Re: [PATCH kvmtool v1 4/5] Add guest_memfd support to kvm__register_mem()

[Suzuki K Poulose <suzuki.poulose@arm.com>]

On 06/07/2026 12:00, Suzuki K Poulose wrote:
> On 06/07/2026 09:35, Fuad Tabba wrote:
>> Add a guest_memfd file descriptor and offset to kvm__register_mem().
>> When guest_memfd >= 0, set KVM_MEM_GUEST_MEMFD and populate the
>> guest_memfd fields, binding the memslot to the file. This prepares
>> for backing guest RAM with guest_memfd on arm64.
>>
>> Signed-off-by: Fuad Tabba <fuad.tabba@linux.dev>
>
> Looks good to me
>
> Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>

Having another look... Thought of adding my comment, no changes to the
above.


>>   int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size,
>> -              void *userspace_addr, enum kvm_mem_type type)
>> +              void *userspace_addr, int guest_memfd,
>> +              u64 guest_memfd_offset, enum kvm_mem_type type)
>>   {
>>       struct kvm_mem_bank *merged = NULL;
>>       struct kvm_mem_bank *bank;
>> @@ -327,6 +328,12 @@ int kvm__register_mem(struct kvm *kvm, u64
>> guest_phys, u64 size,
>>               .userspace_addr        = (unsigned long)userspace_addr,
>>           };
>> +        if (guest_memfd >= 0) {
>> +            mem.flags    |= KVM_MEM_GUEST_MEMFD;
>> +            mem.guest_memfd        = guest_memfd;
>> +            mem.guest_memfd_offset    = guest_memfd_offset;
>> +        }

It may be worth capturing the fd and the offset in the mem_bank here,
but agree that can be added on the need basis and there are no users
of it for now, without conversion support.

Suzuki


>> +
>>           ret = ioctl(kvm->vm_fd, KVM_SET_USER_MEMORY_REGION2, &mem);
>>           if (ret < 0) {
>>               ret = -errno;
>

IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.