On Wed, 19 Jan 2000, Marty Connor wrote: | |It's probably the PCI signature of the image. Edit your src-32/Config |file, and go to about line 123: | |# Newer 90xC revision: |# 0x9200 : 10/100 TPO (3C905C-TXM) |90X_SIGNATURE=0x9200 This was mentioned on the list a bit ago. My drivers were patched seconds later. =) Of course, since it would explain the problem, I will check it again. | |Did you happen to change the PCI _slot_ the card was in? Windows gets |really silly about that. Go to the SYSTEM Control Panel in Windows, and |go into the Device Manager, and DELETE the card from your system. Then |reboot, and let Windows re-install the driver for the card. I have seen |this behavior when I change the slot of a PCI card under Windows. |Sometimes you just have to do this deletion and re-discovery/installation |ritual for no discernable reason. Primitive, that Windows :-) No, my lab is rsynced from a single master copy, so, for example, all netcards have to be in PCI slot 2, all video cards in slot 3, etc... otherwise windows won't find them... You're right, it's primitive. =) However, I'm familiar with that problem, so I deleted the 3c90x device, powered down, and reinstalled. This is the normal solution for this problem. No go. To simplify a long and hairy debug process, it seems that either there's something causing these cards not to get recognised by the motherboard bios (Linux will iteratively ask the PCI bus, so would find card anyway), or the brand new driver from 3com doesn't work. Reasons to believe it's a bios problem: None of the 3c905C-TXM cards ever provided me with the 3com managed-boot message, even though the motherboard is set to boot network devices first. My 3c905B-TXM gives me 3com's little bootmenu in this configuration. Reasons to believe it's a driver problem: The new driver disk apparently contains 'universal' 3c905 drivers, but installing using these drivers gives me the same problems on my otherwise-working 3c905Bs. Symptoms (which are the same on 905B and C) are that Windows thinks everything is just fine, except that a tcpdump shows _NO_ packets sent from this machine, ever. Backing down to the older driver, that doesn't know about 905Cs gives me a working 905B installation. Every one of my lab machines shows the identical symptoms. | |>Anyway, sorry for the offtopic post. | |Not at all. Someone perusing the archives will benefit from all these |public posts. You do us all a service by allowing us to debug in public. Ok, for the sake of the archives, then... =) | |>*I updated this utility to allow me to verify checksums. | |Excellent! Please send Ken or myself the changes. Your work is much |appreciated. Ok, changes will be mailed as soon as I clean them up a bit... I currently make two bad assumptions that are safe in my environment. It _was_ a quick'n'dirty hack, after all... =) | |Regards, | |Marty | |--- | Name: Martin D. Connor |US Mail: Entity Cyber, Inc.; P.O. Box 391827; Cambridge, MA 02139; USA | Voice: (617) 491-6935, Fax: (617) 491-7046 | Email: mdc@thinguin.org | Web: http://www.thinguin.org/ | | -- ========================================================================= Robert Thompson <thompson@cs.swau.edu> Southwestern Adventist University "My opinions are my own. Want one?" Department of Computer Science =========================================================================== This Mail was sent to netboot mailing list by: Robert Thompson <robertt@stargate.cs.swau.edu> To get help about this list, send a mail with 'help' as the only string in it's body to majordomo@baghira.han.de. If you have problems with this list, send a mail to netboot-owner@baghira.han.de.
For requests or suggestions regarding this mailing list archive please write to netboot@gkminix.han.de.