA compatible controller: Neomagic Corporation NM2160=20
[MagicGraph 128XD] (rev 01)
0000:00:03.0 CardBus bridge: Texas Instruments PCI1131 (rev 01)
0000:00:03.1 CardBus bridge: Texas Instruments PCI1131 (rev 01)
It show an other ISA with the bridge d'Intel ?
And for the cat command, this is the results after every 00:xx files on=20
one or two lines :
00:00
PNP0a03
00:02
PNP0c0a
00:04
PNP0100
00:06
PNP0303
00:08
PNP0800
00:0a
PNP0c04
00:0c
PNP0501
00:0f
CSC0000
00:11
PNP0700
00:01
PNP0c01
00:03
PNP0000
00:05
PNP0200
00:07
PNP0f13
00:09
PNP0b00
00:0b
PNP0e03
PNP0e00
00:0e
PNP0401
00:10
CSC0010
So, the last is the CS4237B sound card isa port, isn't it ?!...
>> So, the command pnpdump conclue "No boards found"...
>
> pnpdump is another tool which you should not be using anymore on 2.6. =
It
> should still work though, and it bypasses the kernel. If it isn't
> finding anything, that's very strange seeing as how apparently the
> kernel code _did_ find something. Please also just make sure the
> board(s) is/are inserted properly.
>
> Rene.
>
What is a board ? Must I open my computer to look that ?
>Rene Herman wrote:
>> I suppose that I have the IsaPNP support (how can I be sure ?),=20
because I have the folder /sys/devices/pnp and pnp1.
>By the way, also please try a "zgrep ISAPNP /proc/config.gz". You=20
should see a CONFIG_ISAPNP=3Dy. If the /proc/config.gz file does not=20
exist, go shoot your distribution's kernel maintainer and try "grep=20
ISA_PNP >/boot/config-$(uname -r)" instead.
>For all I know, your pnp1 is PnP-BIOS, with pnp0 being PNPACPI or=20
something...
>Rene.
"zgrep ISAPNP /proc/config.gz" don't work, but "cat=20
/boot/config-2.6.12-10-386 | grep ISAPNP" give a CONFIG_ISAPNP=3Dy
I think you're a great professionnal !
Thibaud.
|
|
From: Rene H. <ren...@ke...> - 2006-05-26 17:03:55
|
Hulin Thibaud wrote: > Just this : I receive two messages, on from Rene Herman, one other > from Alsa user... Yes, this is expected. I used "reply to all", meaning I replied both directly to your email address and to the list address. Since you are also subscribed to the list, you then got that second one delivered to your address as well. Usually people have filters in use that group messages that come in via mailinglists into seperate mail folders, leaving just messages directly to their email address in their inbox. Most people on these linux lists are subscribed to many mailinglists such as alsa-user and for example I personally receive probably more than 1000 messages per day from all of them. Obviously, I do not pay extremely close attention to all of them meaning that I certainly want that duplicate personal copy in my inbox for messages from a conversation I participate in. Otherwise I might overlook them, or see them only after a long time. Some other people will complain about the "duplicates", but it's safe to ignore them... >> The card_id file I said should be one directory up. Is that CS4237 >> card by the way the only ISA-PnP card you have in the machine? > > I'm not perfectly sure. From the documentation of my computer, there not > an other device indicated with the bus ISA as I can rode here : > http://docs.us.dell.com/support/edocs/systems/pmojav/54723.pdf Oh heh, it's a laptop, with the CS4237 onboard. I was expecting a "normal machine", with a discrete ISA-PnP soundboard in one of its slots. > 0000:00:01.0 Bridge: Intel Corp. 82371AB/EB/MB PIIX4 ISA (rev 01) [ ... ] > It show an other ISA with the bridge d'Intel ? Yes, that's normal. This is not important to the issue, but on a PCI PC, the ISA bus (an older expansion bus) hangs off PCI and needs this bridge to forward accesses from the PCI bus. > And for the cat command, this is the results after every 00:xx files on > one or two lines : > 00:0f > CSC0000 That's the CS423x codec > 00:10 > CSC0010 And that's the CTRL port. > What is a board ? Must I open my computer to look that ? Nope, please don't open the laptop... :-) I was assuming a regular computer. > "zgrep ISAPNP /proc/config.gz" don't work, but "cat > /boot/config-2.6.12-10-386 | grep ISAPNP" give a CONFIG_ISAPNP=y Okay, that's fine. I wasn't aware though that your CS4237 was onboard and more importantly, neither was snd-cs4236. It appears to not have support for onboard CS4236+ chips -- the CSC0000 device is driven only by snd-cs4232. Does "modprobe snd-cs4232" work? If not, does it fail with just the same "no devices" message as snd-cs4236 did, or does something more interesting appear at the end of "dmesg" (typing "dmesg" displays the kernel message buffer)? I expect it will work though. It it does, please start "alsamixer" and unmute the Master and PCM channels and crank up the volume a bit before you try playing sound. "Master" is the first slider -- to unmute it, press "M" and then press the up-arrow until the slider is at level 89 (that's the setting where the chip sounds best, 0 dB). PCM is right next to it. Also unmute it, and set that one to 87. > I think you're a great professionnal ! Hey! I resent being called a professional! Rene. |
|
From: Rene H. <ren...@ke...> - 2006-05-26 17:38:21
|
Rene Herman wrote: > Okay, that's fine. I wasn't aware though that your CS4237 was onboard > and more importantly, neither was snd-cs4236. It appears to not have > support for onboard CS4236+ chips -- the CSC0000 device is driven only > by snd-cs4232. Does "modprobe snd-cs4232" work? If not, does it fail > with just the same "no devices" message as snd-cs4236 did, or does > something more interesting appear at the end of "dmesg" (typing "dmesg" > displays the kernel message buffer)? Crap, I see in your original message that you already tried snd-cs4232 as well. I'm not suite sure why that wouldn't work though, so please retry and check that "dmesg" if failed. One more thing to try; let's take PnP out of the picture. Does it work to do: modprobe snd-cs4236 isapnp=0 port=0x534 cport=0x538 irq=5 dma1=1 dma2=0 If not, that same line with snd-cs4232 instead? If both do not work, please try upgrading your kernel (how to do that is dependent on the distribution -- you can probably update through some update tool). I see it's 2.6.12 and that's fairly old. I'm looking at 2.6.16 kernels here. And a last resort for now: modprobe snd-sb8 port=0x220 irq=5 dma8=1 That will probably work, but we should get the crystal driver working instead. Rene. |
|
From: Hulin T. <hul...@wa...> - 2006-06-07 08:46:54
|
