Re: [PATCH v2 06/13] s390x: protvirt: KVM intercept changes

[Janosch Frank <frankja@linux.ibm.com>]

[-- Attachment #1.1: Type: text/plain, Size: 2291 bytes --]

On 12/5/19 6:46 PM, Cornelia Huck wrote:
> On Thu, 5 Dec 2019 18:34:32 +0100
> Janosch Frank <frankja@linux.ibm.com> wrote:
> 
>> On 12/5/19 6:15 PM, Cornelia Huck wrote:
>>> On Fri, 29 Nov 2019 04:48:02 -0500
>>> Janosch Frank <frankja@linux.ibm.com> wrote:
>>>   
>>>> Secure guests no longer intercept with code 4 for an instruction
>>>> interception. Instead they have codes 104 and 108 for secure
>>>> instruction interception and secure instruction notification
>>>> respectively.
>>>>
>>>> The 104 mirrors the 4 interception.
>>>>
>>>> The 108 is a notification interception to let KVM and QEMU know that
>>>> something changed and we need to update tracking information or
>>>> perform specific tasks. It's currently taken for the following
>>>> instructions:
>>>>
>>>> * stpx (To inform about the changed prefix location)
>>>> * sclp (On incorrect SCCB values, so we can inject a IRQ)
>>>> * sigp (All but "stop and store status")
>>>> * diag308 (Subcodes 0/1)
>>>>
>>>> Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
>>>> ---
>>>>  target/s390x/kvm.c | 6 ++++++
>>>>  1 file changed, 6 insertions(+)
>>>>
> 
>>>> @@ -1664,6 +1668,8 @@ static int handle_intercept(S390CPU *cpu)
>>>>              (long)cs->kvm_run->psw_addr);
>>>>      switch (icpt_code) {
>>>>          case ICPT_INSTRUCTION:
>>>> +        case ICPT_PV_INSTR:
>>>> +        case ICPT_PV_INSTR_NOTIFICATION:
>>>>              r = handle_instruction(cpu, run);  
>>>
>>> I'm still a bit uneasy about going through the same path for both 104
>>> and 108. How does the handler figure out whether it should emulate an
>>> instruction, or just process a notification? Is it guaranteed that a
>>> given instruction is always showing up as either a 104 or a 108, so
>>> that the handler can check the pv state?  
>>
>> diag 308 subcode 0/1 are 108, but all other subcodes are defined as a
>> 104 (if they are an exit at all)...
> 
> I think that's a reason to really split 108 from 4/104, or at least add
> an parameter...

And still call the diag 308 handler or have separate handlers?

> 
>>
>>>
>>> [Even if that works, it still feels a bit unclean to me.]
>>>   
>>>>              break;
>>>>          case ICPT_PROGRAM:  
>>>
>>>   
>>
>>
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]