Re: [Qemu-devel] [PATCH for-1.6] isapc: disable kvmvapic

[Paolo Bonzini]

Il 04/10/2013 18:52, Andreas Färber ha scritto:
> Am 04.10.2013 18:47, schrieb Paolo Bonzini:
>> Il 26/08/2013 18:17, Michael S. Tsirkin ha scritto:
>>> On Mon, Aug 26, 2013 at 05:17:07PM +0200, Andreas Färber wrote:
>>>> Am 26.08.2013 17:01, schrieb Paolo Bonzini:
>>>>> Il 26/08/2013 16:34, Andreas Färber ha scritto:
>>>>>> Am 26.08.2013 14:21, schrieb Paolo Bonzini:
>>>>>>> Il 13/08/2013 00:02, Paolo Bonzini ha scritto:
>>>>>>>> vapic requires the VAPIC ROM to be mapped into RAM.  This is not
>>>>>>>> possible without PAM hardware.  This fixes a segmentation fault
>>>>>>>> running with -M isapc.
>>>>>>>>
>>>>>>>> Cc: address@hidden
>>>>>>>> Signed-off-by: Paolo Bonzini <address@hidden>
>>>>>>>> ---
>>>>>>>>
>>>>>>>> diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
>>>>>>>> index a19e172..3aa244a 100644
>>>>>>>> --- a/hw/i386/pc_piix.c
>>>>>>>> +++ b/hw/i386/pc_piix.c
>>>>>>>> @@ -743,6 +743,11 @@ static QEMUMachine isapc_machine = {
>>>>>>>>      .init = pc_init_isa,
>>>>>>>>      .max_cpus = 1,
>>>>>>>>      .compat_props = (GlobalProperty[]) {
>>>>>>>> +        {
>>>>>>>> +            .driver   = "apic-common",
>>>>>>>> +            .property = "kvmvapic",
>>>>>>>> +            .value    = "off",
>>>>>>>> +        },
>>>>>>>>          { /* end of list */ }
>>>>>>>>      },
>>>>>>>>      DEFAULT_MACHINE_OPTIONS,
>>>>>>>>
>>>>>>>
>>>>>>> Ping.
>>>>>>
>>>>>> I had double-checked that my pending qom-test covers isapc and succeeds
>>>>>> with and without your patch, so apparently this is visible at runtime 
>>>>>> only.
>>>>>
>>>>> Yes, it only happens after kvmvapic is actually enabled by the option ROM.
>>>>
>>>> Maybe Michael can pick it up along with the compat refactorings?
>>>
>>> Sure, I'll put it on my tree but I already prepared a pull request
>>> for today, it will go into the next one.
>>
>> Michael, can you post this patch together with the ACPI pull request?
> 
> What about the comment wrt s/kvmvapic/vapic/g? Haven't seen any response
> or v2.

It needs a v2 indeed... weird...

Paolo