Hello Rene ! Sorry for my silence, but I was updating my computer as you asked me. I have now the Ubutu version Dapper 6.06 (related 1st june 2006), with=20 the kernel Linux 2.6.15-23-386. I have the same errors with my old kernel 2.6.12-10 : pnpdump and=20 alsaconf don't fall the soundcard. Rene Herman a =E9crit : > Rene Herman wrote: >=20 >> Okay, that's fine. I wasn't aware though that your CS4237 was onboard=20 >> and more importantly, neither was snd-cs4236. It appears to not have=20 >> support for onboard CS4236+ chips -- the CSC0000 device is driven only= =20 >> by snd-cs4232. Does "modprobe snd-cs4232" work? If not, does it fail=20 >> with just the same "no devices" message as snd-cs4236 did, or does=20 >> something more interesting appear at the end of "dmesg" (typing "dmesg= "=20 >> displays the kernel message buffer)? For the cs4236, I have this message : CS4236+ soundcard not found or device busy FATAL: Error inserting snd_cs4236=20 (/lib/modules/2.6.12-10-386/kernel/sound/isa/cs423x/snd-cs4236.ko): No=20 such device Near the same message with the cs4232. Dmesg has recorded the first line of the message error. >=20 > Crap, I see in your original message that you already tried snd-cs4232=20 > as well. I'm not suite sure why that wouldn't work though, so please=20 > retry and check that "dmesg" if failed. >=20 > One more thing to try; let's take PnP out of the picture. Does it work=20 > to do: >=20 > modprobe snd-cs4236 isapnp=3D0 port=3D0x534 cport=3D0x538 irq=3D5 dma1=3D= 1 dma2=3D0 Idem. >=20 > If not, that same line with snd-cs4232 instead? Idem. If both do not work, > please try upgrading your kernel (how to do that is dependent on the=20 > distribution -- you can probably update through some update tool). I se= e=20 > it's 2.6.12 and that's fairly old. I'm looking at 2.6.16 kernels here. >=20 OK, I have the 2.6.15 now ! > And a last resort for now: >=20 > modprobe snd-sb8 port=3D0x220 irq=3D5 dma8=3D1 Hey ! No error ! With xmms, I have music ! But sound is very bad... How can I improve that ? Thanks you very much, Thibaud. |
|
From: Rene H. <ren...@ke...> - 2006-06-07 22:28:29
|
Hulin Thibaud wrote: > I have now the Ubutu version Dapper 6.06 (related 1st june 2006), > with the kernel Linux 2.6.15-23-386. I have the same errors with my > old kernel 2.6.12-10 : pnpdump and alsaconf don't fall the soundcard. Well, it's rather amazing, seeing as how your computer (a Pentium 1 laptop) isn't exactly new but I'm looking at 2.6.15 here now and it indeed also has no (PnP-BIOS) support for your onboard CS4236. That support is new to 2.6.16. Look at it this way -- your computer is still so advanced it needs the very latest Linux kernel to function completely. Great isn't it? I asked around, and it appears that Ubuntu does not yet have a 2.6.16 kernel available either. Them releasing with 2.6.15 is in fact a bit strange but nothing I can do about that. I'm really quite reluctant to make you compile a new kernel (or ALSA) yourself though. The added bit in 2.6.16 is just the "works magically" PnP-BIOS support so let's try if we can just enable the thing manually (if needed) and then tell the driver to not use pnp. Your Crystal PNPBIOS devices were 00:0f and 00:10. In the directories: /sys/devices/pnp0/00:0f and /sys/devices/pnp0/00:10 there should be files named "resources". If you do a: cat /sys/devices/pnp0/00:0f/resources does it say "state = disabled" and nothing more? If so, try a: echo auto >/sys/devices/pnp0/00:0f/resources It should now look something like (that cat again): state = disabled io 0x534-0x537 io 0x388-0x38b io 0x220-0x22f irq 5 dma 1 dma 0 Then, do a: echo activate >/sys/devices/pnp0/00:0f/resources after which that "state = disabled" should have been changed to "state = active". Ie, you should now have something like state = active io 0x534-0x537 *1 io 0x388-0x38b *2 io 0x220-0x22f *3 irq 5 *4 dma 1 *5 dma 0 *6 If it already looked like this before you echoed anything into the resource file, all the better! For 00:10 (the CTRL port) you should, possibly again after echoing that "auto" and "activate" into the resource file, you should have: cat /sys/devices/pnp0/00:10/resources state = active io 0x538-0x53f *7 All still well? If so, we can try loading the driver. I already made you try this with the values as given here, so I'm hoping your values are different and it will work this time. Try: modprobe snd-cs4236 isapnp=0 port=*1 cport=*7 irq=*4 dma1=*5 dma2=*6 substituting the *N items with the values from the resources files as numbered. Did it work? Please? Aaaaah? And with snd-cs4232 and those same parameters instead? If it did work, you'd need to do that same echoing into that resource file that you did on every boot -- you could do this from your "rc.local" startup file if needed. If you didn't need any echoing (ie, your BIOS already actived everything, as is likely) you can just always load the driver with those parameters. To just need a "modprobe snd-cs4236", you can say in /etc/modprobe.d/alsa-base (I believe that's where ubuntu puts it but I'm not an ubuntu or debian user) options snd-cs4236 isapnp=0 port=*1 cport=*7 irq=*4 dma1=*5 dma2=*6 >> And a last resort for now: >> >> modprobe snd-sb8 port=0x220 irq=5 dma8=1 > > Hey ! No error ! > With xmms, I have music ! > But sound is very bad... How can I improve that ? Well, you could play a bit with volumes in ALSA mixer to see if it improves but this is the wrong driver anyway. It's just that the Crystal chip _has_ a soundblaster compatible part so it can be used, but the chrystal driver should be preferred. It is good to know that it does work since that will mean that at least your BIOS has enabled that part of the chip. I'm hoping that with the above things will work. I can imagine it all looks a bit clumsy if you're a new Linux user but due to some weird reason the PnP-BIOS support for your chip was only added very recently and due to some other weird reason Ubuntu is not using 2.6.16 but 2.6.15. As indicated, on 2.6.16 "modprobe snd-cs4232" should be all you need, and alsaconf should be able to figure that all out as well. Rene. |
|
From: EmIScA <em...@em...> - 2006-06-08 18:22:45
|
Take the debian unstable linux-source-2.6.16 and alsa 1.0.11 source packages and compile them (for the kernel you can use kernel-package, you can search google for an howto about recompile a kernel in debian/ubuntu and package compilation from debian sources). I tried to do that, and works. Be careful to not use binary packages from Debian, you will probably break something, or they just don't work. Tell me if you want more help.... :-D Rene Herman ha scritto: > Hulin Thibaud wrote: > > >> I have now the Ubutu version Dapper 6.06 (related 1st june 2006), >> with the kernel Linux 2.6.15-23-386. I have the same errors with my >> old kernel 2.6.12-10 : pnpdump and alsaconf don't fall the soundcard. >> > > Well, it's rather amazing, seeing as how your computer (a Pentium 1 > laptop) isn't exactly new but I'm looking at 2.6.15 here now and it > indeed also has no (PnP-BIOS) support for your onboard CS4236. That > support is new to 2.6.16. > > Look at it this way -- your computer is still so advanced it needs the > very latest Linux kernel to function completely. Great isn't it? > > I asked around, and it appears that Ubuntu does not yet have a 2.6.16 > kernel available either. Them releasing with 2.6.15 is in fact a bit > strange but nothing I can do about that. I'm really quite reluctant to > make you compile a new kernel (or ALSA) yourself though. The added bit > in 2.6.16 is just the "works magically" PnP-BIOS support so let's try if > we can just enable the thing manually (if needed) and then tell the > driver to not use pnp. > > Your Crystal PNPBIOS devices were 00:0f and 00:10. In the directories: > > /sys/devices/pnp0/00:0f > > and > > /sys/devices/pnp0/00:10 > > there should be files named "resources". If you do a: > > cat /sys/devices/pnp0/00:0f/resources > > does it say "state = disabled" and nothing more? If so, try a: > > echo auto >/sys/devices/pnp0/00:0f/resources > > It should now look something like (that cat again): > > state = disabled > io 0x534-0x537 > io 0x388-0x38b > io 0x220-0x22f > irq 5 > dma 1 > dma 0 > > Then, do a: > > echo activate >/sys/devices/pnp0/00:0f/resources > > after which that "state = disabled" should have been changed to "state = > active". Ie, you should now have something like > > state = active > io 0x534-0x537 *1 > io 0x388-0x38b *2 > io 0x220-0x22f *3 > irq 5 *4 > dma 1 *5 > dma 0 *6 > > If it already looked like this before you echoed anything into the > resource file, all the better! > > For 00:10 (the CTRL port) you should, possibly again after echoing that > "auto" and "activate" into the resource file, you should have: > > cat /sys/devices/pnp0/00:10/resources > state = active > io 0x538-0x53f *7 > > All still well? If so, we can try loading the driver. I already made you > try this with the values as given here, so I'm hoping your values are > different and it will work this time. Try: > > modprobe snd-cs4236 isapnp=0 port=*1 cport=*7 irq=*4 dma1=*5 dma2=*6 > > substituting the *N items with the values from the resources files as > numbered. > > Did it work? Please? Aaaaah? And with snd-cs4232 and those same > parameters instead? > > If it did work, you'd need to do that same echoing into that resource > file that you did on every boot -- you could do this from your > "rc.local" startup file if needed. If you didn't need any echoing (ie, > your BIOS already actived everything, as is likely) you can just always > load the driver with those parameters. To just need a "modprobe > snd-cs4236", you can say in /etc/modprobe.d/alsa-base (I believe that's > where ubuntu puts it but I'm not an ubuntu or debian user) > > options snd-cs4236 isapnp=0 port=*1 cport=*7 irq=*4 dma1=*5 dma2=*6 > > >>> And a last resort for now: >>> >>> modprobe snd-sb8 port=0x220 irq=5 dma8=1 >>> >> Hey ! No error ! >> With xmms, I have music ! >> But sound is very bad... How can I improve that ? >> > > Well, you could play a bit with volumes in ALSA mixer to see if it > improves but this is the wrong driver anyway. It's just that the Crystal > chip _has_ a soundblaster compatible part so it can be used, but the > chrystal driver should be preferred. It is good to know that it does > work since that will mean that at least your BIOS has enabled that part > of the chip. > > I'm hoping that with the above things will work. I can imagine it all > looks a bit clumsy if you're a new Linux user but due to some weird > reason the PnP-BIOS support for your chip was only added very recently > and due to some other weird reason Ubuntu is not using 2.6.16 but > 2.6.15. As indicated, on 2.6.16 "modprobe snd-cs4232" should be all you > need, and alsaconf should be able to figure that all out as well. > > Rene. > > > _______________________________________________ > Alsa-user mailing list > Als...@li... > https://lists.sourceforge.net/lists/listinfo/alsa-user > > |
|
From: Hulin T. <hul...@wa...> - 2006-06-11 13:18:19
|
Rene Herman a =E9crit : > Hulin Thibaud wrote: >=20 >> I have now the Ubutu version Dapper 6.06 (related 1st june 2006), >> with the kernel Linux 2.6.15-23-386. I have the same errors with my >> old kernel 2.6.12-10 : pnpdump and alsaconf don't fall the soundcard. >=20 > Well, it's rather amazing, seeing as how your computer (a Pentium 1=20 > laptop) isn't exactly new but I'm looking at 2.6.15 here now and it=20 > indeed also has no (PnP-BIOS) support for your onboard CS4236. That=20 > support is new to 2.6.16. It's a pentium 1 MMX, 233 Mhz... > Look at it this way -- your computer is still so advanced it needs the=20 > very latest Linux kernel to function completely. Great isn't it? Of course, that's great ! > I asked around, and it appears that Ubuntu does not yet have a 2.6.16=20 > kernel available either. Them releasing with 2.6.15 is in fact a bit=20 > strange but nothing I can do about that. I'm really quite reluctant to=20 > make you compile a new kernel (or ALSA) yourself though. The added bit=20 > in 2.6.16 is just the "works magically" PnP-BIOS support so let's try i= f=20 > we can just enable the thing manually (if needed) and then tell the=20 > driver to not use pnp. >=20 > Your Crystal PNPBIOS devices were 00:0f and 00:10. In the directories: >=20 > /sys/devices/pnp0/00:0f >=20 > and >=20 > /sys/devices/pnp0/00:10 >=20 > there should be files named "resources". If you do a: >=20 > cat /sys/devices/pnp0/00:0f/resources I obtain : state =3D active io 0x210-0x217 > does it say "state =3D disabled" and nothing more? If so, try a: >=20 > echo auto >/sys/devices/pnp0/00:0f/resources I obtain nothing in sudo mode, or, in user mode : -bash: /sys/devices/pnp0/00:0f/resources: Permission denied > It should now look something like (that cat again): >=20 > state =3D disabled > io 0x534-0x537 > io 0x388-0x38b > io 0x220-0x22f > irq 5 > dma 1 > dma 0 >=20 > Then, do a: >=20 > echo activate >/sys/devices/pnp0/00:0f/resources Idem for the both mode > after which that "state =3D disabled" should have been changed to "stat= e =3D=20 > active". Ie, you should now have something like >=20 > state =3D active > io 0x534-0x537 *1 > io 0x388-0x38b *2 > io 0x220-0x22f *3 > irq 5 *4 > dma 1 *5 > dma 0 *6 >=20 > If it already looked like this before you echoed anything into the=20 > resource file, all the better! Good ! > For 00:10 (the CTRL port) you should, possibly again after echoing that= =20 > "auto" and "activate" into the resource file, you should have: >=20 > cat /sys/devices/pnp0/00:10/resources > state =3D active > io 0x538-0x53f *7 OK, I actived that and I obtain with the cat command : state =3D active io 0x530-0x537 io 0x388-0x38b io 0x220-0x22f irq 5 dma 1 dma 0 > All still well? If so, we can try loading the driver. I already made yo= u=20 > try this with the values as given here, so I'm hoping your values are=20 > different and it will work this time. Try: >=20 > modprobe snd-cs4236 isapnp=3D0 port=3D*1 cport=3D*7 irq=3D*4 dma1=3D*5 = dma2=3D*6 >=20 > substituting the *N items with the values from the resources files as=20 > numbered. I supose this is a question about io numbers, but which choose ? I used sudo modprobe snd-cs4236 isapnp=3D0 port=3D0x530 cport=3D0x388 irq= =3D5=20 dma1=3D1 dma2=3D0 I obtained this : CS4232+ soundcard not found or device busy FATAL: Error inserting snd_cs4236=20 (/lib/modules/2.6.15-23-386/kernel/sound/isa/cs423x/snd-cs4236.ko): No=20 such device > Did it work? Please? Aaaaah? And with snd-cs4232 and those same=20 Aaaah...! And with snd-cs4232 : ALSA=20 /home/hulin/alsa-driver-1.0.11rc5/isa/cs423x/../../alsa-kernel/isa/cs423x= /cs4236.x:294:=20 CS4232 WSS PnP manual resources are invalid, using auto config CS4232 WSS PnP configure failed for WSS (out of resources ?) PnP BIOS detection failed for CS4232 CS4232 soundcard not found or device busy FATAL: Error inserting snd_cs4232=20 (/lib/modules/2.6.15-23-386/kernel/sound/isa/cs423x/snd-cs4232.ko): No=20 such device > If it did work, you'd need to do that same echoing into that resource=20 > file that you did on every boot -- you could do this from your=20 > "rc.local" startup file if needed. If you didn't need any echoing (ie,=20 > your BIOS already actived everything, as is likely) you can just always= =20 > load the driver with those parameters. To just need a "modprobe=20 > snd-cs4236", you can say in /etc/modprobe.d/alsa-base (I believe that's= =20 > where ubuntu puts it but I'm not an ubuntu or debian user) >=20 > options snd-cs4236 isapnp=3D0 port=3D*1 cport=3D*7 irq=3D*4 dma1=3D*5 d= ma2=3D*6 >=20 >>> And a last resort for now: >>> >>> modprobe snd-sb8 port=3D0x220 irq=3D5 dma8=3D1 >> Hey ! No error ! >> With xmms, I have music ! >> But sound is very bad... How can I improve that ? >=20 > Well, you could play a bit with volumes in ALSA mixer to see if it=20 > improves but this is the wrong driver anyway. It's just that the Crysta= l=20 > chip _has_ a soundblaster compatible part so it can be used, but the=20 > chrystal driver should be preferred. It is good to know that it does=20 > work since that will mean that at least your BIOS has enabled that part= =20 > of the chip. >=20 > I'm hoping that with the above things will work. I can imagine it all=20 > looks a bit clumsy if you're a new Linux user but due to some weird=20 > reason the PnP-BIOS support for your chip was only added very recently=20 > and due to some other weird reason Ubuntu is not using 2.6.16 but=20 > 2.6.15. As indicated, on 2.6.16 "modprobe snd-cs4232" should be all you= =20 > need, and alsaconf should be able to figure that all out as well. >=20 > Rene. New after using alsaconf 1.0.11cr : The soundcard selection found two=20 cards on my system : cs4232 snd-cs4232 legacy Probe legacy ISA (non-PnP) chips It modify /etc/modprobe.d/sound and /etc/modprobe.conf In the first, I read alias snd-card-0 snd-cs4232 and alias sound-slot-0=20 snd-cs4232 The second don't exist. And no information about io ! I continu with alsa conf. I obtain : "OK, sound driver is configured"..=20 for adjustements volume, use your favorite mixer So, I type alsamixer, and I see : alsamixer: function snd_ctl_open failed for default: No such device And XMMS don't work. Ahhhh ! Near of the goal ? |
|
From: Rene H. <ren...@ke...> - 2006-06-11 19:01:09
|
Hulin Thibaud wrote: >> cat /sys/devices/pnp0/00:0f/resources > > I obtain : > state = active > io 0x210-0x217 >> cat /sys/devices/pnp0/00:10/resources > > OK, I actived that and I obtain with the cat command : > state = active > io 0x530-0x537 > io 0x388-0x38b > io 0x220-0x22f > irq 5 > dma 1 > dma 0 In a previous post you showed 00:0f to be the codec (to have id CSC0000, that is) and 00:10 to be the CTRL port (CSC0010). Seems like it's the other way around now... > I used sudo modprobe snd-cs4236 isapnp=0 port=0x530 cport=0x388 irq=5 > dma1=1 dma2=0 Try: modprobe snd-cs4236 isapnp=0 port=0x530 cport=0x210 irq=5 dma1=1 dma2=0 With both devices actived as you showed above, this should really work. If not, also try 0x534 instead of 0x530 for the port= value; I remember some confusion there with sometimes a +4 offset and sometimes not. That 0x388 is just the onboard OPL3 -- you can ignore it, or pass it as fm_port=0x388. sb_port=0x220 can also be passed, but isn't necessary either. > And with snd-cs4232 : > ALSA > /home/hulin/alsa-driver-1.0.11rc5/isa/cs423x/../../alsa-kernel/isa/cs423x/cs4236.x:294: > CS4232 WSS PnP manual resources are invalid, using auto config > CS4232 WSS PnP configure failed for WSS (out of resources ?) > PnP BIOS detection failed for CS4232 > CS4232 soundcard not found or device busy > FATAL: Error inserting snd_cs4232 > (/lib/modules/2.6.15-23-386/kernel/sound/isa/cs423x/snd-cs4232.ko): No > such device With just "modprobe snd-cs4232", or using the same parameters as above with snd-cs4236? (please also comment out any paramters you may have placed in /etc/modprobe.conf or /etc/modprobe.d/sound while you try) With the 1.0.11-rc5 driver, things should be detected. Unless ofcourse the device IDs really are swapped between codec and CTRL port; this would be very odd. snd-cs4232 expects CSC0000 (or CSC0100 or GIM0100) to be the codec. CSC0010 is always the CTRL port... Please verify by supplying the output of: cat /sys/devices/pnp0/00:0f/id cat /sys/devices/pnp0/00:0f/options cat /sys/devices/pnp0/00:0f/resources cat /sys/devices/pnp0/00:10/id cat /sys/devices/pnp0/00:10/options cat /sys/devices/pnp0/00:10/resources If there really is something odd there, we can add your IDs to snd-cs4236 in the future. You have now installed the 1.0.11-rc5 drivers onto your system (why -rc5 and not 1.0.11 final, by the way?). Given that the 1.0.11-rc5 snd-cs4232 thinks it should be driving CSC0000 you'd probably still get complaints about PnP failures, but: modprobe snd-cs4232 isapnp=0 port=0x530 cport=0x210 irq=5 dma1=1 dma2=0 should still work as well. It's better to use snd-cs4236 though. After things work, I'll be sure to get PnP BIOS support with these IDs into snd-cs4236 as well. For now, specifying the values hopefully works okay. Rene. |
|
From: Hulin T. <hul...@wa...> - 2006-06-13 23:13:20
|
Hello !
