#hpa@transmeta.com (H. Peter Anvin) writes:
#
#> Also, we should take the intersection of all the capabilities in an
#> SMP system.
#Aren't the capabilities of all processors identical in an *S*MP
#system?
Well, that depends. They SHOULD be. By Intel 'standards' *laff*, they have
to be identical stepping, indicating identical features. However, I've
seen it with an earlier rev CPU as cpu0, and a later one as cpu1 on a dual
board, working fine. Here's how to do it, extremely simplified.
SMP?
->yes
->check features on cpu0
->do NOT check features on cpu1
->assign features on cpu0 to cpu1 variables
->no
->check features on cpu0
done
Basically, this allows you to use two different CPUs, if the board will
allow it, and keep the OS from oopsing because of the mismatched features.
Seen it in action, and the way the person implemented it worked VERY well.
I'll see if I have a current email and get a diff for everyone if I can.
:)
-Phil R. Jaenke (kernel@nls.net / prj@nls.net)
TheGuyInCharge(tm), Ketyra Designs - We get paid to break stuff :)
Linux pkrea.ketyra.INT 2.0.33 #15 Sat Apr 18 00:40:21 EDT 1998 i586
Linux eiterra.nls.net 2.1.98 #15 Fri May 1 18:21:00 EDT 1998 i586
- Linus says for 'brave people only.' I say 'keep a backup.' - :)
! I reserve the right to bill spammers for my time and disk space !
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu