Re: [Xen-devel] [PATCH] vm_event: Implement ARM SMC events

[Corneliu ZUZU <czuzu@xxxxxxxxxxxxxxx>]

On 4/13/2016 1:52 PM, Julien Grall wrote:
> Hello Corneliu,
>
> On 13/04/16 09:55, Corneliu ZUZU wrote:
> > On 4/12/2016 8:24 PM, Tamas K Lengyel wrote:
> > Another issue came to my mind: "HVC #imm", if handled through the
> > hvm-ops code, currently requires setting other registers to predefined
> > values before the HVC is actually issued. That would imply additional
> > effort to save/restore those registers if an external privileged domain
> > would want to set guest breakpoints. Given that, if we were to use HVC
> > for sw-bkpts, IMO it would be nice if the hvm-ops code architecture
> > would be slightly changed such that -lone- "HVM #imm" calls would be
> > achievable for some use cases, such as this.
> That is not true. All the hypercalls are using the same #imm 
> (XEN_HYPERCALL_TAG = 0xEA1), which indeed requires specific value in 
> various registers to differentiate the HVM-ops.
> It's up to us to define the requirements for the other #imm. For 
> instance there are some #imm allocated for debugging (see 
> do_debug_trap) that will dump the content of the registers.
That's nice, I wonder why we didn't do that for guest-request vm-events...

> > > So what I will do instead of issuing a software_breakpoint vm_event
> > > for SMCs, I'll introduce a new type, say
> > > VM_EVENT_REASON_PRIVILEGED_CALL, that can be used to forward both
> > > hypercalls and SMCs to a monitoring guest. This would also allow us to
> > > use the software_breakpoint type for the actual software breakpoint
> > > events in the future.
> > >
> > > Tamas
> > Isn't the HVC-part already achieved by guest-request vm-events? Maybe
> > tying this vm-event specifically to SMC (in which case the name could be
> > something like VM_EVENT_REASON_SECURE_CALL) and thus making it
> > ARM-specific would avoid that redundancy?
> I would create 2 distinct events for HVC and SMC. It would make easier 
> to differentiate them in userspace.
> Regards,
As stated, HVC events are already implemented as guest-request vm-events 
(hence the redundancy that a VM_EVENT_REASON_PRIVILEGED_CALL would add).
So indeed, that would leave us with the option of simply creating 
another event for SMC.
Corneliu.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel