I'll start the testing responses to this, your most recent, posting, and move to your previous posting after this one. I wanted to ask about udev before running the tests in your previous post on the assumption that its presence might change something fundamental in testing procedures. The tests I ran where I "got the same responses as before" were preliminary explorations to find out if that might be the case, and were not intended to be the responses you asked for earlier.
Right. I just wanted to save a round later, in case I hadn't been clear on this point.
Overall comment: The only real possibility of error I see in what you did is misspecification of the SB ioport. If that isn't wrong, I have nothing to suggest based on this report ... I'm probably misunderstanding how the driver works. Sorry. Go to Fry's, I guess.
On Thu, 7 Apr 2005, Ray Olszewski wrote:
At this point, my best suggestion is to follow the step-by-step procedure I described in my prior message, using /dev/.static/dev/dsp1 (or audio1) as the target device. Once I see the full report of the test, as I outlined it in my prior message, I (and others) should be able to indicate if there is an error of detail in your procedure or if the sound driver just does not work the way I think it does.
Here is the output when I try the /dev/.static/dev directories using 123ogg:
[EMAIL PROTECTED]:~/MyMusic$ sudo ogg123 -d oss -o dsp:/dev/.static/dev/dsp1 03\ -\ Norman\ Blake\ \ \ You\ Are\ My\ Sunshine.ogg
Audio Device: OSS audio driver output
Playing: 03 - Norman Blake You Are My Sunshine.ogg Ogg Vorbis stream: 2 channel, 44100 Hz Title: Norman Blake / You Are My Sunshine Artist: Various Track number: 3 Tracktotal: 19 Album: O Brother, Where Art Thou? Comment: Ripped with Sound Juicer Error: Cannot open device oss.
[EMAIL PROTECTED]:~/MyMusic$ sudo ogg123 -d oss -o dsp:/dev/.static/dev/audio1 03\ -\ Norman\ Blake\ \ \ You\ Are\ My\ Sunshine.ogg
Audio Device: OSS audio driver output
Playing: 03 - Norman Blake You Are My Sunshine.ogg Ogg Vorbis stream: 2 channel, 44100 Hz Title: Norman Blake / You Are My Sunshine Artist: Various Track number: 3 Tracktotal: 19 Album: O Brother, Where Art Thou? Comment: Ripped with Sound Juicer Error: Cannot open device oss.
OK. In context, this error means ogg123 cannot open (for example) dsp:/dev/.static/dev/audio1 ... either there is no physical device associated with the logical device or you don't have permission to open it. (I tested this here to be sure.)
[EMAIL PROTECTED]:~/MyMusic$
PS, -- In addition to what I asked for before, please include the EXACT output of both "ls -l /dev/dsp*" -AND- "ls -l /dev/.static/dev/dsp*".
[EMAIL PROTECTED]:~$ ls -l /dev/dsp* crw-rw---- 1 root audio 14, 3 2005-04-07 08:44 /dev/dsp crw-rw---- 1 root audio 14, 5 2005-04-07 08:44 /dev/dspW [EMAIL PROTECTED]:~$
[EMAIL PROTECTED]:~$ sudo ls -l /dev/.static/dev/dsp* ls: /dev/.static/dev/dsp*: No such file or directory [EMAIL PROTECTED]:~$
On the other hand, when I eliminate the wildcard and use real numbers I get:
[EMAIL PROTECTED]:~$ sudo ls -l /dev/.static/dev/dsp1 crw-rw---- 1 root audio 14, 19 2004-09-18 06:52 /dev/.static/dev/dsp1 [EMAIL PROTECTED]:~$ sudo ls -l /dev/.static/dev/dsp2 crw-rw---- 1 root audio 14, 35 2004-09-18 06:52 /dev/.static/dev/dsp2 [EMAIL PROTECTED]:~$ sudo ls -l /dev/.static/dev/dsp3 crw-rw---- 1 root audio 14, 51 2004-09-18 06:52 /dev/.static/dev/dsp3 [EMAIL PROTECTED]:~$ sudo ls -l /dev/.static/dev/dsp4 ls: /dev/.static/dev/dsp4: No such file or directory [EMAIL PROTECTED]:~$ sudo ls -l /dev/.static/dev/dsp crw-rw---- 1 root audio 14, 3 2004-09-18 06:52 /dev/.static/dev/dsp [EMAIL PROTECTED]:~$
Your guess is as good as mine as to why wildcards don't work here. Actually, your guess is probably better than mine :)
OK. This part looks OK. dsp1 has the right major and minor numbers. Assuming sudo works to give you permission to write to this device ... I normally do this by adding the userid I'm using to group audio (in /etc/group) ... this is not the source of your problem. If it works to access /dev/dsp, it is probably OK here too.
The "ls" problem bothers me a little bit. It probably means ls cannot read the directory /dev/.static/dev to expand the wildcard. Just a guess, though.
