Dear Uwe,
Sorry for there is no errata sheet. Maybe I can add more PCI IDs for our chipset first. I can provide some recommend setting written at internal document, if needed.
Morgan
----- Original Message ----- From: "Uwe Hermann" uwe@hermann-uwe.de To: "Morgan Tsai /SiS" my_tsai@sis.com Cc: "bari" bari@onelabs.com; "Corey Osgood" corey.osgood@gmail.com; linuxbios@linuxbios.org Sent: Friday, November 09, 2007 2:10 AM Subject: Re: [LinuxBIOS] Walmart/Everex GPC TC2502 (VIA CN700 +VT8237RChipset), PC2500E
Hi Morgan!
On Thu, Nov 08, 2007 at 10:52:19AM +0800, Morgan Tsai /SiS wrote:
SiS965(L) vs 966(L)
Identical device ID: LAN, USB Different device ID: SATA, Audio(966 has HD Audio, but still reserved AC97)
Although they are funtionally alike, such like register map. There is somewhare different value to be filled for bug fix.
That doesn't sound too bad, IMHO. If the only real differences are PCI IDs of some devices and small changes in some registers, then the 966(L) code could easily be adapted to also work with 965(L), correct?
Adding more PCI IDs is trivial and we can do that without datasheet access I think, but the "different value to be filled for bug fix" could be a problem if it's not documented publically...
Is there some errata sheet we can look at and/or do you know how to fix the 966(L) code in svn to make it work for 965(L), too?
Thanks, Uwe.