RE: [Xen-ia64-devel] Meeting Summary taken from Xen-ia64 Next StepsDiscussion during Xen Summit

["Dong, Eddie" <eddie.dong@xxxxxxxxx>]

Tristan Gingold wrote:
> Le Vendredi 20 Janvier 2006 15:51, Dong, Eddie a écrit :
>> Thanks for the summary and one more thing:
>>      We are all agreed the mechanism to send event channel should use
>> callback/failsafe eventually instead of using a pseudo physical IRQ,
>> as the latter one has some potential issues.
>>      Actually same issues happen in PPC side too.
>> Thanks Dan, Thanks Ian and Keir for these great results.
>> Eddie
> Hi,
> 
> just a beginner question: what is the callback/failsafe mechanism ?
> I am still planning to use IRQ for IOSAPIC virtualization.  Maybe I
> can move to event channel as a second step and then to cb/fs after.
> 
> Thanks,
> Tristan.

Tristan:
        In Xen/X86, the event channel mechanism is based on callback, see 
event_callback_cs/event_callback_eip fields in  arch.guest_context of VCPU. 
When an event channel is detected in (x86_32/entry.s) test_all_events, the 
TRAPBOUNCE created is based on above cs:eip. Current Xen/IA64 uses pseudo 
physical IRQ that is OK now but has potential issues like mask mechanism, 
priority etc. 
        Yes, it can be a later after your current virtual interrupt controller 
stuff :-)
        Failsafe is used for guest application fault, it may be not implemented 
in Xen/IA64 yet :-(
Eddie

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel