Apparently 2.0.x kernels have problems with the VXtwo and VXpro+ boards based
on the UMC8886 and 8887 chipset. I have two boards with each chipset.
Here's the easiest way to tell which one you have without looking at the
board itself:
UMC8886 - bogomips will report as the clockrate without the multiplier,
usually 10MHz over. 50MHz reports as 50 to 60 bogomips. This
varies wildly from boot to boot.
UMC8887 - bogomips will generally report as the clockrate with multiplier,
minus the multipler/~2. So, a board running at 150MHz will
report back between 109 bogomips and 120 bogomips. A board at
133MHz (66*2) will report back between 90 and 100 bogomips. This
stays about the same with maybe 2 or 3 bogomips of difference at
each boot.
This appears to be a Cyrix-only issue, but I can't be sure. I don't have
any intel CPUs at home to test it with, only a Cx6x86 PR200L+ and Cx6x86
PR166L+. I recall seeing a patch somewhere for UMC8886/8887
identification, so it would be greatly appreciated if someone could
confirm or deny this, before I set to work on a patch of my own. It
would also be greatly appreciated if someone could test this with
an intel CPU, non-MMX preferrably. The 8886 board is an Amptron
PM-8600, so if any of you have one of those lying around, crank it
up, give it a shot. Thanks much! :)
-Phillip R. Jaenke [InterNIC Handle: PRJ5] (kernel@prj.pcimporters.com)
MIS Department, PC Importers, Inc. 800.319.9284, x4262
Head of Development, The Improvement Linux Project.
Penguin0: Cyrix Cx6x86 PR200L+, 64M, 1.6G, 1.2G, 4.3G, EE/Pro, 2.0.30
Penguin1: Intel P5-100, 32M, 4G, NE2k, 2.1.60-Improvement