On Jul 18, 2014, at 12:38 AM, Marcelo Tosatti wrote:
> On Thu, Jul 17, 2014 at 08:18:03PM +0300, Nadav Amit wrote:
>> Small question if I may regarding kvm_mmu_pin_pages:
>>
...
>> I understand that the current use-case is for pinning only few
>> pages. Yet, wouldn't it be better (for performan
On 07/22/2014 05:46 AM, Xiao Guangrong wrote:
>> +if (is_pinned_spte(*sptep)) {
>> +/* don't nuke pinned sptes if page aging: return
>> + * young=yes instead.
>> + */
>> +if (age)
>> +
Hi Marcelo,
On Jul 10, 2014, at 3:12 AM, mtosa...@redhat.com wrote:
> struct kvm_vcpu_arch {
> /*
>* rip and regs accesses must go through
> @@ -392,6 +402,9 @@
> struct kvm_mmu_memory_cache mmu_page_cache;
> struct kvm_mmu_memory_cache mmu_page_header_cache;
>
> +
On Thu, Jul 17, 2014 at 08:18:03PM +0300, Nadav Amit wrote:
> Small question if I may regarding kvm_mmu_pin_pages:
>
> On 7/9/14, 10:12 PM, mtosa...@redhat.com wrote:
> >+
> >+static int kvm_mmu_pin_pages(struct kvm_vcpu *vcpu)
> >+{
> >+struct kvm_pinned_page_range *p;
> >+int r = 1;
> >+
Small question if I may regarding kvm_mmu_pin_pages:
On 7/9/14, 10:12 PM, mtosa...@redhat.com wrote:
+
+static int kvm_mmu_pin_pages(struct kvm_vcpu *vcpu)
+{
+ struct kvm_pinned_page_range *p;
+ int r = 1;
+
+ if (is_guest_mode(vcpu))
+ return r;
+
+ if (!v
Allow vcpus to pin spte translations by:
1) Creating a per-vcpu list of pinned ranges.
2) On mmu reload request:
- Fault ranges.
- Mark sptes with a pinned bit.
- Mark shadow pages as pinned.
3) Then modify the following actions:
- Page age => skip spte flush.