Given that I spent a lot of time getting a XFS enabled kernel merged
with -aa series kernel, I'd like to ask if anyone else has tried out the
-aa series without XFS? If so, and you have liked what you
see/use/whatever, I'd like to ask SGI if there is a way to work-out
between the XFS tree and -AA tree a possibly easier migration path.
I found a lot of things in the -AA patch I applied directly to a the
most recent XFS patched 2.4.17 kernel was that many parts of the -AA
patch that was rejected were in fact already in the XFS kernel to begin
with.
The biggest point here is that there were only about 12 reject files.
Some were very large >200 lines of replaced code(it was getting harry at
that point lemme tell you), but for the most part I found that many of
the VM functions which were changed were just that. Simple changes that
make a function or set of functions within the code do things in a
different order, rather than just raw rip-and-replace actions.
So, is it possible to do anything around this? The main reason I ask is
even though I'm not 100% sure I did everything 100% right, the kernel
has performed very well from a stability/memory usage standpoint. It
still uses a lot, but not ALL available RAM goes to cache now. This was
causing my eepro100 to throw console messages(card reports no
resources).
Problem was though, is it was all memory related. Now those problems
have gone away.
--
Austin Gonyou
Systems Architect, CCNA
Coremetrics, Inc.
Phone: 512-698-7250
email: austin@xxxxxxxxxxxxxxx
"It is the part of a good shepherd to shear his flock, not to skin it."
Latin Proverb
signature.asc
Description: This is a digitally signed message part
|