On 27/11/2007, at 7:59 AM, Daniel Ouellet wrote:
Hi,
Please correct me if my understanding is wrong. I am trying to trace
a bug and I am not sure where it is really, but I got a possible
idea I want to check for if that make sense.
My understanding's is that all drives are using an abstraction layer
between the kernel and the drivers itself.
Now, I don't know if there is a difference between drivers for a
single processor kernel and a multiple core kernel. So, first
question is if there is a difference or not. This for the SAS
drivers in Sun server.
If there is a difference there, then I thought for a test and trying
to isolate the source of the problem, or the section where it might
be in anyway by trying to compile a kernel for mp and only change
the driver use if that's even possible. May be a very stupid idea,
but how to go at finding where the source of the bug is, is my goal
and I thought that may be, if that's not totally stupid, I could try
that and see if it might be isolate there.
The only thing I am sure of is that so far I can reproduce the bug
at will with the mp and do not see if without, everything else stay
equal.
what is the bug you're able to reproduce?
As far as I can tell, it's either in the drivers for the SAS
drivers, or may be some part in the PCI drivers may be, but I can't
be sure either way.
Any clue stick would be welcome.
So far I am trying to compile reduce kernel to make sure there isn't
any part of it that might create the problem, but no success so far.
I would love to find this bug, or isolate it more, but so far, all
my tries are not successful and I try to isolate the differences
between the mp kernel and the single one, however, I am not sure of
the differences between them to know where to look yet.
Thanks
Daniel
