2008-10-22 13:01:25     AD1981BL problem

Document created by Aaronwu Employee on Aug 8, 2013
Version 1Show Document
  • View in full screen mode

2008-10-22 13:01:25     AD1981BL problem

Andreas Bellgardt (GERMANY)

Message: 64056   


I have trouble with the AD1981BL on SPORT0 of a custom board together with BF537.

The board is similar to the BF537-Stamp and the sound circuit is similar to STAMP AD1981 AC97 Add-On card.

First some points to hardware diagnostics:

Oscillator of the AD1981 is up and running at 24,5MHz, BIT_CLK (pin 6) provides a 12,3MHz clock.

As I start the system with monitoring SDATA_IN, SDATA_OUT and SYNC I see no signals on the line, they are all 0!

I get the following messages from the driver:


Advanced Linux Sound Architecture Driver Version 1.0.12rc1 (Thu Jun 22 13:55:50 2006 UTC).

ASoC version 0.13.2

dma rx:3 tx:4, err irq:45, regs:ffc00800

AD1980 SoC Audio Codec

asoc: AC97 <-> bf5xx-ac97 mapping ok

AD1980 AC97 reset failed

ALSA device list:

  No soundcards found.


I searched in this forum and found tracker #4011 that describes this behaviour.

But the ticket is closed since 2008-09-12 so it should be fixed, right?


My SVN source revision is:


# svn info

Path: .

URL: svn://sources.blackfin.uclinux.org/linux-kernel/trunk

Repository Root: svn://sources.blackfin.uclinux.org/linux-kernel

Repository UUID: 526b6c2d-f592-4532-a319-5dd88ccb003d

Revision: 5462

Node Kind: directory

Schedule: normal

Last Changed Author: vapier

Last Changed Rev: 5462

Last Changed Date: 2008-10-21 07:59:51 +0200 (Tue, 21 Oct 2008)


Maybe my kernel config is wrong. Do I have to choose "Blackfin 53x SPORT support" ( BFIN_SPORT) in "character devices" together with the Blackfin ASOC support?

Please see my attached kernel config.











2008-10-22 14:46:50     Re: AD1981BL problem

Cameron Barfield (UNITED STATES)

Message: 64064   


In your config, you have:




Do you have the AD1980's reset line tied to a GPIO on the BF537? I believe the reference design has the reset line going to GPIO 5.




2008-10-23 02:43:48     Re: AD1981BL problem

Andreas Bellgardt (GERMANY)

Message: 64088   


Thanks Cameron,


I have the reset line connected to the system reset that is asserted on power-on by a reset chip.

That reset chip resets AD1981BL and the BF537. In the reference design there are both possibilities, system reset and a GPIO, but the GPIO is connected through a resistor stated as "DNP" that is "Do Not Populate" for me, right?


I tried to change the source of the driver to not to make a reset by a GPIO but when the driver tries to get a "VENDOR ID1" it reads 0 and exits. I looked at the datasheet of the AD1981 and I don't see why it could be necessary to reset the chip before communication.


Any suggestions?