Re: Trident problems.

[Takashi Iwai <tiwai@suse.de>]

[-- Attachment #1: Type: text/plain, Size: 2491 bytes --]

At Thu, 28 Aug 2003 00:57:41 -0700,
David Parker wrote:
> 
> Hi,
> 
> A long time ago (many months... sorry I don't know exactly when), my Trident
> card stopped working with ALSA on a Debian unstable box. It still works with
> the kernel OSS module, but ALSA has problems with the AC'97 codec. Here's what
> I tried and the resulting log messages/output:
> 
> hypnotic:/home/dap# modprobe snd-trident
> ALSA ../../alsa-kernel/pci/trident/trident_main.c:3394: AC'97 codec ready error
> Trident 4DWave PCI soundcard not found or device busy
> devfs_register(unknown): could not append to parent, err: -17
> ALSA ../../alsa-kernel/pci/ac97/ac97_codec.c:2096: AC'97 0:0 does not respond - RESET
> Trident 4DWave PCI soundcard not found or device busy
> 
> hypnotic:/home/dap# modprobe trident
> Trident 4DWave/SiS 7018/ALi 5451,Tvia CyberPro 5050 PCI Audio, version 0.14.10h, 18:35:01 Jul 19 2003
> trident: Trident 4DWave DX found at IO 0x1400, IRQ 19
> ac97_codec: AC97  codec, id: ""D68 (Unknown)
> 
> hypnotic:/home/dap# uname -a
> Linux hypnotic 2.4.22-pre7 #1 SMP Sat Jul 19 18:29:55 PDT 2003 i686 GNU/Linux
> 
> hypnotic:/proc/asound# cat version 
> Advanced Linux Sound Architecture Driver Version 0.9.4.
> Compiled on Jul 19 2003 for kernel 2.4.22-pre7 (SMP) with versioned symbols.
> 
> hypnotic:/home/dap# lspci -n -v -s 00:0b.0
> 00:0b.0 Class 0401: 1023:2000 (rev 02)
>         Subsystem: 1023:2000
>         Flags: bus master, medium devsel, latency 64, IRQ 19
>         I/O ports at 1400 [size=256]
>         Memory at e8002000 (32-bit, non-prefetchable) [size=4K]
>         Capabilities: [48] Power Management version 1
> 
> I saw someone else had this same problem on the alsa-devel list a while ago
> (http://www.mail-archive.com/alsa-devel@lists.sourceforge.net/msg06500.html)
> but it doesn't look like it was ever resolved. Is there something I need to
> tweak, or is this a valid bug? If it would help to test specific old versions
> of ALSA I could do that too.

could you please try the attached patch?
if it still doesn't solve the problem, try to increase the length of
timeout in snd_trident_4d_dx_init(), e.g. 

	end_time = jiffies + HZ * 2;

it will wait for 2 seconds.
try this with and without the patch.

if the same problem happens, try to commend out the line
"return -EIO;" in the same function, so that the probing proceeds even
if the check fails.  let's see whether the further ac97 probing
works.

if still fails...  remove the warm reset part.


Takashi

[-- Attachment #2: trident-4d-fix.dif --]
[-- Type: application/octet-stream, Size: 1082 bytes --]

Index: alsa-kernel/pci/trident/trident_main.c
===================================================================
RCS file: /suse/tiwai/cvs/alsa/alsa-kernel/pci/trident/trident_main.c,v
retrieving revision 1.39
diff -u -r1.39 trident_main.c
--- alsa-kernel/pci/trident/trident_main.c	2 Sep 2003 18:04:54 -0000	1.39
+++ alsa-kernel/pci/trident/trident_main.c	10 Sep 2003 12:24:46 -0000
@@ -3387,9 +3387,6 @@
 	outl(0x00000001, TRID_REG(trident, DX_ACR2_AC97_COM_STAT));
 	udelay(100);
 	outl(0x00000000, TRID_REG(trident, DX_ACR2_AC97_COM_STAT));
-	/* DAC on, disable SB IRQ and try to force ADC valid signal */
-	trident->ac97_ctrl = 0x0000004a;
-	outl(trident->ac97_ctrl, TRID_REG(trident, DX_ACR2_AC97_COM_STAT));
 	/* wait, until the codec is ready */
 	end_time = (jiffies + (HZ * 3) / 4) + 1;
 	do {
@@ -3401,6 +3398,10 @@
 	return -EIO;
 
  __dx_ok:
+	/* DAC on, disable SB IRQ and try to force ADC valid signal */
+	trident->ac97_ctrl = 0x0000004a;
+	outl(trident->ac97_ctrl, TRID_REG(trident, DX_ACR2_AC97_COM_STAT));
+
 	snd_trident_stop_all_voices(trident);
 
 	return 0;