Re: Cyrix 6x86MX and Centaur C6 CPUs in 2.1.102

Mike Jagdis (
Thu, 21 May 1998 13:03:00 +0100 (GMT/BST)

On Wed, 20 May 1998, Martin Mares wrote:

> > c) The code that shows Cyrix CPU steppings does not conform to Cyrix
> > stepping numbering. 6x86MX steppings are documented as 1.3, 1.4, 1.5,
> > 1.6, etc... in both the Cyrix/NS and IBM documentation, _not_ 0.3, 0.4,
> > 0.5, 0.6 as the present Linux kernel code would have it.
> >
> > A similar mistake is done for K6 steppings, which in the AMD
> > documentation are described as steppings A, B, C.
> >
> > This causes confusion with users that do "cat /proc/cpuinfo" and can't
> > figure out what CPU stepping they have. So I wish you could get it
> > right.
> I didn't implement it in the original CPU detection code as it seemed very
> hairy to me and I didn't like to substantially enlarge the code by lots of
> weird stepping naming conversions.
> Anyway, just send me a patch :-)

The problem is that the step/revision numbers that are printed on
the chips use different numbering/conventions from is actually
programmed in to the silicon. The detection code uses the numbers
in the silicon. If you want to go from that to what is printed
on the chip you need to put look up tables/conversion code in
the kernel to do it - then you need to watch for new CPU revisions
(which may not even be announced, never mind news worthy) and update
the kernel code as necessary before people start asking why it
misdetects their CPU. This is not good...

Better would be to add the mappings in README files. Complaints
should be forwarded to the CPU vendors. A kernel is not really
the place to work around the fact that their marketing department
can't get together with their engineers :-).


|  Mike Jagdis                  |  Internet:   |
|  Roan Technology Ltd.         |                                      |
|  54A Peach Street, Wokingham  |  Telephone:  +44 118 989 0403        |
|  RG40 1XG, ENGLAND            |  Fax:        +44 118 989 1195        |

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to