Re: [Qemu-devel] [kvmtool test PATCH 22/24] kvmtool: arm64: Add support for guest physical address size

[Marc Zyngier]

Hi Eric,

On 05/07/18 14:46, Auger Eric wrote:
> Hi Marc,
> 
> On 07/05/2018 03:20 PM, Marc Zyngier wrote:
>> On 05/07/18 13:47, Julien Grall wrote:
>>> Hi Will,
>>>
>>> On 04/07/18 16:52, Will Deacon wrote:
>>>> On Wed, Jul 04, 2018 at 04:00:11PM +0100, Julien Grall wrote:
>>>>> On 04/07/18 15:09, Will Deacon wrote:
>>>>>> On Fri, Jun 29, 2018 at 12:15:42PM +0100, Suzuki K Poulose wrote:
>>>>>>> Add an option to specify the physical address size used by this
>>>>>>> VM.
>>>>>>>
>>>>>>> Signed-off-by: Suzuki K Poulose <address@hidden>
>>>>>>> ---
>>>>>>>   arm/aarch64/include/kvm/kvm-config-arch.h | 5 ++++-
>>>>>>>   arm/include/arm-common/kvm-config-arch.h  | 1 +
>>>>>>>   2 files changed, 5 insertions(+), 1 deletion(-)
>>>>>>>
>>>>>>> diff --git a/arm/aarch64/include/kvm/kvm-config-arch.h 
>>>>>>> b/arm/aarch64/include/kvm/kvm-config-arch.h
>>>>>>> index 04be43d..dabd22c 100644
>>>>>>> --- a/arm/aarch64/include/kvm/kvm-config-arch.h
>>>>>>> +++ b/arm/aarch64/include/kvm/kvm-config-arch.h
>>>>>>> @@ -8,7 +8,10 @@
>>>>>>>                         "Create PMUv3 device"),                         
>>>>>>> \
>>>>>>>         OPT_U64('\0', "kaslr-seed", &(cfg)->kaslr_seed,                 
>>>>>>> \
>>>>>>>                         "Specify random seed for Kernel Address Space " 
>>>>>>> \
>>>>>>> -                       "Layout Randomization (KASLR)"),
>>>>>>> +                       "Layout Randomization (KASLR)"),                
>>>>>>> \
>>>>>>> +       OPT_INTEGER('\0', "phys-shift", &(cfg)->phys_shift,             
>>>>>>> \
>>>>>>> +                       "Specify maximum physical address size (not "   
>>>>>>> \
>>>>>>> +                       "the amount of memory)"),
>>>>>>
>>>>>> Given that this is a shift value, I think the help message could be more
>>>>>> informative. Something like:
>>>>>>
>>>>>>  "Specify maximum number of bits in a guest physical address"
>>>>>>
>>>>>> I think I'd actually leave out any mention of memory, because this does
>>>>>> actually have an effect on the amount of addressable memory in a way 
>>>>>> that I
>>>>>> don't think we want to describe in half of a usage message line :)
>>>>> Is there any particular reasons to expose this option to the user?
>>>>>
>>>>> I have recently sent a series to allow the user to specify the position
>>>>> of the RAM [1]. With that series in mind, I think the user would not 
>>>>> really
>>>>> need to specify the maximum physical shift. Instead we could automatically
>>>>> find it.
>>>>
>>>> Marc makes a good point that it doesn't help for MMIO regions, so I'm 
>>>> trying
>>>> to understand whether we can do something differently there and avoid
>>>> sacrificing the type parameter.
>>>
>>> I am not sure to understand this. kvmtools knows the memory layout 
>>> (including MMIOs) of the guest, so couldn't it guess the maximum 
>>> physical shift for that?
>>
>> That's exactly what Will was trying to avoid, by having KVM to compute
>> the size of the IPA space based on the registered memslots. We've now
>> established that it doesn't work, so what we need to define is:
>>
>> - whether we need another ioctl(), or do we carry on piggy-backing on
>> the CPU type,
> kvm type I guess

I really meant target here. Whatever you pass as a "-cpu" on your QEMU
command line.

>> - assuming the latter, whether we can reduce the number of bits used in
>> the ioctl parameter by subtly encoding the IPA size.
> Getting benefit from your Freudian slip, how should guest CPU PARange
> and maximum number of bits in a guest physical address relate?

Freudian? I'm not on the sofa yet... ;-)

> My understanding is they are not correlated at the moment and our guest
> PARange is fixed at the moment. But shouldn't they?
> 
> On Intel there is
>    qemu-system-x86_64 -M pc,accel=kvm -cpu SandyBridge,phys-bits=36
> or
>    qemu-system-x86_64 -M pc,accel=kvm -cpu SandyBridge,host-phys-bits=true
> 
> where phys-bits, as far as I understand has a a similar semantics as the
> PARange.

I think there is value in having it global, just like on x86. We don't
really support heterogeneous guests anyway.

Independently, we should also repaint/satinize PARange so that the guest
observes the same thing, no matter what CPU it runs on (an A53/A57
system could be confusing in that respect).

Thanks,

        M.
-- 
Jazz is not dead. It just smells funny...