Re: [PATCH v6 20/28] s390x: Guest support for Secure-IPL Code Loading Attributes Facility (SCLAF)

[Thomas Huth <thuth@redhat.com>]

On 29/09/2025 14.25, Thomas Huth wrote:
> On 18/09/2025 01.21, Zhuoying Cai wrote:
> ...
>> diff --git a/target/s390x/gen-features.c b/target/s390x/gen-features.c
>> index bd2060ab93..c3e0c6ceff 100644
>> --- a/target/s390x/gen-features.c
>> +++ b/target/s390x/gen-features.c
>> @@ -722,6 +722,7 @@ static uint16_t full_GEN16_GA1[] = {
>>       S390_FEAT_UV_FEAT_AP_INTR,
>>       S390_FEAT_CERT_STORE,
>>       S390_FEAT_SIPL,
>> +    S390_FEAT_SCLAF,
>>   };
>  >
>>   static uint16_t full_GEN17_GA1[] = {
>> @@ -924,6 +925,7 @@ static uint16_t qemu_MAX[] = {
>>       S390_FEAT_EXTENDED_LENGTH_SCCB,
>>       S390_FEAT_CERT_STORE,
>>       S390_FEAT_SIPL,
>> +    S390_FEAT_SCLAF,
>>   };
> 
> In the cover letter you wrote "All actions must be performed on a KVM 
> guest" ... so does this feature depend on KVM or not? If you cannot use the 
> feature with TCG, I think you should not add this to the "qemu_MAX" CPU model?

Ok, after reading through a bunch of patches now, it seems like this feature 
also works with TCG, right? So the modification to qemu_MAX should stay, 
never mind me previous comment here, I was just confused by the term "KVM 
guest" in the cover letter.

  Thomas