Re: [XenPPC] Overflow in decrementer restore

[Jimi Xenidis <jimix@xxxxxxxxxxxxxx>]

On Sep 1, 2006, at 4:28 PM, Segher Boessenkool wrote:

> > The problem turned out to be that we don't sync the timebases between
> > the processors.
> Does this happen on JS2x?  SLOF is supposed to ensure that all
> CPUs start with the timebases already synchronised.
Not sure how to interpret this but linux _does_ spit out a lot of  
output about sync'ing timebase, tho it could be a no-op
Here is the Linux output from a JS21 SLOF: JS2XFW6331

Processor 1 found.
Synchronizing timebase
Got ack
score 299, offset 1000
score 299, offset 500
score 299, offset 250
score 299, offset 125
score 299, offset 62
score 299, offset 31
score 299, offset 15
score 217, offset 7
score -181, offset 3
score 63, offset 5
score -93, offset 4
Min 4 (score -57), Max 5 (score 31)
Final offset: 5 (25/300)
Processor 2 found.
Synchronizing timebase
Got ack
score 299, offset 1000
score 299, offset 500
score 299, offset 250
score 299, offset 125
score 299, offset 62
score 299, offset 31
score 299, offset 15
score 189, offset 7
score -169, offset 3
score 19, offset 5
score -79, offset 4
Min 4 (score -69), Max 5 (score 13)
Final offset: 5 (27/300)
Processor 3 found.
Synchronizing timebase
Got ack
score 299, offset 1000
score 299, offset 500
score 299, offset 250
score 299, offset 125
score 299, offset 62
score 299, offset 31
score 299, offset 15
score 287, offset 7
score -89, offset 3
score -9, offset 5
score 133, offset 6
score -41, offset 5
Min 5 (score 3), Max 6 (score 101)
Final offset: 5 (19/300)
Brought up 4 CPUs



-JX

_______________________________________________
Xen-ppc-devel mailing list
Xen-ppc-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ppc-devel