On Tue, 18 Oct 2005 22:01:58 -0400 Lawrence Teo <[EMAIL PROTECTED]> wrote:
> Ken Gunderson wrote: > > Hello All: > > > > I'm having some issues w/an Intel PRO/1000 MT Dual Port Server Adapter > > in an OBSD-3.7 box. The card is in the pci-x riser on one of > > these puppies; > > > > <http://www.tyan.com/products/html/gt20b5151.html> > > > > Dmesg complains the "The EEPROM Checksum Is Not Valid". I've tested > > the card in a Winblows machine and it works. Any insights as to what > > is going on here greatly appreciated. > > I experienced the same problem with that card. And yes, it's driving me > nuts too. :) > > What's strange is that I have two older cards (same model) and they > work fine with OpenBSD. > > I am testing the problematic card in a Dell Precision 380 workstation > (no PCI-X riser here :)). > > dmesg snippet: > > em0 at pci5 dev 2 function 0 "Intel PRO/1000MT (82546GB)" rev 0x01: irq > 3em0: The EEPROM Checksum Is Not Valid > em0: Unable to initialize the hardware > em1 at pci5 dev 2 function 1 "Intel PRO/1000MT (82546GB)" rev 0x01: irq > 9em1: The EEPROM Checksum Is Not Valid > em1: Unable to initialize the hardware > > Full dmesg (slightly dated) at: > > http://marc.theaimsgroup.com/?l=openbsd-misc&m=112664653901887&w=2 > (see "Problem 3") > > At first, I thought the problem could be resolved by using the latest > Intel drivers. OpenBSD recently sync'ed the em(4) driver with Intel's > latest, and I just tried a recent October snapshot containing that > driver. Unfortunately, the problem is still there. Ditto. > If you still have the packaging of your card, could you please > let me know the date and version of your card? I'm just trying to see > if there's a certain batch of cards exhibiting that problem. Mine are: > > Date: 10/18/2004 > Version: A85400-006 God, I'll see if I can dig it up. Other parts were back ordered so the card has been on hand a good month or two. The sku #'s on the white label are: 0004239EFB93 463AD C41421 - 003 > Also, a kind soul on the Internet sent me the following link showing > how it can be resolved with FreeBSD, but I haven't had time to > investigate it in detail yet. > > http://lists.freebsd.org/pipermail/freebsd-mobile/2004-July/004302.html Thanks for the link but I'd also tried this w/FBSD prior to posting. > If you do find a solution, please let the list know. I'll do the same. > Thanks! np. I've resubmitted to Intel tech support via email and this time even rec'd an automated reply referring me to all the pdf docs that I'd already rummaged through and advising "in the unlikely event that we were unable to reply with a document that pertains to your inquiry, please use one (and only one) of the following two methods to escalate your issue to Intel Technical Support..." please via ioblah, blah... So I did and rec'd. yet another automated reply informing me: "Due to inbound e-mail volumes, you should expect a response to your question within two business days". Good to know... Pretty encouraging. All in html'ized email of course. Glad I didn't spring for the 4 port version.... I've also got a tech at Tyan looking into it. Subsequently I have tried the card in an Opteron based Tyan Transport TA26 w/same results. So just when I was thinking it was something particular to the GT20/Intel Pro 1000 combo the same problem on totally different beast. About only thing in common is the name Tyan and that it's on a pci riser... So, methinks Intel is up to their usual tricks and changed something on the card w/o telling us. In any event, looks more and more to be pointing to a driver issue? Bottom line is that, for now at least, others should probably want to stay away from these cards. Sutff I've tried: OBSD-3.7 OBSD-3.8 snap FBSD-5.4 release (w/FBSD driver from Intel's site) FBSD-6.0RC1 Ubuntu 3 different set of hardware: PCI-Riser: Tyan GT20 (B5151) Tyan TA26 (B2882) Legacy BX board Abit BX6r2 in PCI slot Same results. XP works, anything else doesn't. This thing appears to be totally foobarred.... -- Best regards, Ken Gunderson Q: Because it reverses the logical flow of conversation. A: Why is putting a reply at the top of the message frowned upon?