I tested yours advices in a script and this is the result (curiously,=20
the command modprobe has not writting in the file result but there=20
produce always the same message : error inserting... no such device).
The problem is that the command 00:10 (the NCTRL port) is not active but=20
close, and I can't active it ! Have I forgot how do ? I don't believe...
# =3D command in root mode (or coment)
# cat /etc/modprobe.d/sound
# alias snd-card-0 snd-cs4232 (coment line)
# alias sound-slot-0 snd-cs4232 (coment line)
# modprobe snd-cs4236 isapnp=3D0 port=3D0x530 cport=3DOx210 irq=3D5 dma1=3D=
1 dma2=3D0
# modprobe snd-cs4236 isapnp=3D0 port=3D0x534 cport=3DOx210 irq=3D5 dma1=3D=
1 dma2=3D0
# modprobe snd-cs4232
# modprobe snd-cs4236
# cat /sys/devices/pnp0/00:0f/id
CSC0010
# cat /sys/devices/pnp0/00:0f/options
Dependent: 01 - Priority acceptable
port 0x210-0x210, align 0xf, size 0x8, 16-bit address decoding
Dependent: 02 - Priority acceptable
port 0x340-0x340, align 0xf, size 0x8, 16-bit address decoding
Dependent: 03 - Priority acceptable
port 0x540-0x540, align 0xf, size 0x8, 16-bit address decoding
Dependent: 04 - Priority acceptable
port 0x120-0x120, align 0xf, size 0x8, 16-bit address decoding
# cat /sys/devices/pnp0/00:0f/resources
state =3D active
io 0x210-0x217
# cat /sys/devices/pnp0/00:10/id
CSC0000
# cat /sys/devices/pnp0/00:10/options
Dependent: 01 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x388-0x388, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 02 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x388-0x388, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5 High-Edge
dma 0,1,3 8-bit compatible
Dependent: 03 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x388-0x388, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 04 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x388-0x388, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
Dependent: 05 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x398-0x398, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 06 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x398-0x398, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
Dependent: 07 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x3a8-0x3a8, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 08 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x3a8-0x3a8, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
Dependent: 09 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x3b8-0x3b8, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 10 - Priority acceptable
port 0x530-0x530, align 0x3, size 0x8, 16-bit address decoding
port 0x3b8-0x3b8, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
Dependent: 11 - Priority acceptable
port 0x604-0x604, align 0x3, size 0x8, 16-bit address decoding
port 0x388-0x388, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 12 - Priority acceptable
port 0x604-0x604, align 0x3, size 0x8, 16-bit address decoding
port 0x388-0x388, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
Dependent: 13 - Priority acceptable
port 0x604-0x604, align 0x3, size 0x8, 16-bit address decoding
port 0x398-0x398, align 0x7, size 0x4, 16-bit address decoding
port 0x220-0x280, align 0x1f, size 0x10, 16-bit address decoding
irq 5,7,2/9,11,15 High-Edge
dma 0,1,3 8-bit compatible
dma 0,1,3 8-bit compatible
Dependent: 14 - Priority acceptable
# cat /sys/devices/pnp0/00:10/resources
state =3D disabled
io 0x530-0x537
io 0x388-0x38b
io 0x220-0x22f
irq 5
dma 1
dma 0
# modprobe snd-cs4232 isapnp=3D0 port=3D0x530 cport=3D0x210 irq=3D5 dma1=3D=
1 dma2=3D0
And a dmesg indicate this :
[4317362.523000] CS4236+ soundcard not found or device busy
[4317363.250000] **** SET: Misaligned resource pointer: c0892242 Type 04=20
Len 42
[4317363.253000] **** SET: Misaligned resource pointer: c08922c6 Type 00=20
Len 42
[4317363.255000] **** SET: Misaligned resource pointer: c089234a Type 01=20
Len 42
[4317363.399000] pnp: Device 00:10 activated.
[4317363.401000] ALSA=20
/home/hulin/alsa-driver-1.0.11rc5/isa/cs423x/../../alsa-kernel/isa/cs423x=
/cs4236.c:294:=20
CS4232 WSS PnP manual resources are invalid, using auto config
[4317363.406000] CS4232 WSS PnP configure failed for WSS (out of resource=
s?)
[4317363.408000] PnP BIOS detection failed for CS4232
[4317363.421000] pnp: Device 00:10 disabled.
[4317363.423000] cs4232-pnpbios: probe of 00:10 failed with error -16
[4317363.438000] CS4232 soundcard not found or device busy
Thanks,
Thibaud.
Rene Herman a =E9crit :
> Hulin Thibaud wrote:
>
>>> cat /sys/devices/pnp0/00:0f/resources
>> I obtain :
>> state =3D active
>> io 0x210-0x217
>
>>> cat /sys/devices/pnp0/00:10/resources
>> OK, I actived that and I obtain with the cat command :
>> state =3D active
>> io 0x530-0x537
>> io 0x388-0x38b
>> io 0x220-0x22f
>> irq 5
>> dma 1
>> dma 0
>
> In a previous post you showed 00:0f to be the codec (to have id CSC000=
0,
> that is) and 00:10 to be the CTRL port (CSC0010). Seems like it's the
> other way around now...
>
>> I used sudo modprobe snd-cs4236 isapnp=3D0 port=3D0x530 cport=3D0x388=
irq=3D5
>> dma1=3D1 dma2=3D0
>
> Try:
>
> modprobe snd-cs4236 isapnp=3D0 port=3D0x530 cport=3D0x210 irq=3D5 dma1=
=3D1 dma2=3D0
>
> With both devices actived as you showed above, this should really work=
.
> If not, also try 0x534 instead of 0x530 for the port=3D value; I remem=
ber
> some confusion there with sometimes a +4 offset and sometimes not.
>
> That 0x388 is just the onboard OPL3 -- you can ignore it, or pass it a=
s
> fm_port=3D0x388. sb_port=3D0x220 can also be passed, but isn't necessa=
ry=20
either.
>
>> And with snd-cs4232 :
>> ALSA
>>=20
/home/hulin/alsa-driver-1.0.11rc5/isa/cs423x/../../alsa-kernel/isa/cs423x=
/cs4236.x:294:=20
>> CS4232 WSS PnP manual resources are invalid, using auto config
>> CS4232 WSS PnP configure failed for WSS (out of resources ?)
