On 08.05.2018 09:32, Christian Borntraeger wrote:
>
>
> On 05/08/2018 09:30 AM, Christian Borntraeger wrote:
>>
>>
>> On 05/02/2018 04:28 PM, David Hildenbrand wrote:
>>> On 02.05.2018 14:42, Heiko Carstens wrote:
>>>> On Wed, May 02, 2018 at 01:08:14PM +0200, David Hildenbrand wrote:
>>>>> This makes it certainly more readable.
>>>>>
>>>>> Signed-off-by: David Hildenbrand <david@xxxxxxxxxx>
>>>>> ---
>>>>> arch/s390/kvm/vsie.c | 10 +++++-----
>>>>> 1 file changed, 5 insertions(+), 5 deletions(-)
>>>>>
>>>>> diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c
>>>>> index 969882b54266..7c51a9dc0ec8 100644
>>>>> --- a/arch/s390/kvm/vsie.c
>>>>> +++ b/arch/s390/kvm/vsie.c
>>>>> @@ -557,7 +557,7 @@ static int pin_blocks(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page)
>>>>> if (test_kvm_cpu_feat(vcpu->kvm, KVM_S390_VM_CPU_FEAT_64BSCAO))
>>>>> gpa |= (u64) READ_ONCE(scb_o->scaoh) << 32;
>>>>> if (gpa) {
>>>>> - if (!(gpa & ~0x1fffUL))
>>>>> + if (gpa < 8192)
>>>>
>>>> 2 * PAGE_SIZE, please. That's how it is done on various other places too.
>>>>
>>>
>>> Christian, does the documentation use the wording "lowcore" or "8192" / 8k?
>>>
>>> If it is lowcore, I agree to use 2 * PAGE_SIZE. If not, I prefer it to
>>> directly match documentation.
>>>
>>> Anyhow, I leave this decision to Janosch and Christian. Thanks for
>>> having alook!
>>>
>> prefix area and reverse prefix area.
>> In fact what we check here is actually the reversed prefix area so for completeness we
>> would need an additional check.
>
> We seem to check both for SCA so we are good for SCA.
>
I remember that this check only had to be performed for the SCA.
I'll resend, using PAGE_SIZE * 2 then. Thanks!
--
Thanks,
David / dhildenb