It's a longshot, but you might try creating the node /dev/dsp1 just as a test ("mknod /dev/dsp1 c 14 19"; "chmod 660 /dev/dsp1"; "chgrp audio /dev/dsp1" -- then check that it looks like this in ls -l: "crw-rw---- 1 root audio 14, 19 Jun 13 2001 /dev/dsp1") and now see if ogg123 can open it. It's a long shot, though.
2. Create the device entries if they are not present. (I was a bit surprised that they are not present, but if they are not, you are wrong that they "should be auto-created on boot if they are needed" ... well, maybe not wrong about "should*, but should != are, and in fact they are not.) Use mknod to create them by hand. (BTW, is /dev/dsp or /dev/dsp0 present? If not, you may be using devfs, and that puts the devices somewhere different ... I forget where since I don't use it, but maybe /dev/sound/dsp1 and so on. If /dev/dsp is present, see if it is s symlink to something other than /dev/dsp0 ... this might help you track down dsp1.) The relevant values for creating dsp1 are major 14, minor 19 (man mknod will explain this). Also make sure the userid you are using for testing has permission to write to this device ... making its permissions match /dev/dsp will probably suffice.
Skipped since I'm not sure if these nodes need to be created and/or copied, and I'm not sure if you're sure either. Or whether this step may be invalidated by the presence of udev. Corrections welcome.
Probably OK, as I said above.
3. Try one of the programs I suggested with the output directed to the appropriate dsp1 entry (which depends on where you find it, or put it, in step 2).
The output from trying to use /dev/.static/dev/dsp1 and /dev/.static/dev/audio1 above suffices, I presume?
If sound still does not work, include in your report:
A. The EXACT entry from /etc/modules .
[EMAIL PROTECTED]:~$ cat /etc/modules # /etc/modules: kernel modules to load at boot time. # # This file should contain the names of kernel modules that are # to be loaded at boot time, one per line. Comments begin with # a "#", and everything on the line after them are ignored. # pas2 io=220 irq=7 dma=1 # sb io=0x230 irq=5 dma=1 mpu_io=0x330
ide-cd ide-detect sd_mod pas2 io=0x388 irq=3 dma=5 sb_io=0x230 sb_irq=5 sb_dma=1 opl3 io=0x388 [EMAIL PROTECTED]:~$
Is 0x230 the right ioport? I thought you said in an earlier message that you'd moved the SB emulation to 0x220 (but I don't have the old messages to check). Did you move it back? Or is my memory failing?
I do recall that sb_irq= 5 is right. I don't remember what you did about sb_dma.
B. ALL dmesg output related to sound.
I'll try, presuming I can identify what's relevant:
[EMAIL PROTECTED]:~$ dmesg |more
<snip>
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
serio: i8042 AUX port at 0x60,0x64 irq 12
serio: i8042 KBD port at 0x60,0x64 irq 1
Serial: 8250/16550 driver $Revision: 1.90 $ 48 ports, IRQ sharing enabled
ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
<snip>
hdd: ATAPI 40X CD-ROM CD-R/RW drive, 8192kB Cache, UDMA(33)
Uniform CD-ROM driver Revision: 3.20
Pro Audio Spectrum driver Copyright (C) by Hannu Savolainen 1993-1996
YM3812 and OPL-3 driver Copyright (C) by Hannu Savolainen, Rob Hooft 1993-1996
Capability LSM initialized
kjournald starting. Commit interval 5 seconds
<snip>
[drm] Initialized sis 1.1.0 20030826 on minor 0: Silicon Integrated Systems [SiS] 300/305 PCI/AGP VGA Display Adapter
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
Sound: DMA (output) timed out - IRQ/DRQ config error?
This may indicate a mismatch between the card's hardware settings and the values in /etc/modules. Especially if you get these messages only when you try to specify sb_* values.
[EMAIL PROTECTED]:~$
C. Relevant parts of /proc/interrupts and /proc/ioports from AFTER step 3. (Interrupts don't show up until they are accessed.)
