Re: Cyrix patch : Proposal?

Andre Derrick Balsa (andrewbalsa@usa.net)
Fri, 07 Nov 1997 21:13:50 +0100


Hi Nigel,

Nigel Metheringham wrote:
>
> linker@nightshade.z.ml.org said:
> } A very good post. I would assume that you approve of minor patching so
> } that cpuinfo displays the correct information?
>
> If this was done then userlevel tools could twiddle happily knowing that
> they had the right CPU to play with.

This is correct, but it's not a good enough reason to implement a patch:
it's much easier to write a good user-level utility that will correctly
detect the CPU type, revision, speed, etc... than to put all of this in
the kernel.

IMHO there are 2 valid reasons for kernel patching when it comes to
Cyrix CPUs:

1) If the kernel will either not boot at all, or will boot but setup its
working parameters incorrectly (e.g. wrong bogomips/udelay settings may
cause problems with some drivers that use udelay for timing loops). This
was a point I discussed at length with Pavel Machek.

2) There is a feature that *must* be set in the kernel, because it
affects basic kernel operation (e.g. VSPM, but I have already explained
that VSPM brings nothing in terms of performance, and is a one-of-a-kind
implementation of paging).

These two issues are adressed by a very short patch available at:
http://www.tux.org/~balsa/linux/cyrix/Cyrix.patch
This patch adds less than 50 bytes to the executable and about a dozen
lines to the source code. It's totally inocuous on non-6x86 systems.

I would like to have more linux users download and test it. It will
patch against 2.0.29, 2.0.30 and 2.0.31. I haven't tried patching
against 2.1.6x, but it should patch OK with little or no changes, since
it's so simple. It's a subset of the 2.1.39 patch with some small
modifications.

I am still trying to sort out a nasty oops that occurs on my 6x86MX rev.
1.3, but the above patch also contains a one-line workaround for the
problem. As soon as the cause of the oops will have been exactly
established and an adequate solution will have been found, I'll propose
a patch for inclusion in the VGER kernel source.

Cheers,

========================================================
Andrew D. Balsa
Home Page: http://www.tux.org/~balsa
andrewbalsa@usa.net
========================================================