On 08/02/2012 09:14 PM, Marcelo Tosatti wrote:
> On Sun, Jul 29, 2012 at 04:18:58PM +0800, Xiao Guangrong wrote:
>> After commit a2766325cf9f9, the error pfn is replaced by the
>> error code, it need not be released anymore
>>
>> [ The patch is compiling tested for powerpc ]
>>
>> Signed-off-by: Xiao Guangrong <xiaoguangrong@xxxxxxxxxxxxxxxxxx>
>> ---
>> arch/powerpc/kvm/e500_tlb.c | 1 -
>> arch/x86/kvm/mmu.c | 6 +++---
>> arch/x86/kvm/mmu_audit.c | 4 +---
>> arch/x86/kvm/paging_tmpl.h | 8 ++------
>> virt/kvm/iommu.c | 1 -
>> virt/kvm/kvm_main.c | 14 ++++++++------
>> 6 files changed, 14 insertions(+), 20 deletions(-)
>>
>> diff --git a/arch/powerpc/kvm/e500_tlb.c b/arch/powerpc/kvm/e500_tlb.c
>> index c8f6c58..09ce5ac 100644
>> --- a/arch/powerpc/kvm/e500_tlb.c
>> +++ b/arch/powerpc/kvm/e500_tlb.c
>> @@ -524,7 +524,6 @@ static inline void kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500,
>> if (is_error_pfn(pfn)) {
>> printk(KERN_ERR "Couldn't get real page for gfn %lx!\n",
>> (long)gfn);
>> - kvm_release_pfn_clean(pfn);
>> return;
>> }
>>
>> diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
>> index 320a781..949a5b8 100644
>> --- a/arch/x86/kvm/mmu.c
>> +++ b/arch/x86/kvm/mmu.c
>> @@ -2498,7 +2498,9 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
>> rmap_recycle(vcpu, sptep, gfn);
>> }
>> }
>> - kvm_release_pfn_clean(pfn);
>> +
>> + if (!is_error_pfn(pfn))
>> + kvm_release_pfn_clean(pfn);
>> }
>
> Can it ever be error_pfn? Seems a problem if so.
>
Yes, the no-slot-pfn, we will cache the mmio access into spte.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
