On Sat, 8 Jan 2000, Dan Hollis wrote:
> > The problem I think is that this isn't a true MVP3.
> You do know that the MVP3 lets you change the PCI IDs through software? It
> seems some BIOS definitely do this, changing the ID of the southbridge to
> an older controller ID for some unexplainable reason.
Yes, I'm aware of that possibility, but even chips that are marked
598 on the board on hybrids don't act the same way a real 598 does.
Programming them as a 597 doesn't solve any problem. In fact it creates
them. Known issues with 598s pretending to be 597s include..
o Unreliable (or usually completely broken) AGP x2
o Failure to bus-master cards which work fine in other systems
with true MVP3 chipset.
More confusing, the 598 and 597 contain a second PCI->PCI bridge device.
Here are the real IDs for this 2nd device:
82C598 - 8598
82C597 - 8597
Guess which ID the hybrid has? 8598.
So to sum up, we have a chip that contains an old northbridge
incapable of AGP x2 (and other problems) identifying itself as
an older bridge, incorporating a second device which is really from
the MVP3 chipset. Brain damaged?
It makes my head hurt just thinking about this stuff.
This combination of north/south bridge parts doesn't seem to be supported
by VIA in any of their documentation, and I really cannot understand the
reasons for including such a strange combination of chipset parts.
I've noticed most of these hybrid chipset combos tend to be from
manufacturers such as PC-Chips. None of the big-name super7 vendors
(such as FIC for example) produce such wierdness. It makes me wonder
whether or not there are some taiwanese companies out to save some money
in production costs by including sub-standard parts.
I've been in touch with VIA about this sometime ago when I first
encountered such wierdness during developing powertweak-linux, but it
seems their technical support email doesn't get read.
> The only way to know for sure is to look on the motherboard itself. The
> data from software can't be trusted.
It can be determined from the revision of the 586 whether it's true MVP3
or not. This field afaik cannot be reprogrammed in software, and yields
0->3 on vp3, 4 on mvp3, and 41 on hybrid mvp3.
(Theres also a rev 40 I think, but I've not seen this personally)
normally a 586 & 597 == a VP3, but if it's a rev 41 586, it's hybrid MVP3.
-- Dave.- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Jan 15 2000 - 21:00:13 EST