Re: 3.0.3 on Durango: some problems
Thu, 14 Nov 1996 11:03:01 -0500

>>> geerten kuiper said:
> Yesterday I finally put together my new hardware, and installed RedHat 3.0.3
> (using a PC164 kernel from 4.0, and apropriately new MILO/LinLoad: thanks to
> Wolfgang Dreyer). Only problem during install was, I had to do mke2fs from
> the second VC by hand.
> So now I've got it running, but some problems/oddities are remaining.
> - the firmware clock had not updated the date this morning. Time seems to be
> OK though.
> - MILO complains about not detecting the NVRAM. (I do not know if this is a
> problem)

THIS IS NOT A PROBLEM! It simply means that MILO saw no evidence that you'd
ever put anything into NVRAM using MILO, like environment variables.

> - Mouse: Firmware sees a serial mouse on COM2, which is correct. Kernel
> claims it sees a PS/2 mouse, which is not correct. In the XF86Config I set
> up the serial mouse, but it doesn't work: when I move it the pointer stays
> in place for a while (many seconds, maybe a minute) then suddenly jumps to a
> new position. I have tried replacing the mouse, but to no avail.

Kernel doesn't really probe for a PS/2 mouse, just kinda assumes its there
if configured in.

For serial mice, you *MUST* have the protocol correct or you *will* see
bizarre behaviour.

> - Ethernet: I have an ACCTON NE2000-clone in one of the ISA slots, and the
> kernel should see it as "eth0: NE2000 found at 0x300, using IRQ 5." It does,
> but only occasionally (one time out of five or worse) The other times it
> says "NE*000 ethercard probe at 0x300: failed to detect IRQ line."
> I have two of those cards, both work fine in the Pentium, both fail in the
> Alpha, so I guess there's nothing wrong with the card itself.

NE2000's are known to work, but also known is some uncertainty about all
the clones.

> - There are some other things that work sometimes, but more often do not:
> Ctrl-Alt-Del or "shutdown -r now" may cycle through MILO and reboot, but
> usually hangs.

This may be fixed by going to the latest MILO (test-images/ directory).

> "date" may show the right date, but usually does not. (even after setting
> CLOCKMODE="ARC") (it is consistent from boot to shutdown.)

Flakey clock reads on the PC164 are a known problem.

> - Doing "time" will get the Elapsed time wrong. (I believe this one has been
> on the list before.)

A matter of using the latest LIBC and utilities.

> As some of these problems seem to be related to interrupts, I checked
> /proc/interrupts.
> It shows
> 1: keyboard
> 5: NE2000 (only when the card was detected during boot)
> 19: 53c8xx
> no serial, parallel ports, no floppy (floppy seems to work fine though)

It'll only show them if there have been some since you booted.

> /proc/pci shows:
> Bus 0, device 8, function 0:
> Non-VGA device: Intel 82378IB (rev 67).
> Medium devsel. Master Capable. No bursts.
> What's the funny device 8

This is the PCI-to-ISA bridge chip, needed for the ISA slots and devices.


American Non Sequitur Society: we don't make sense, but we do like pizza...

Jay A Estabrook Alpha Migration Tools - LINUX Project
Digital Equipment Corp. (508) 952-4202 or (DTN) 227-4202
151 Taylor Street - TAY1-2 enet:
Littleton, MA 01460-1407 decnet: tallis::jestabro

To unsubscribe: send e-mail to with
'unsubscribe' as the subject.  Do not send it to

Feedback | Store | News | Support | Product Errata | About Us | Linux Info | Search | JumpWords
No Frames | Show Frames

Copyright © 1995-1997 Red Hat Software. Legal notices