[OT] Intel chipset development documents

[OT] Intel chipset development documents

[Greg Sheard]

Hi all. Apologies for OT, but this seems the best place to ask.
Intel have removed documentation for the 430VX (Triton) chipset from
their developer site. I'm trying to check that the access details for
the Southbridge are the same as for the 440BX chipset, since I'm working
on some code for direct PCI access. If they're not, could somebody
please let me have the relevant documentation?

Thanks in advance,

Greg Sheard.
ECSC Ltd.

Re: [OT] Intel chipset development documents

[Martin Mares]

Hello!

> Hi all. Apologies for OT, but this seems the best place to ask.
> Intel have removed documentation for the 430VX (Triton) chipset from
> their developer site. I'm trying to check that the access details for
> the Southbridge are the same as for the 440BX chipset, since I'm working
> on some code for direct PCI access. If they're not, could somebody
> please let me have the relevant documentation?

I guess I could have some 430VX documentation at home (will check
tomorrow).

As far as I remember, Configuration Type 1 should be supported since
the earliest Intel chipset, Type 2 could vary.

				Have a nice fortnight
-- 
Martin `MJ' Mares   <mj@ucw.cz>   http://atrey.karlin.mff.cuni.cz/~mj/
Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
God is real, unless declared integer.

Re: [OT] Intel chipset development documents

[Greg Sheard]

On Fri, 2001-11-02 at 17:38, Martin Mares wrote:
> Hello!
> 
> > Hi all. Apologies for OT, but this seems the best place to ask.
> > Intel have removed documentation for the 430VX (Triton) chipset from
> > their developer site. I'm trying to check that the access details for
> > the Southbridge are the same as for the 440BX chipset, since I'm working
> > on some code for direct PCI access. If they're not, could somebody
> > please let me have the relevant documentation?
> 
> I guess I could have some 430VX documentation at home (will check
> tomorrow).
> 
> As far as I remember, Configuration Type 1 should be supported since
> the earliest Intel chipset, Type 2 could vary.
> 

Thanks for that Martin, much appreciated.

I already have the configuration type down (it's 1), but the 430VX and
also the VIA 585 seem only to report host bridges. I'm unable to spot
the piece of code which does different PCI-related things for these
chipsets in the kernel. Does anybody know if a workaround is applied?

Regards,
Greg.

Re: [OT] Intel chipset development documents

[Charles Marslett]

Greg Sheard wrote:
> 
> On Fri, 2001-11-02 at 17:38, Martin Mares wrote:
> > Hello!
> >
> > > Hi all. Apologies for OT, but this seems the best place to ask.
> > > Intel have removed documentation for the 430VX (Triton) chipset from
> > > their developer site. I'm trying to check that the access details for
> > > the Southbridge are the same as for the 440BX chipset, since I'm working
> > > on some code for direct PCI access. If they're not, could somebody
> > > please let me have the relevant documentation?
> >
> > I guess I could have some 430VX documentation at home (will check
> > tomorrow).
> >
> > As far as I remember, Configuration Type 1 should be supported since
> > the earliest Intel chipset, Type 2 could vary.
> >
> 
> Thanks for that Martin, much appreciated.
> 
> I already have the configuration type down (it's 1), but the 430VX and
> also the VIA 585 seem only to report host bridges. I'm unable to spot
> the piece of code which does different PCI-related things for these
> chipsets in the kernel. Does anybody know if a workaround is applied?
> 
> Regards,
> Greg.

Martin has it backwards, I think.  The 486 chip sets and (maybe) some
of the early Pentium chip sets from Intel are the ones that used Type 2.

All the PPro and newer chip sets use Type 1 so far as I am aware.  Also,
I don't know of any chips that support both -- you get one or the other.

So Type 1 does sound right for the 430VX.

--Charles

Re: [OT] Intel chipset development documents

[Maciej W. Rozycki]

On Fri, 2 Nov 2001, Charles Marslett wrote:

> All the PPro and newer chip sets use Type 1 so far as I am aware.  Also,
> I don't know of any chips that support both -- you get one or the other.

 The i430LX/NX chipset supports both.  Docs for the chipset are still
available from the Intel site. 

-- 
+  Maciej W. Rozycki, Technical University of Gdansk, Poland   +
+--------------------------------------------------------------+
+        e-mail: macro@ds2.pg.gda.pl, PGP key available        +

Re: [OT] Intel chipset development documents

[Martin Mares]

Hello!