>> PnP BIOS detection failed for CS4232
>> CS4232 soundcard not found or device busy
>> FATAL: Error inserting snd_cs4232
>> (/lib/modules/2.6.15-23-386/kernel/sound/isa/cs423x/snd-cs4232.ko): N=
o
>> such device
>
> With just "modprobe snd-cs4232", or using the same parameters as above
> with snd-cs4236? (please also comment out any paramters you may have
> placed in /etc/modprobe.conf or /etc/modprobe.d/sound while you try)
>
> With the 1.0.11-rc5 driver, things should be detected. Unless ofcourse
> the device IDs really are swapped between codec and CTRL port; this
> would be very odd. snd-cs4232 expects CSC0000 (or CSC0100 or GIM0100)
> to be the codec. CSC0010 is always the CTRL port...
>
> Please verify by supplying the output of:
>
> cat /sys/devices/pnp0/00:0f/id
> cat /sys/devices/pnp0/00:0f/options
> cat /sys/devices/pnp0/00:0f/resources
>
> cat /sys/devices/pnp0/00:10/id
> cat /sys/devices/pnp0/00:10/options
> cat /sys/devices/pnp0/00:10/resources
>
> If there really is something odd there, we can add your IDs to
> snd-cs4236 in the future.
>
> You have now installed the 1.0.11-rc5 drivers onto your system (why -r=
c5
> and not 1.0.11 final, by the way?). Given that the 1.0.11-rc5 snd-cs42=
32
> thinks it should be driving CSC0000 you'd probably still get complaint=
s
> about PnP failures, but:
>
> modprobe snd-cs4232 isapnp=3D0 port=3D0x530 cport=3D0x210 irq=3D5 dma1=
=3D1 dma2=3D0
>
> should still work as well. It's better to use snd-cs4236 though. After
> things work, I'll be sure to get PnP BIOS support with these IDs into
> snd-cs4236 as well. For now, specifying the values hopefully works oka=
y.
>
> Rene.
>
>
>
> _______________________________________________
> Alsa-user mailing list
> Als...@li...
> https://lists.sourceforge.net/lists/listinfo/alsa-user
>
>
|
|
From: Rene H. <ren...@ke...> - 2006-06-14 22:15:16
|
Hulin Thibaud wrote: > Hum, I'm not sure to understand... My english and my computers > knowlege are not very good ! Oh, I see... But no, never mind, it's just that the driver is now being loaded without using PnP (you are providing all the parameters manually on the modprobe command line). Best would be if that were not necessary. If you use the 1.0.11-rc5 snd-cs4232 driver that you compiled and installed, and you use simply "modprobe snd-cs4232", not specifying any parameters, it does not load? It gives you those PnP WSS configure failures in dmesg that I saw earlier? If so, I said I will not bother trying to debug/fix that and the manual route that now works will do. > I tested mplayer and that works, I can see divx ! And hear them. An old > pentium II 233Mhz in multimedia A Pentium 1 (MMX) probably in fact. "cat /proc/cpuinfo" will tell. > that's a bad news for hardware constructors, no ?!... You'll get skips and stuff yet; a 233 MHz CPU is extremely minimal. But generally, yes, for a number of years now non-gamers easily get by for quite a long time with their hardware. This in fact is only getting better; the power required for games versus year-of-publication seems much more non-linear then for non-games meaning that the non-gamer is buying a more and more ridiculously overpowered PC every time he upgrades -- which then lasts him longer. Rene. |
|
From: Hulin T. <hul...@wa...> - 2006-06-15 13:58:48
|
Rene Herman a =E9crit : > Hulin Thibaud wrote: >=20 >> Hum, I'm not sure to understand... My english and my computers >> knowlege are not very good ! >=20 > Oh, I see... But no, never mind, it's just that the driver is now being= =20 > loaded without using PnP (you are providing all the parameters manually= =20 > on the modprobe command line). Best would be if that were not necessary= . >=20 > If you use the 1.0.11-rc5 snd-cs4232 driver that you compiled and=20 > installed, and you use simply "modprobe snd-cs4232", not specifying any= =20 > parameters, it does not load? It gives you those PnP WSS configure=20 > failures in dmesg that I saw earlier? After reboot, modprobe snd-cs4232 give the usual error. Last 7 lines of=20 dmesg : [4297869.345000] pnp: Device 00:10 activated. [4297869.346000] ALSA=20 /home/hulin/alsa-driver-1.0.11rc5/isa/cs423x/../../alsa-kernel/isa/cs423x= /cs4236.c:294:=20 CS4232 WSS PnP manual resources are invalid, using auto config [4297869.348000] CS4232 WSS PnP configure failed for WSS (out of resource= s?) [4297869.349000] PnP BIOS detection failed for CS4232 [4297869.364000] pnp: Device 00:10 disabled. [4297869.364000] cs4232-pnpbios: probe of 00:10 failed with error -16 [4297869.372000] CS4232 soundcard not found or device busy > If so, I said I will not bother trying to debug/fix that and the manual= =20 > route that now works will do. Sure, thank you. But : do you think that manually debug at every start=20 is a good thing ? Is it difficult to fix the pnp=20 /sys/devices/pnp0/00:10/resources opened and automatize it ? >> I tested mplayer and that works, I can see divx ! And hear them. An ol= d=20 >> pentium II 233Mhz in multimedia >=20 > A Pentium 1 (MMX) probably in fact. "cat /proc/cpuinfo" will tell. Yes, it's a Pentium I. Hopeful... >> that's a bad news for hardware constructors, no ?!... >=20 > You'll get skips and stuff yet; a 233 MHz CPU is extremely minimal. But= =20 > generally, yes, for a number of years now non-gamers easily get by for=20 > quite a long time with their hardware. This in fact is only getting=20 > better; the power required for games versus year-of-publication seems=20 > much more non-linear then for non-games meaning that the non-gamer is=20 > buying a more and more ridiculously overpowered PC every time he=20 > upgrades -- which then lasts him longer. I agree, so much of people throw their old hardware because the new=20 windows is arrived... And Ubuntu or Mandrake, that are very popular for=20 linux big public, require new (or more ridiculously overpowered)=20 hardware. Idem for OpenOffice, Firefox, Gnome, KDE... This race to power=20 do live the market of computers, necessary sometimes but not often. So,=20 my reflexion is that developing tools and howtos for old (!) computers=20 can help people, democratize informatic and help linux to grow. Thibaud. |
|
From: Rene H. <ren...@ke...> - 2006-06-14 01:57:23
|
Hulin Thibaud wrote: > # modprobe snd-cs4236 isapnp=0 port=0x530 cport=Ox210 irq=5 dma1=1 dma2=0 <NUL>x210, not <OOH>x210 > # cat /sys/devices/pnp0/00:0f/id > CSC0010 > # cat /sys/devices/pnp0/00:0f/resources > state = active > io 0x210-0x217 > # cat /sys/devices/pnp0/00:10/id > CSC0000 > # cat /sys/devices/pnp0/00:10/resources > state = disabled > io 0x530-0x537 > io 0x388-0x38b > io 0x220-0x22f > irq 5 > dma 1 > dma 0 In an earlier posting you showed 00:0f to be CSC0000 and 00:10 to be CSC0010. But oh well, this way things are as expected. In another earlier posting I showed how to enable disabled devices. Please don't do any of these things "from a script" by the way. Just open a so called root shell (in stead of "shell", it might be called a "console" in the menus). If there's no seperate "root shell/console" entry in the menus, open a normal shell/console from the menus and type "su". You should now be asked for your root password, upon the supplying of which you will have become the root user. The state you showed above was after you tried loading drivers already which might have enabled/disabled things. Please do exactly as follows: 1) turn off your computer 2) turn on your computer and let it boot 3) open that root shell. 4) cat /sys/devices/pnp0/00:0f/resources does it still say (for all these, with possibly different values): (I) state = active io 0x210-0x217 if so, great, goto 7. If not, does it say: state = disabled io 0x210-0x217 if so, semi-great, goto 6. If it just says state = disabled do 5) echo auto >/sys/devices/pnp0/00:0f/resources 6) echo activate >/sys/devices/pnp0/00:0f/resources It should now look as indicated at (I) 7) cat /sys/devices/pnp0/00:0f/resources does it say (II) state = active io 0x530-0x537 io 0x388-0x38b io 0x220-0x22f irq 5 dma 1 dma 0 if so, great, goto 10. If not, does it say: state = disabled io 0x530-0x537 io 0x388-0x38b io 0x220-0x22f irq 5 dma 1 dma 0 if so, semi-great, goto 9. If it just says state = disabled do 8) echo auto >/sys/devices/pnp0/00:0f/resources 9) echo activate >/sys/devices/pnp0/00:0f/resources It should now look as indicated at (II). Using the values from those resource files (ie, substitute if they are different) do: 10) modprobe snd-cs4236 isapnp=0 port=0x530 cport=0x210 irq=5 dma1=1 dma2=0 11) If it loaded, great, goto 14. 12) modprobe snd-cs4236 isapnp=0 port=0x534 cport=0x210 irq=5 dma1=1 dma2=0 13) if it did not load, goto 18 14) start "alsamixer" 15) unmute (press "M" while on them) and turn up the master and PCM volumes if you have an hardware way of setting the volume (a knob) as well, it's adviced to set the master volume to 89 and the PCM volume to 87 and leave them at that, using the knob to make it louder/quieter. 16) play music and be jolly 17) stop here 18) repeatedly bang your head on the laptop until it's broken beyond any hope of salvation 19) stop here I really need to know first if things work when you do as described. If so, we can look at PnP to automate this. If not, I'm out of advice. Rene. |
|
From: Rene H. <ren...@ke...> - 2006-06-14 02:06:47
|
Rene Herman wrote: > 7) cat /sys/devices/pnp0/00:0f/resources > 8) echo auto >/sys/devices/pnp0/00:0f/resources > 9) echo activate >/sys/devices/pnp0/00:0f/resources Sorry, substitute 00:10 for 00:0f in step 7, 8 and 9. Rene. |
|
From: Hulin T. <hul...@wa...> - 2006-06-14 12:23:28
|
Yeaaaah, that's a great day ! It works !!!! You're really the best informatician that I know ! Congratulations Rene ! Excuse-me, I jump the step 18, because my wife don't share my passion=20 about computers and would be jealous of this practice... ! I'm astonish about your patience, your job is to assure support of alsa=20 ? Any way, I'm very happy to have discussed with you about Alsa. I'll write a how-to to configure sound with Alsa on Dell Latitude CP MMX=20 233 Pentium II, but before, I have one last thing to do. I wanted obtain Alsa because that's more quickly and there is a best=20 quality to listen music. And, I project to see video ! Mplayer=20 conceptors recommand to use Alsa for his rapidity. Now, I will try to see divx ! A new chalenge and a new adventure, and=20 become, one day, a real geek as you ?... Oups, I forget to automatise this ! OK, I'm listening you ! Rene Herman a =E9crit : > Hulin Thibaud wrote: >=20 >> # modprobe snd-cs4236 isapnp=3D0 port=3D0x530 cport=3DOx210 irq=3D5 dm= a1=3D1 dma2=3D0 >=20 > <NUL>x210, not <OOH>x210 >=20 >> # cat /sys/devices/pnp0/00:0f/id >> CSC0010 >> # cat /sys/devices/pnp0/00:0f/resources >> state =3D active >> io 0x210-0x217 >=20 >> # cat /sys/devices/pnp0/00:10/id >> CSC0000 >> # cat /sys/devices/pnp0/00:10/resources >> state =3D disabled >> io 0x530-0x537 >> io 0x388-0x38b >> io 0x220-0x22f >> irq 5 >> dma 1 >> dma 0 >=20 > In an earlier posting you showed 00:0f to be CSC0000 and 00:10 to be=20 > CSC0010. But oh well, this way things are as expected. In another=20 > earlier posting I showed how to enable disabled devices. >=20 > Please don't do any of these things "from a script" by the way. Just=20 > open a so called root shell (in stead of "shell", it might be called a= =20 > "console" in the menus). If there's no seperate "root shell/console"=20 > entry in the menus, open a normal shell/console from the menus and type= =20 > "su". You should now be asked for your root password, upon the supplyin= g=20 > of which you will have become the root user. >=20 > The state you showed above was after you tried loading drivers already=20 > which might have enabled/disabled things. Please do exactly as follows: >=20 > 1) turn off your computer > 2) turn on your computer and let it boot > 3) open that root shell. > 4) cat /sys/devices/pnp0/00:0f/resources >=20 > does it still say (for all these, with possibly different values): >=20 > (I) state =3D active > io 0x210-0x217 >=20 > if so, great, goto 7. If not, does it say: >=20 > state =3D disabled > io 0x210-0x217 >=20 > if so, semi-great, goto 6. If it just says >=20 > state =3D disabled >=20 > do >=20 > 5) echo auto >/sys/devices/pnp0/00:0f/resources > 6) echo activate >/sys/devices/pnp0/00:0f/resources >=20 > It should now look as indicated at (I) >=20 > 7) cat /sys/devices/pnp0/00:0f/resources >=20 > does it say >=20 > (II) state =3D active > io 0x530-0x537 > io 0x388-0x38b > io 0x220-0x22f > irq 5 > dma 1 > dma 0 >=20 > if so, great, goto 10. If not, does it say: >=20 > state =3D disabled > io 0x530-0x537 > io 0x388-0x38b > io 0x220-0x22f > irq 5 > dma 1 > dma 0 >=20 > if so, semi-great, goto 9. If it just says >=20 > state =3D disabled >=20 > do >=20 > 8) echo auto >/sys/devices/pnp0/00:0f/resources > 9) echo activate >/sys/devices/pnp0/00:0f/resources >=20 > It should now look as indicated at (II). >=20 > Using the values from those resource