[PATCH v7 01/45] KVM: Prepare for handling only shared mappings in mmu_notifier events

[Steven Price]

On 02/03/2025 23:36, Gavin Shan wrote:
> On 2/14/25 2:13 AM, Steven Price wrote:
>> From: Sean Christopherson <seanjc at google.com>
>>
>> Add flags to "struct kvm_gfn_range" to let notifier events target only
>> shared and only private mappings, and write up the existing mmu_notifier
>> events to be shared-only (private memory is never associated with a
>> userspace virtual address, i.e. can't be reached via mmu_notifiers).
>>
>> Add two flags so that KVM can handle the three possibilities (shared,
>> private, and shared+private) without needing something like a tri-state
>> enum.
>>
>> Link: https://lore.kernel.org/all/ZJX0hk+KpQP0KUyB@google.com
>> Signed-off-by: Sean Christopherson <seanjc at google.com>
>> Signed-off-by: Steven Price <steven.price at arm.com>
>> ---
>>   include/linux/kvm_host.h | 2 ++
>>   virt/kvm/kvm_main.c      | 7 +++++++
>>   2 files changed, 9 insertions(+)
>>
>> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
>> index 3cb9a32a6330..0de1e485452c 100644
>> --- a/include/linux/kvm_host.h
>> +++ b/include/linux/kvm_host.h
>> @@ -266,6 +266,8 @@ struct kvm_gfn_range {
>>       gfn_t end;
>>       union kvm_mmu_notifier_arg arg;
>>       enum kvm_gfn_range_filter attr_filter;
>> +    bool only_private;
>> +    bool only_shared;
>>       bool may_block;
>>   };
> 
> The added members (only_private and only_shared) looks duplicated to the
> member of attr_filter, which can be set to KVM_FILTER_SHARED,
> KVM_FILTER_PRIVATE,
> or both of them. More details can be found from the following commit where
> attr_filter was by dca6c88532322 ("KVM: Add member to struct kvm_gfn_range
> to indicate private/shared").
> 
> I'm guessing Sean's suggestion was given before dca6c88532322 showed up.

Thanks for pointing this out - you are absolutely right. I need to
switch the code in the following patches over to use attr_filter
instead. I hadn't realised when rebasing that attr_filter is the
replacement.

Thanks,
Steve

>>   bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range);
>> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
>> index faf10671eed2..4f0136094fac 100644
>> --- a/virt/kvm/kvm_main.c
>> +++ b/virt/kvm/kvm_main.c
>> @@ -593,6 +593,13 @@ static __always_inline kvm_mn_ret_t
>> __kvm_handle_hva_range(struct kvm *kvm,
>>                * the second or later invocation of the handler).
>>                */
>>               gfn_range.arg = range->arg;
>> +
>> +            /*
>> +             * HVA-based notifications aren't relevant to private
>> +             * mappings as they don't have a userspace mapping.
>> +             */
>> +            gfn_range.only_private = false;
>> +            gfn_range.only_shared = true;
>>               gfn_range.may_block = range->may_block;
>>               /*
>>                * HVA-based notifications aren't relevant to private
> 
> Thanks,
> Gavin
>