Re: kernel inclusion planned?

"Also sprach Stephan von Krawczynski:"
> The problem at my side is: I do not have spare boxes available for such tests.
> Every now and then I get my hands on some Xeon or Opteron boxes with 4-8 GB
> ram and raid1 or raid5 disks.

It's likely to be a problem in itself, since I do not have that
architecture (I have recently got a UP AMD 64 bit machine working,
using spare parts and ebay ...).


> But they do not stay particularly long so I cannot perform real testings. What
> I do is install the stuff based on 2.4 smp kernel (last available from

I haven't worked on 2.4 for a long time, but I presume it has stayed
stable from when I did work on it :).  I assume you are running 32 bit
xeons (if that is possible?) since I never did any 64 bit development
for 2.4.  It's only fairly recently, with the assembly of my 64 bit AMD
machine from scrounged parts, that I have been able to debug 64 bit at
all in 2.6.  Previously I was able to check the compilation (on
sorceforge's compile farm) but never to do any dynamic testing or
debugging.

I don't understand how opterons can run on 2.4. I didn't realize the
64 bit AMD architecture was supported there.  I have never tried it.
Now you tell me that it is possible I can go and try it and debug.

Running so much memory also puts you in a range where I cannot test - I
don't have that much memory. Once you have that much, questions of
"lower" and "upper" memory mappings begin to be more significant, and I
have never been able to test that. Memory for buffers should come
preferentially from "lower" memory, but under stress, upper memory may
be supplied, requiring more special treatment. That may be a problem.

If you give me a login - better, arrange to loan me equipment, or
arrange to have me come and visit and run on your equipment in situ, 
then I can do more.

Peter