files (ie, substitute if they are=20 > different) do: >=20 > 10) modprobe snd-cs4236 isapnp=3D0 port=3D0x530 cport=3D0x210 irq=3D5 d= ma1=3D1 dma2=3D0 > 11) If it loaded, great, goto 14. >=20 > 12) modprobe snd-cs4236 isapnp=3D0 port=3D0x534 cport=3D0x210 irq=3D5 d= ma1=3D1 dma2=3D0 > 13) if it did not load, goto 18 >=20 > 14) start "alsamixer" > 15) unmute (press "M" while on them) and turn up the master and PCM vol= umes >=20 > if you have an hardware way of setting the volume (a knob) as well, it'= s=20 > adviced to set the master volume to 89 and the PCM volume to 87 and=20 > leave them at that, using the knob to make it louder/quieter. >=20 > 16) play music and be jolly > 17) stop here >=20 > 18) repeatedly bang your head on the laptop until it's broken beyond an= y=20 > hope of salvation > 19) stop here >=20 > I really need to know first if things work when you do as described. If= =20 > so, we can look at PnP to automate this. If not, I'm out of advice. >=20 > Rene. >=20 >=20 >=20 > _______________________________________________ > Alsa-user mailing list > Als...@li... > https://lists.sourceforge.net/lists/listinfo/alsa-user >=20 >=20 |
|
From: Rene H. <ren...@ke...> - 2006-06-14 20:55:10
|
Hulin Thibaud wrote: > A new chalenge and a new adventure, and become, one day, a real geek > as you ?... I'd advice against it. As to the PnP thing; if those PnP configure failures you got were with the 1.0.11-rc5 snd-cs4232 driver when you tried to load it without giving it any parameters let's just leave it at that -- debugging that via email would be too much of a hassle. Glad things are working... Rene. |
|
From: Hulin T. <hul...@wa...> - 2006-06-14 21:20:34
|
Hum, I'm not sure to understand... My english and my computers knowlege=20 are not very good ! I exactly followed yours advices : no failures, sound work. Debugging is=20 easy to end now ! Do you want I try with the Alsa 1.0.11 stable ? So, when I restart the PC, the port 00:10 is closed, and I must type=20 again in a root console : 8) echo auto >/sys/devices/pnp0/00:10/resources 9) echo activate >/sys/devices/pnp0/00:10/resources 12) modprobe snd-cs4236 isapnp=3D0 port=3D0x534 cport=3D0x210 irq=3D5 dma= 1=3D1 dma2=3D0 12) can be recorded in the /etc/modprobe.conf file. I can do a script as=20 /etc/init.d/alsa_pnp with 8) and 9), but maybe you have a best idea ? I tested mplayer and that works, I can see divx ! And hear them. An old=20 pentium II 233Mhz in multimedia, that's a bad news for hardware=20 constructors, no ?!... Rene Herman a =E9crit : > Hulin Thibaud wrote: >=20 >> A new chalenge and a new adventure, and become, one day, a real geek >> as you ?... >=20 > I'd advice against it. >=20 > As to the PnP thing; if those PnP configure failures you got were with=20 > the 1.0.11-rc5 snd-cs4232 driver when you tried to load it without=20 > giving it any parameters let's just leave it at that -- debugging that=20 > via email would be too much of a hassle. >=20 > Glad things are working... >=20 > Rene. >=20 >=20 >=20 |
|
From: Rene H. <ren...@ke...> - 2006-06-10 20:49:55
|
Hulin Thibaud wrote: > OK, I was installing the new kernel 2.6.16.20. I tried twice, but I > have errors at booting, that's concerning the soundcard... You have more trouble than just with the soundcard... > Kernel panic - not syncing: VFS: Unable to mount root fs on > unknown-block(0,0) That one is important for example... this is why I said I was reluctant to make you compile a new kernel. Did you try the /sys things on 2.6.15 that I described? Rene. |
|
From: Lee R. <rlr...@jo...> - 2006-06-07 23:22:41
|
On Thu, 2006-06-08 at 00:28 +0200, Rene Herman wrote: > I asked around, and it appears that Ubuntu does not yet have a 2.6.16 > kernel available either. Them releasing with 2.6.15 is in fact a bit > strange but nothing I can do about that. It's not really strange. 2.6.16 was released March 20th and the Dapper release was originally scheduled for April 15th. It was pushed back to June 1st so bugs could be fixed. That's not nearly enough time to regression test a new kernel. The Ubuntu kernel maintainers do pick up and backport kernel bug fixes that are important to their users, for example all the HDA intel fixes between ALSA 1.0.11-rc2 and 1.0.11 are in the Dapper kernel. I suspect this one simply did not make the cut as so few people have the hardware. Of course, they may pick it up if someone sends them a patch. Are you aware of any showstopper bugs missing from the Dapper kernel that are fixed in 2.6.16? Lee |
|
From: Rene H. <ren...@ke...> - 2006-06-07 23:33:27
|
Lee Revell wrote: > Are you aware of any showstopper bugs missing from the Dapper kernel > that are fixed in 2.6.16? No showstoppers directly but ever since Andrew posted that 2.6.16 was going into the distributions, people have been working their ass of on it. I'd have delayed dapper a bit more and shipped 2.6.16... Rene. |
|
From: Lee R. <rlr...@jo...> - 2006-06-08 00:42:59
|
On Thu, 2006-06-08 at 01:30 +0200, Rene Herman wrote: > Lee Revell wrote: > > > Are you aware of any showstopper bugs missing from the Dapper kernel > > that are fixed in 2.6.16? > > No showstoppers directly but ever since Andrew posted that 2.6.16 was > going into the distributions, people have been working their ass of on > it. I'd have delayed dapper a bit more and shipped 2.6.16... I think the slippery slope effect would have been an issue - if they had delayed the release to ship a new kernel for no clear reason, then people would have made a lot of noise about why it can't be delayed for their pet project. Personally I would like to have seen the release delayed to get Evolution into an acceptable state ;-) I don't think any more work went into 2.6.16 than 2.6.15 or any other kernel. Lee |
|
From: Sergei S. <ste...@li...> - 2006-06-14 22:28:51
|
On Thu, 15 Jun 2006 00:15:55 +0200 Rene Herman <ren...@ke...> wrote: > You'll get skips and stuff yet; a 233 MHz CPU is extremely minimal. But > generally, yes, for a number of years now non-gamers easily get by for > quite a long time with their hardware. This in fact is only getting > better; the power required for games versus year-of-publication seems > much more non-linear then for non-games meaning that the non-gamer is > buying a more and more ridiculously overpowered PC every time he > upgrades -- which then lasts him longer. > > Rene. > > A 600Mhz P3 was sufficient to play DVDs; a 300MHz was also sufficient to play DVDs, though using ATI's HW assist. |
