Re: Music with sb64pci? (final statement)

From: Gerhard Thimm ([email protected])
Date: Thu Sep 28 2000 - 14:24:01 CEST


Michele Andreoli schrieb:
>
> On Wed, Sep 27, 2000 at 05:07:54PM +0200, Gerrit Lammert nicely wrote:
> > Hi Michele (and all),
> >
> > just to bring this to an end and solve any misunderstanding.
> > Two things brought me to my first mail with this topic:
> > 1. I earlier (1 year or so) played mp3s to the systemspeaker with mulinux.
> > It sounded awfull, but I was impressed, what this little peep-device can be
> > forced to do. But in 10r0 when I wanted to try again, all was quiet.
> > 2. The (stupid) idea: With an pci-card you don't need irq/io, might it be
> > possible to change the sb16-driver/script slightly to work with this card.
> >
> > Nr. 1: When I removed the "no387" line everything worked fine, so that
> > "problem" is solved. :-)
>
> What problem? The pc-speaker problem, or SB problem?
>
> Anycase, we should return back to reality: the sound patch idea is
> for ISA card. The sb module in muLinux do not work with PCI cards:
> indeed, ioport is a ISA concept, or i'm wrong?

The Intel-Processors support two physical-address-spaces. A
memory-adress-space, which has a size of 1MB in real-mode and 4GB in
protected-mode. Additionally the intel-processors have a
IO-address-space of 64KB. To access this two Address-spaces there a
different control-signal necessary. To access memory there are the
signals Memory-Read and Memory-Write. For he IO-Space there are IO-Read
and IO-Read. In the assembly-language there are different Commands
(Operation-Codes) to perform this Memory-IO-accesses. "load" and "store"
for memory and "input" and "output" for IO.

I think the io-concept also exists in the pci-world (and also in the
agp-world). Look at a graphics-card, which has the video-memory in
memory-space and the video-control-registers in io-space. You can run
the oldest dos-pacman on all this cards, so i think ioports are an
intel-concept, not isa.

Gerhard

>
> Michele

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]



This archive was generated by hypermail 2.1.6 : Sat Feb 08 2003 - 15:27:16 CET