On Tue, Mar 24, 2009 at 11:47:33AM +0200, Avi Kivity wrote:
> Marcelo Tosatti wrote:
>>> Maybe it's best to resync when relinking a global page?
>>>
>>
>> How about this. It will shorten the unsync period of global pages,
>> unfortunately.
>>
>> diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
>> index 2a36f7f..bccdcc7 100644
>> --- a/arch/x86/kvm/mmu.c
>> +++ b/arch/x86/kvm/mmu.c
>> @@ -1238,6 +1238,10 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct
>> kvm_vcpu *vcpu,
>> set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests);
>> kvm_mmu_mark_parents_unsync(vcpu, sp);
>> }
>> + if (role.level != PT_PAGE_TABLE_LEVEL &&
>> + !list_empty(&vcpu->kvm->arch.oos_global_pages))
>> + set_bit(KVM_REQ_MMU_GLOBAL_SYNC,
>> &vcpu->requests);
>> +
>> pgprintk("%s: found\n", __func__);
>> return sp;
>> }
>> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
>> index 2ea8262..48169d7 100644
>> --- a/arch/x86/kvm/x86.c
>> +++ b/arch/x86/kvm/x86.c
>> @@ -3109,6 +3109,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu,
>> struct kvm_run *kvm_run)
>> kvm_write_guest_time(vcpu);
>> if (test_and_clear_bit(KVM_REQ_MMU_SYNC, &vcpu->requests))
>> kvm_mmu_sync_roots(vcpu);
>> + if (test_and_clear_bit(KVM_REQ_MMU_GLOBAL_SYNC,
>> &vcpu->requests))
>> + kvm_mmu_sync_global(vcpu);
>> if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
>> kvm_x86_ops->tlb_flush(vcpu);
>> if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS
>
> Windows will (I think) write a PDE on every context switch, so this
> effectively disables global unsync for that guest.
>
> What about recursively syncing the newly linked page in FNAME(fetch)()?
> If the page isn't global, this becomes a no-op, so no new overhead. The
> only question is the expense when linking a populated top-level page,
> especially in long mode.
Yes, I started doing that but it touches the nice fastpath in fetch().
I'll see if I can come up with something and with numbers.
Thanks
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