> I already have the configuration type down (it's 1), but the 430VX and
> also the VIA 585 seem only to report host bridges. I'm unable to spot
> the piece of code which does different PCI-related things for these
> chipsets in the kernel. Does anybody know if a workaround is applied?

It's quite strange -- can you send me 'lspci -vvx -MH1' output, please?

				Have a nice fortnight
-- 
Martin `MJ' Mares   <mj@ucw.cz>   http://atrey.karlin.mff.cuni.cz/~mj/
Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
"Beware of bugs in the above code; I have only proved it correct, not tried it." -- D.E.K.

Re: [OT] Intel chipset development documents

[Greg Sheard]

On Sat, 2001-11-03 at 11:04, Martin Mares wrote:
> Hello!
> 
> > I already have the configuration type down (it's 1), but the 430VX and
> > also the VIA 585 seem only to report host bridges. I'm unable to spot
> > the piece of code which does different PCI-related things for these
> > chipsets in the kernel. Does anybody know if a workaround is applied?
> 
> It's quite strange -- can you send me 'lspci -vvx -MH1' output, please?
> 

Well I would, but with the assistance of Martin Bligh from IBM I've
found my bug. Major thinko - there seems to be something funky in the
440BX chipset that allowed me to do something like:

  outl(PCI_CONF1_ADDRESS(bus, 0, dev, fn), 0xCF8);

but still returned the correct PCI information! The problem was that I'd
done my own inline for the addressing and, since I wasn't using reg, I'd
left it out. Somehow I'd managed to shift the important stuff around...

Thanks to everyone who's offered suggestions and documents, it's much
appreciated.

Cheers,
Greg.

Re: [OT] Intel chipset development documents

[Marc Haber]

On Fri, Nov 02, 2001 at 05:10:50PM +0000, Greg Sheard wrote:
> Hi all. Apologies for OT, but this seems the best place to ask.
> Intel have removed documentation for the 430VX (Triton) chipset from
> their developer site.

The famous Triton chipset was numbered 430_F_X. The 430VX was one of
low-cost successor.

Greetings
Marc

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Karlsruhe, Germany |  lose things."    Winona Ryder | Fon: *49 721 966 32 15
Nordisch by Nature |  How to make an American Quilt | Fax: *49 721 966 31 29

Re: [OT] Intel chipset development documents

[Robert Love]

On Fri, 2001-11-02 at 13:03, Marc Haber wrote:
> The famous Triton chipset was numbered 430_F_X. The 430VX was one of
> low-cost successor.

Actually, I believe all of the i430 series were "Triton" .. with Triton
I being the famous i430FX and Triton II being the lousy i430VX.

	Robert Love

Re: [OT] Intel chipset development documents

[Greg Sheard]

On Fri, 2001-11-02 at 20:31, Robert Love wrote:
> On Fri, 2001-11-02 at 13:03, Marc Haber wrote:
> > The famous Triton chipset was numbered 430_F_X. The 430VX was one of
> > low-cost successor.
> 
> Actually, I believe all of the i430 series were "Triton" .. with Triton
> I being the famous i430FX and Triton II being the lousy i430VX.
> 

According to some information I found on the releases (sadly nothing on
the datasheets) the HX was Triton II, and the VX was Triton III. Except
Intel then dropped the Triton name shortly after they released the VX.

Cheers,
Greg

Re: [OT] Intel chipset development documents

[Robert Love]

On Fri, 2001-11-02 at 16:07, Greg Sheard wrote:
> > Actually, I believe all of the i430 series were "Triton" .. with Triton
> > I being the famous i430FX and Triton II being the lousy i430VX.
> 
> According to some information I found on the releases (sadly nothing on
> the datasheets) the HX was Triton II, and the VX was Triton III. Except
> Intel then dropped the Triton name shortly after they released the VX.

Hm, you are right.  I forgot about the HX, which was the best chip in
the series IMO.  Anyhow, they are all Triton's :)

	Robert Love

Re: [OT] Intel chipset development documents

[Maciej W. Rozycki]

On 2 Nov 2001, Greg Sheard wrote:

> Intel have removed documentation for the 430VX (Triton) chipset from
> their developer site. I'm trying to check that the access details for
> the Southbridge are the same as for the 440BX chipset, since I'm working
> on some code for direct PCI access. If they're not, could somebody
> please let me have the relevant documentation?

 Get the ISA bridge specs from 
'http://developer.intel.com/design/intarch/datashts/290550.htm'.

-- 
+  Maciej W. Rozycki, Technical University of Gdansk, Poland   +
+--------------------------------------------------------------+
+        e-mail: macro@ds2.pg.gda.pl, PGP key available        +