[Gllug] Kernel source on Red Hat

[Vincent AE Scott]

Dave Jones(davej at suse.de)@Thu, Jun 06, 2002 at 04:03:17PM +0200:
> 
>  > but it's performance is heavily dependant on having host and UML using /tmpfs.
> 
> Hmm, another way could be NFS exporting the host and mounting it inside
> the UML instance perhaps.

yep, thats certainly another way i have considered using it.  but i'm
less than convinced of using linux as an NFS server.  The last time i
tried it ( a while ago now ) it was far less performant than a solaris
box.


>  > and nope, theres currently no direct hardware access.
> 
> Something that could be /very/ interesting, would be valgrind+UML.
> a. It could find leaks and the like in kernel code.
> b. Valgrinds virtualisation of the CPU (whilst slowing things down
>    slightly) could be useful for trapping hardware poking, and
>    redirecting it to the host kernel.
>    Eg, a cpuid instruction gets transformed by uml into an open() &
>    read() of /dev/cpu/0/cpuid 
> 
> Jeff has been talking with the valgrind author, so this might come
> to light one day..

yep, that would be way cool.  hell for some simple stuff, i'd imagine
you could even have a config file which mapped certain get() type calls
to read's from /proc.  that way it'd be easy to add on new support in an
instance.  <g> you could even have an XML config file...


>  > IIRC, UML is being integrated into the 2.5 tree, dave?
> 
> Jeff last pushed it for inclusion circa 2.5.5 iirc. I'm not sure
> how up to date it is wrt the current tree, but it'd be nice to see
> it merged at some point.

cool.  UML does appear to have almost daily patches[1] tho.  it seems to
get fixed and patched almost continuously.  not a problem for those that
compile it from scratch and keep tabs on its development.  far more of
an issue for those that download pre-compiled binaries, which often lag
far behind the source releases.  a common thing on the UML list whenever
someone finds a bug in the UML bins is to 'upgrade to the latest
version'.  which is both nice, as its always being improved, but bad
from a mass adoption standpoint.

-v

[1] well, i only ever look at the homepage every few weeks, and i'm
always several versions behind, with plenty of fixes and features having
been added.

-- 
keys:  http://codex.net/pgp/gpg.asc http://codex.net/pgp/pgp.asc

 Keyboard error. Hit F1 to resume. 


-- 
Gllug mailing list  -  Gllug at linux.co.uk
http://list.ftech.net/mailman/listinfo/gllug