On Fri, 2021-12-10 at 15:53 +0100, Paolo Bonzini wrote:
> On 12/10/21 13:25, David Woodhouse wrote:
> > On Thu, 2021-12-09 at 23:34 +0100, Paolo Bonzini wrote:
> > > Compared to the review it's missing this hunk:
> > >
> > > @@ -265,7 +265,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm,
> >
On 12/10/21 13:25, David Woodhouse wrote:
On Thu, 2021-12-09 at 23:34 +0100, Paolo Bonzini wrote:
Compared to the review it's missing this hunk:
@@ -265,7 +265,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct
gfn_to_pfn_cache *gpc)
gpc->valid = false;
- old_khva =
On Thu, 2021-12-09 at 23:34 +0100, Paolo Bonzini wrote:
>
> Compared to the review it's missing this hunk:
>
> @@ -265,7 +265,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct
> gfn_to_pfn_cache *gpc)
>
> gpc->valid = false;
>
> - old_khva = gpc->khva;
> + old
On Thu, 2021-12-09 at 23:34 +0100, Paolo Bonzini wrote:
> Compared to the review it's missing this hunk:
>
> @@ -265,7 +265,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct
> gfn_to_pfn_cache *gpc)
>
> gpc->valid = false;
>
> - old_khva = gpc->khva;
> + old_khva
On 12/9/21 21:40, David Woodhouse wrote:
NP, very useful fixes. Thanks. Incremental patch looks like this. It
passes the xen_shinfo_test self-test; will test it with real Xen guests
tomorrow and repost based on your kvm/next tree once it shows up.
Compared to the review it's missing this hunk:
On Thu, 2021-12-09 at 19:34 +0100, Paolo Bonzini wrote:
> Sorry for the late review...
NP, very useful fixes. Thanks. Incremental patch looks like this. It
passes the xen_shinfo_test self-test; will test it with real Xen guests
tomorrow and repost based on your kvm/next tree once it shows up.
di
Sorry for the late review...
On 11/21/21 13:54, David Woodhouse wrote:
+EXPORT_SYMBOL_GPL(kvm_gfn_to_pfn_cache_check);
+
+static void __release_gpc(struct kvm *kvm, kvm_pfn_t pfn, void *khva,
+ gpa_t gpa, bool dirty)
+{
+ /* Unmap the old page if it was mapped befor
From: David Woodhouse
This can be used in two modes. There is an atomic mode where the cached
mapping is accessed while holding the rwlock, and a mode where the
physical address is used by a vCPU in guest mode.
For the latter case, an invalidation will wake the vCPU with the new
KVM_REQ_GPC_INVA
