Re: [Qemu-devel] [PATCH] target-ppc: Multiple/String Word alignment exception

[Laurent Vivier]

On 31/03/2016 09:15, Alexander Graf wrote:
> 
> 
> On 31.03.16 09:06, Laurent Vivier wrote:
>>
>>
>> On 31/03/2016 08:54, Alexander Graf wrote:
>>>
>>>
>>> On 31.03.16 01:29, David Gibson wrote:
>>>> On Wed, 30 Mar 2016 19:13:00 +0200
>>>> Laurent Vivier <address@hidden> wrote:
>>>>
>>>>> If the processor is in little-endian mode, an alignment interrupt must
>>>>> occur for the following instructions: lmw, stmw, lswi, lswx, stswi or 
>>>>> stswx.
>>>>>
>>>>> This is what happens with KVM, so change TCG to do the same.
>>>>>
>>>>> As the instruction can be emulated by the kernel, enable the change
>>>>> only in softmmu mode.
>>>>>
>>>>> Signed-off-by: Laurent Vivier <address@hidden>
>>>>
>>>> I guess this makes sense given the existing hardware behaviour, even
>>>> though it seems a bit perverse to me to make the emulator strictly less
>>>> functional.
>>>>
>>>> Alex, what do you think?
>>>
>>> In general we only implement strict checks if it breaks guests not to
>>> have them. Are you aware of any such case?
>>
>> No, it does not break anything. The idea was to have the same behavior
>> with TCG as with a real CPU (or kvm). But if it is not the rule, we can
>> drop this patch.
> 
> I guess if you really care about same behavior, we'd need to have risu
> ported to ppc. However, that's a huge can of worms. Once we start that,
> we'd have to verify risu against every single CPU type we support
> because they all interpret certain corner cases differently.

I didn't know risu: it seems to be a wonderful tool!
> 
> That's basically what David was trying to say with POWER9. How do you
> know that POWER9 still requires strong alignment checks for indexed LE
> instructions? If it doesn't, we'd have to add a case in TCG to not the
> the checks again. These multiply very quickly :).

I understand. So just forget this patch :)

Thanks,
Laurent