Re: [Xen-devel] [PATCH V2 2/3] xen/vm_event: Support for guest-requested events

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 24.06.15 at 16:56, <rcojocaru@xxxxxxxxxxxxxxx> wrote:
> On 06/15/2015 12:03 PM, Razvan Cojocaru wrote:
>> Added support for a new class of vm_events: VM_EVENT_REASON_REQUEST,
>> sent via HVMOP_request_vm_event. The guest can request that a
>> generic vm_event (containing only the vm_event-filled guest registers
>> as information) be sent to userspace by setting up the correct
>> registers and doing a VMCALL. For example, for a 64-bit guest, this
>> means: EAX = 34 (hvmop), EBX = 24 (HVMOP_request_vm_event).
>> 
>> Signed-off-by: Razvan Cojocaru <rcojocaru@xxxxxxxxxxxxxxx>
>> 
>> ---
>> Changes since V1:
>>  - Using "curr" consistently in hvm_event_requested().
>>  - Renamed VM_EVENT_REASON_REQUEST to VM_EVENT_REASON_GUEST_REQUEST.
> 
> In the meantime, my colleague Corneliu Zuzu working on an ARM-related
> project with Xen has tested my patch and discovered that some registers
> get clobbered when Xen is compiled in debug mode and a GUEST_REQUEST
> event is being sent out.
> 
> Here's the ARM part, in do_trap_hypercall(), xen/arch/arm/traps.c:
> 
> 1399 #ifndef NDEBUG
> 1400     /*
> 1401      * Clobber argument registers only if pc is unchanged, otherwise
> 1402      * this is a hypercall continuation.
> 1403      */
> 1404     if ( orig_pc == regs->pc )
> 1405     {
> 1406         switch ( arm_hypercall_table[*nr].nr_args ) {
> 1407         case 5: HYPERCALL_ARG5(regs) = 0xDEADBEEF;
> 1408         case 4: HYPERCALL_ARG4(regs) = 0xDEADBEEF;
> 1409         case 3: HYPERCALL_ARG3(regs) = 0xDEADBEEF;
> 1410         case 2: HYPERCALL_ARG2(regs) = 0xDEADBEEF;
> 1411         case 1: /* Don't clobber x0/r0 -- it's the return value */
> 1412             break;
> 1413         default: BUG();
> 1414         }
> 1415         *nr = 0xDEADBEEF;
> 1416     }
> 1417 #endif
> 
> Would it be fair to say that HVMOP_request_vm_event should be an
> exception to this rule (i.e. something along the lines of "if (
> unlikely(r12 == HVMOP_request_vm_event) && orig_pc == regs->pc )",
> etc.)? Even in debug mode, a GUEST_REQUEST vm_event should keep the
> state of the guest intact.

It's a hypercall, and hence should follow hypercall rules. If the
guest wants certain registers preserved, it should do so itself.

Jan


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