Alan Cox <[email protected]> writes:
> > > I guess in the mean time SIS board are best avoided. Who knows if they'll
> > > ever actually deliver
> >
> > Why suggesting to avoid this board that looks pretty fast, stable and is
> > very cheap? A reasonable and low cost Sound board will fit and will give
> > far better sound quality for less CPU load.
>
> because so many people have bought things on the promise of vendor support
> that never appeared or was at best minimal
Indeed. Could you all please hassle SiS for the datasheet for the 7012
integrated audio controller in the SiS 735?
They say you should contact their marketing people about datasheets
using a web based form
http://www.sis.com/contact/salescontact.htm
Alternatively, random people at SiS can be spammed about the lack of
datasheet or open linux driver. It'd be nice if someone could give
them a physical telephone call, but I don't have access to a landline
so it'd be really expensive for me :-(
>From the SiS website:
China
Ms. Ellie Yin
Tel:886+2+29161619 ext.346
E-mail: [email protected].
Europe, Taiwan, Japan, Korea:
Ms. Jessie Lee
Tel:886+2+29161619 ext.341
E-mail: [email protected]
North America, Oceania:
Miss Michele Huang
Tel:886+2+29161619 ext.345
E-mail: [email protected].
All other regions(S.E. Asia, India, Middle-East, Latin America, Russia):
Ms. Ellie Yin
Tel:886+2+29161619 ext.346
E-mail: [email protected]
[...]
--
http://john.fremlin.de
> Alternatively, random people at SiS can be spammed about the lack of
> datasheet or open linux driver. It'd be nice if someone could give
> them a physical telephone call, but I don't have access to a landline
> so it'd be really expensive for me :-(
If you don't have a specific reason to ask then don't hassle them. On the
whole SiS work very well. But if you have a 735 board and want audio then
yes - can't hurt to make them realise how many intel boards they may be
selling 8)
> Indeed. Could you all please hassle SiS for the datasheet for the 7012
> integrated audio controller in the SiS 735?
Sis is quite good at writing Linux drivers and they release the source
under GPL - just search through google for bug reports for the sis900
network driver.
And it's probably the only driver with a large list of the PHY's that
are used by the mobo manufacturers with the nic, and the various ways to
get at the correct negotiation result. That's something you won't be
able to write even with the sis datasheet.
Just wait a bit, or try to convince nvidia that they should release the
source of their driver if you want to do something now.
--
Manfred
On Sun, 4 Nov 2001, Manfred Spraul wrote:
> > Indeed. Could you all please hassle SiS for the datasheet for the 7012
> > integrated audio controller in the SiS 735?
>
> Sis is quite good at writing Linux drivers and they release the source
> under GPL - just search through google for bug reports for the sis900
> network driver.
Agreed, but we need more generic drivers when possible anyway. Just think
about Tekram drivers for DC-390 boards. They just ignored anything
different from Tekram adapters. Btw, my Netgear FA311 board is not handled
by the sis driver of linux-2.2.20 and my little finger tells me that it
could be so given a few code addition.
> And it's probably the only driver with a large list of the PHY's that
> are used by the mobo manufacturers with the nic, and the various ways to
> get at the correct negotiation result. That's something you won't be
> able to write even with the sis datasheet.
Let me adver^H^Hocate: :-)
Just some pricing from France:
- K7S5A (without NIC option) : FF 590
- ASUS A7M (example) : FF 1300
- CM8378 Mentor sound board : FF 119
K7S5A + Mentor sound board -> FF 709
ASUS A7M (with bogus VIA 686B) -> FF 1300
The K7S5A is a great Mobo, good quality, very fast and very cheap, as long
as you donnot want to use on-board sound. :)
The K7S5A I have is used by my children under Windows, but I have stressed
it a lot under Linux and FreeBSD prior to leave them play with it.
> Just wait a bit, or try to convince nvidia that they should release the
> source of their driver if you want to do something now.
Good point.
G?rard.
G?rard Roudier wrote:
> different from Tekram adapters. Btw, my Netgear FA311 board is not handled
> by the sis driver of linux-2.2.20 and my little finger tells me that it
> could be so given a few code addition.
Unless you have a really strange board I haven't seen, NetGear FA311 are
the natsemi DP83815/6 chips, handling by either "natsemi" or "fa311"
drivers, not "sis900" driver...
What's the PCI id?
Jeff
--
Jeff Garzik | Only so many songs can be sung
Building 1024 | with two lips, two lungs, and one tongue.
MandrakeSoft | - nomeansno
Jeff Garzik wrote:
>
> G?rard Roudier wrote:
> > different from Tekram adapters. Btw, my Netgear FA311 board is not handled
> > by the sis driver of linux-2.2.20 and my little finger tells me that it
> > could be so given a few code addition.
>
> Unless you have a really strange board I haven't seen, NetGear FA311 are
> the natsemi DP83815/6 chips, handling by either "natsemi" or "fa311"
> drivers, not "sis900" driver...
>
sis900 and natsemi are similar, probably both could be handled with one
driver.
e.g. freebsd has one driver for natsemi and sis900.
But I'm not a big fan of huge drivers that handle multiple 99%
compatible controllers and always break for one controller if you try to
fix another controller, so I won't try to merge them.
