Re: [Openvortex-dev] Aureal Advantage (8810) problem

[Manuel Jander]

Hi,

Thanks for reporting this. It would be very interested in knowing more
about your computer (CPU, mainboard), specific Aureal Vortex sound card,
any other hardware you have installed (output of "lspci -vvv" would be
very useful).

On Mon, 2004-01-12 at 15:48, Kovacs Krisztian wrote:
>   Hi,
> 
>   I have some problems using ALSA 1.0rc2 and the Aureal driver. The driver
> detects the card, however, I have a lot of warning messages in the kernel
> log, and sometimes the sound is crappy. (High pitches, a lot of
> distortion.)

>   When loading the driver, I get the following messages:
> kecske kernel: PCI: Found IRQ 9 for device 0000:00:0f.0
> kecske kernel: Vortex: hardware init.... <6>done.
> kecske kernel: SDAC detected <3>vortex: unknown irq source 4

This warnings are OK, except "vortex: unknown irq source 4". Since we
reverse engineered the driver from the windows binary driver, we don't
have any documentation about the chip. That means we have still no idea
what the IRQ source flag "4" stands for, hence the warning.

>   I had a look at the source, it seems that vortex_interrupt() handler is
> getting interrupts from an unknown source. When the handler gets the
> source from the VORTEX_IRQ_SOURCE register, it gets 4... I've added a few
> debug messages, and it looks that sometimes it even gets 1004. (1000 would
> be IRQ_SOURCE_TIMER, but what is 4?)
> 
>   Even worse, sometimes I get a lot of ALSA warning messages:
> 
> kecske kernel: ALSA sound/core/pcm_lib.c:186: Unexpected hw_pointer value
> [1] (stream = 0, delta: -597, max jitter = 8192): wrong interrupt acknowledge?
> kecske kernel: ALSA sound/core/pcm_lib.c:186: Unexpected hw_pointer value
> [1] (stream = 0, delta: -84, max jitter = 8192): wrong interrupt acknowledge?
> kecske kernel: ALSA sound/core/pcm_lib.c:186: Unexpected hw_pointer value
> [1] (stream = 0, delta: -850, max jitter = 8192): wrong interrupt acknowledge?

How often do you get them ? Every then and when or at a higher rate
(more than one per second) ?

>   Apart from these problems, the driver works. When the pitches come,
> stopping and starting again usually helps, at least for a few minutes.
> However, these bugs are really annoying, so I'd be really glad if they
> could somehow be solved.  How can I help in this process? I have no idea
> what IRQ source 4 could be, neither how I could find it out.

There is are changes to come in the routing setup of the driver, once we
integrate 3D positional audio into the stable driver branch.
But if you want, you could try this driver snapshot:
http://galadriel.mat.utfsm.cl/~mjander/aureal/alsa/alsa-new-23-10-2003.tgz

>   Besides Linux, I have Windows XP on this computer, and that driver works
> without problems. How can I use Jeff Muizelaar's driver logger? Would
> those debug outputs be useful?

Regarding the driver logger ask Jeff Muizelaar. I have some older
snapshots of that on:
http://galadriel.mat.utfsm.cl/~mjander/aureal/re/


I hope you can provide me more info, to be able to track down this bug.

Best Regards

Manuel