[PATCH] powerpc: kvm: optimize "sc 0" as fast return

[Liu ping fan]

On Fri, Nov 8, 2013 at 12:19 PM, Liu ping fan <kernelfans at gmail.com> wrote:
> On Fri, Nov 8, 2013 at 11:10 AM, Alexander Graf <agraf at suse.de> wrote:
>>
>> On 08.11.2013, at 03:44, Liu Ping Fan <kernelfans at gmail.com> wrote:
>>
>>> syscall is a very common behavior inside guest, and this patch
>>> optimizes the path for the emulation of BOOK3S_INTERRUPT_SYSCALL,
>>> so hypervisor can return to guest without heavy exit, i.e, no need
>>> to swap TLB, HTAB,.. etc
>>
>> The syscall exit you touch here only happens when you do an sc > 0 with MSR_PR set inside the guest. The only case you realistically see this is when you run PR KVM inside of an HV KVM guest.
>>
> Maybe I misunderstood the ISA spec, but refer for "6.5.14 System Call
> Interrupt", no description about the MSR_PR when sc trigger a syscall
> interrupt. So I think, guest application "sc 0" will also fall to the
> kernel who owns hypervisor mode.  Am I right?
>
Some further comment: I think the essential of the problem is whether
we switch RMA from guest to HV when interrupts raise.
DSI/ISI will be redirected to HDSI and RMA switch.  But what about
SYSCALL, and DEC, external interrupt, ...etc?

>> I don't think we should optimize for that case. Instead, we should rather try to not bounce to the 1st hypervisor in the first place in that scenario :).
>>
> Sorry, but just want to make clear about the idiom:  0 -> kernel run
> with NV, and 1st -> kernel run on HV-KVM and provide PR-KVM to up
> layer? Right?
>
> When you say "try to not bounce to the 1st hypervisor ", what is the
> exact meaning and how can we achieve this?  I am a quite newer on
> powerpc, and hope that I can get more clear figure about it  :)
>

Thanks
Pingfan