--
Manfred
On Sun, 4 Nov 2001, Manfred Spraul wrote:
> Jeff Garzik wrote:
> >
> > G?rard Roudier wrote:
> > > different from Tekram adapters. Btw, my Netgear FA311 board is not handled
> > > by the sis driver of linux-2.2.20 and my little finger tells me that it
> > > could be so given a few code addition.
> >
> > Unless you have a really strange board I haven't seen, NetGear FA311 are
> > the natsemi DP83815/6 chips, handling by either "natsemi" or "fa311"
> > drivers, not "sis900" driver...
> >
> sis900 and natsemi are similar, probably both could be handled with one
> driver.
> e.g. freebsd has one driver for natsemi and sis900.
>
> But I'm not a big fan of huge drivers that handle multiple 99%
> compatible controllers and always break for one controller if you try to
> fix another controller, so I won't try to merge them.
So you would have preferred, for example, to have dozens of different
drivers for SYM53C8XX chips and probably as many for Adaptec aic7xxx ones.
And, probably, one set of different drivers per O/S. And why not one set
per O/S major version and even per adjacent ones of the same O/S.
Given all the different brands that use similar or compatibles chips, the
way you want drivers to be developped and maintained looks just
unrealistic to me.
G?rard.
G?rard Roudier wrote:
>
> On Sun, 4 Nov 2001, Manfred Spraul wrote:
>
>
>>Jeff Garzik wrote:
>>
>>>G?rard Roudier wrote:
>>>
>>>>different from Tekram adapters. Btw, my Netgear FA311 board is not handled
>>>>by the sis driver of linux-2.2.20 and my little finger tells me that it
>>>>could be so given a few code addition.
>>>>
>>>Unless you have a really strange board I haven't seen, NetGear FA311 are
>>>the natsemi DP83815/6 chips, handling by either "natsemi" or "fa311"
>>>drivers, not "sis900" driver...
>>>
>>>
>>sis900 and natsemi are similar, probably both could be handled with one
>>driver.
>>e.g. freebsd has one driver for natsemi and sis900.
>>
>>But I'm not a big fan of huge drivers that handle multiple 99%
>>compatible controllers and always break for one controller if you try to
>>fix another controller, so I won't try to merge them.
>>
>
> So you would have preferred, for example, to have dozens of different
> drivers for SYM53C8XX chips and probably as many for Adaptec aic7xxx ones.
> And, probably, one set of different drivers per O/S. And why not one set
> per O/S major version and even per adjacent ones of the same O/S.
>
> Given all the different brands that use similar or compatibles chips, the
> way you want drivers to be developped and maintained looks just
> unrealistic to me.
Jeff has intimate knowledge of the Tulip driver, one of the more complex
drivers that supports a bazillion different cards. And also one of the hardest
to get (and keep) working on all of the devices it seems....
I think he has a very valid point....
Ben
>
> G?rard.
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
>
--
Ben Greear <[email protected]> <Ben_Greear AT excite.com>
President of Candela Technologies Inc http://www.candelatech.com
ScryMUD: http://scry.wanfear.com http://scry.wanfear.com/~greear
On Mon, 5 Nov 2001, Ben Greear wrote:
> G?rard Roudier wrote:
>
> >
> > On Sun, 4 Nov 2001, Manfred Spraul wrote:
> >
> >
> >>Jeff Garzik wrote:
> >>
> >>>G?rard Roudier wrote:
> >>>
> >>>>different from Tekram adapters. Btw, my Netgear FA311 board is not handled
> >>>>by the sis driver of linux-2.2.20 and my little finger tells me that it
> >>>>could be so given a few code addition.
> >>>>
> >>>Unless you have a really strange board I haven't seen, NetGear FA311 are
> >>>the natsemi DP83815/6 chips, handling by either "natsemi" or "fa311"
> >>>drivers, not "sis900" driver...
> >>>
> >>>
> >>sis900 and natsemi are similar, probably both could be handled with one
> >>driver.
> >>e.g. freebsd has one driver for natsemi and sis900.
> >>
> >>But I'm not a big fan of huge drivers that handle multiple 99%
> >>compatible controllers and always break for one controller if you try to
> >>fix another controller, so I won't try to merge them.
> >>
> >
> > So you would have preferred, for example, to have dozens of different
> > drivers for SYM53C8XX chips and probably as many for Adaptec aic7xxx ones.
> > And, probably, one set of different drivers per O/S. And why not one set
> > per O/S major version and even per adjacent ones of the same O/S.
> >
> > Given all the different brands that use similar or compatibles chips, the
> > way you want drivers to be developped and maintained looks just
> > unrealistic to me.
>
>
> Jeff has intimate knowledge of the Tulip driver, one of the more complex
> drivers that supports a bazillion different cards. And also one of the hardest
> to get (and keep) working on all of the devices it seems....
>
> I think he has a very valid point....
Ok. I do understand the point, now.
Sorry for the misunderstanding. I caught the remark as a too general
statement.
May thanks for the efforts maintaining tulip drivers.
Regards,
G?rard.
> So you would have preferred, for example, to have dozens of different
> drivers for SYM53C8XX chips and probably as many for Adaptec aic7xxx on=
> es.
The problem with tulip clones is a bit different. Imagine if each SYS53C8XX
chip had ten clone versions that each understood 90% of the official
instruction set and had different magically unique bugs some of which were
not documented.
Dealing with things like GPIO variants, LED wiring, custom NVRAM is a bit
cleaner. The very core of the tulip clones is variable
Alan