2003-05-16 13:49:48

by Vojtech Pavlik

[permalink] [raw]
Subject: [patch] Fix support for SiS5581/5582/5596 IDE

Hi!

The last patch for SiS96* also added support for SiS5581, SiS5582 and
SiS5596. However, the PCI IDs for these three chips were incorrect by
mistake (cut and paste problem). This patch, on top of the previous one
fixes it, and thus adds proper support for these old chips.

--
Vojtech Pavlik
SuSE Labs, SuSE CR


Attachments:
(No filename) (316.00 B)
sis5581-2.4.diff (1.70 kB)
sis5581-2.5.diff (1.70 kB)
Download all attachments

2003-05-16 14:00:32

by Lionel Bouton

[permalink] [raw]
Subject: Re: [patch] Fix support for SiS5581/5582/5596 IDE

Vojtech Pavlik wrote:

>Hi!
>
>The last patch for SiS96* also added support for SiS5581, SiS5582 and
>SiS5596. However, the PCI IDs for these three chips were incorrect by
>mistake (cut and paste problem). This patch, on top of the previous one
>fixes it, and thus adds proper support for these old chips.
>
>

Thanks for the work.
I've yet to read carefully each change (will do this week-end), but I
agree on the principles.

The only thing that disturbed me was the removal of the DEBUG code (as
it often helped me find out bugs and understand new chips' behavior) but
I think it's best for me to maintain a patch which adds this code in the
very rare cases where it can be usefull.

LB.

2003-05-16 14:08:10

by Vojtech Pavlik

[permalink] [raw]
Subject: Re: [patch] Fix support for SiS5581/5582/5596 IDE

On Fri, May 16, 2003 at 04:13:17PM +0200, Lionel Bouton wrote:

> >The last patch for SiS96* also added support for SiS5581, SiS5582 and
> >SiS5596. However, the PCI IDs for these three chips were incorrect by
> >mistake (cut and paste problem). This patch, on top of the previous one
> >fixes it, and thus adds proper support for these old chips.
>
> Thanks for the work.
> I've yet to read carefully each change (will do this week-end), but I
> agree on the principles.

Great. And thanks for veryfing the patch.

> The only thing that disturbed me was the removal of the DEBUG code (as
> it often helped me find out bugs and understand new chips' behavior) but
> I think it's best for me to maintain a patch which adds this code in the
> very rare cases where it can be usefull.

I sort of expected that when I was removing it. ;) Debugging code comes
very handy when one runs into problems, but clutters up the source in
all other cases. So the final decision to keep it or have a separate
debug patch is of course in your hands.

--
Vojtech Pavlik
SuSE Labs, SuSE CR