Re: [Qemu-devel] [PATCH v1] s390x/cpumodel: allow to enable "idtes" feature for TCG

[Thomas Huth]

On 28.06.2017 19:02, David Hildenbrand wrote:
> On 28.06.2017 16:21, Thomas Huth wrote:
>> On 27.06.2017 18:10, David Hildenbrand wrote:
>>> STFL bit 4 and 5 are just indications to the guest, which TLB entries an
>>> IDTE call will clear. These are performance indicators for the guest.
>>>
>>> STFL bit 4:
>>>     INVALIDATE DAT TABLE ENTRY (IDTE) performs
>>>     the invalidation-and-clearing operation by
>>>     selectively clearing TLB segment-table entries
>>>     when a segment-table entry or entries are
>>>     invalidated. IDTE also performs the clearing-by-
>>>     ASCE operation. Unless bit 4 is one, IDTE simply
>>>     purges all TLBs. Bit 3 is one if bit 4 is one.
>>>
>>> We can simply set STFL bit 4 ("idtes") and still purge the complete TLB.
>>> Purging more than advertised is never bad. E.g. Linux doesn't even care
>>> about this bit. We can optimized this later.
>>
>> Not sure, but why do we need this bit in add_qemu_cpu_model_features()
>> if Linux does not care about it? We will get it automatically once we
>> support the z9 in TCG...
> 
> The idea is to use this as a list we support in addition to the z900
> features. This is later helpful when actually switching to a new model
> (z9 might still take some time). Nobody has to do go over all features
> again and see if they are implemented.

OK, I agree, that makes sense.

However, I'm not sure whether you can simply ignore the clearing-by-ASCE
stuff in this case. For example, according to the PoP:

"When the clearing-by-ASCE-option bit (bit 52 of gen-
 eral register R2 is one), the M4 field is ignored."

And the idte helper function currently always takes the M4 field into
account...

 Thomas