[EMAIL PROTECTED]:~$ cat /proc/interrupts CPU0 0: 17539757 XT-PIC timer 1: 33620 XT-PIC i8042 2: 0 XT-PIC cascade 3: 2244 XT-PIC PAS16 7: 1 XT-PIC parport0 8: 1 XT-PIC rtc 10: 8647 XT-PIC eth0 11: 192875 XT-PIC uhci_hcd 12: 44880 XT-PIC i8042 14: 18338 XT-PIC ide0 15: 35222 XT-PIC ide1 NMI: 0 LOC: 0 ERR: 0 MIS: 0 [EMAIL PROTECTED]:~$
[EMAIL PROTECTED]:~$ cat /proc/ioports 0000-001f : dma1 0020-0021 : pic1 0040-0043 : timer0 0050-0053 : timer1 0060-006f : keyboard 0070-0077 : rtc 0080-008f : dma page reg 00a0-00a1 : pic2 00c0-00df : dma2 00f0-00ff : fpu 0170-0177 : ide1 01f0-01f7 : ide0 0376-0376 : ide1 0378-037a : parport0 037b-037f : parport0 0388-038b : Yamaha OPL3 03c0-03df : vga+ 03f6-03f6 : ide0 03f8-03ff : serial 0778-077a : parport0 0800-083f : 0000:00:07.3 0800-083f : pnp 00:00 0840-085f : 0000:00:07.3 0850-085f : pnp 00:00 0850-0857 : piix4-smbus 0cf8-0cff : PCI conf1 d880-d8ff : 0000:00:0e.0 dc00-dc7f : 0000:00:0d.0 dc00-dc7f : 0000:00:0d.0 dce0-dcff : 0000:00:07.2 dce0-dcff : uhci_hcd e000-efff : PCI Bus #01 ec00-ecff : 0000:01:00.0 ffa0-ffaf : 0000:00:07.1 ffa0-ffa7 : ide0 ffa8-ffaf : ide1 [EMAIL PROTECTED]:~$
Hey, why are ide0 and parport0 showing up thrice? Note that no address in the 200 range are showing up.
D. The EXACT line you used in running the music program you test with.
See my first answer above. Does it suffice? Copied right from the terminal where I ran it.
E. Any STDOUT or STDERR output from the test music program (like you did before).
I think it's there with the command, right?
F. The output of "ls -l /dev/dsp*" (or the equivalent if you are using devfs).
See above.
The response above does cover all 3 of these.
One last detail -- you say you are testing with this command:
[EMAIL PROTECTED]:~/MyMusic$ sudo mpg123 -a /dev/audio1 03\ -\ Norman\ Blake\ \ \ You\ Are\ My\ Sunshine.ogg
Last time I checked, mpg123 does not play ogg files. You need to match the app to the file type ... use ogg123 with *.ogg files, mpg123 with *.mp3 files. This is not the cause of the problem you report though ... most likely that is the absence of /dev/dsp1 .
Ok. Registered. I never used mpg123 before and was expecting it to give me some indication if I sent it the wrong kind of file.
In your tests, mpg123 never got around to attempting to play the file (and Linux apps in general don't pay attention to extensions), so mpg123 never "noticed" that it wasn't in mp3 format. For example:
[EMAIL PROTECTED]:~$ mpg123 /home/music/ogg/Bob_Seger_\&_The_Silver_Bullet_Band/Live_Bullet/Katmandu.ogg
High Performance MPEG 1.0/2.0/2.5 Audio Player for Layer 1, 2 and 3.
Version 0.59r (1999/Jun/15). Written and copyrights by Michael Hipp.
Uses code from various people. See 'README' for more!
THIS SOFTWARE COMES WITH ABSOLUTELY NO WARRANTY! USE AT YOUR OWN RISK!
Directory: /home/music/ogg/Bob_Seger_&_The_Silver_Bullet_Band/Live_Bullet/
Playing MPEG stream from Katmandu.ogg ...
Junk at the beginning 4f676753
Free format not supported: (head ffff0703)
Junk at the beginning 766f7262
MPEG 1.0 layer I, 320 kbit/s, 48000 Hz dual-channel
Illegal Audio-MPEG-Header 0x7017380c at offset 0x1709.
Skipped 3152148 bytes in input.
Illegal Audio-MPEG-Header 0xcb0ff566 at offset 0x30315d.
Of course, we folks who never make mistakes normally don't get to see the error messages, so how would we know what they say?
James
PS Miscellaneous observations: the opl3 module will assign itself to io 0x388 while the pas2 module will not--regardless of the presence or absence of the opl3 driver with its io parameter (tried it both with and without). The kernel doc seemed to assign pas2 and opl3 (where needed, and I'm not sure I need it: seems to make no difference in sound output whether it's there or not) the same io unabashedly, so I did the same. Assigning the pas2 module to io 0x230 made no difference in sound characteristics that I could observe--whether or not a sb_io=0x230 was present or absent. And the io still didn't show up in cat /proc/ioports output in any case. Leaving the io paramter absent from the pas2 line results in an error message, as follows:
Mar 28 20:57:29 localhost kernel: Pro Audio Spectrum driver Copyright (C) by Hannu Savolainen 1993-1996 Mar 28 20:57:29 localhost kernel: I/O, IRQ, DMA and type are mandatory
--regardless of whether I've specified sb_io= or not on that same line.
PPS Whew! I should get an honorary geek doctorate for this, a good citizenship award, or at least be taught the secret handshake of the Geek International Cadre. How 'bout it, Ray?
Shhh. There is no secret Geek International Cadre, let alone a secret handshake. Nothing to see here folks ... move along. And pay no attention to the man behind that curtain ....
- To unsubscribe from this list: send the line "unsubscribe linux-newbie" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.linux-learn.org/faqs
