Re: [Qemu-devel] [qemu-s390x] [PATCH 2/3] s390x/cpumodel: also change name of vxbeh

[Christian Borntraeger <borntraeger@de.ibm.com>]

On 16.07.19 09:24, David Hildenbrand wrote:
> On 15.07.19 18:12, Christian Borntraeger wrote:
>>
>>
>> On 15.07.19 17:50, Christian Borntraeger wrote:
>>>
>>>
>>> On 15.07.19 17:02, Thomas Huth wrote:
>>>> On 15/07/2019 16.23, Christian Borntraeger wrote:
>>>>> David suggested to keep everything in sync as 4.1 is not yet released.
>>>>> This patch fixes the name "vxbeh" into "vxp".
>>>>>
>>>>> To simplify the backports this patch will not change VECTOR_BCD_ENH as
>>>>> this is just an internal name. That will be done by an extra patch that
>>>>> does not need to be backported.
>>>>>
>>>>> Suggested-by: David Hildenbrand <david@redhat.com>
>>>>> Fixes: d05be57ddc2e ("s390: cpumodel: fix description for the new vector facility")
>>>>> Fixes: 54d65de0b525 ("s390x/cpumodel: vector enhancements")
>>>>> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
>>>>> ---
>>>>>  target/s390x/cpu_features_def.inc.h | 2 +-
>>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>>>
>>>>> diff --git a/target/s390x/cpu_features_def.inc.h b/target/s390x/cpu_features_def.inc.h
>>>>> index 3118a9f89228..99f58a89318a 100644
>>>>> --- a/target/s390x/cpu_features_def.inc.h
>>>>> +++ b/target/s390x/cpu_features_def.inc.h
>>>>> @@ -104,7 +104,7 @@ DEF_FEAT(CMM_NT, "cmmnt", STFL, 147, "CMM: ESSA-enhancement (no translate) facil
>>>>>  DEF_FEAT(VECTOR_ENH2, "vxeh2", STFL, 148, "Vector Enhancements facility 2")
>>>>>  DEF_FEAT(ESORT_BASE, "esort-base", STFL, 150, "Enhanced-sort facility (excluding subfunctions)")
>>>>>  DEF_FEAT(DEFLATE_BASE, "deflate-base", STFL, 151, "Deflate-conversion facility (excluding subfunctions)")
>>>>> -DEF_FEAT(VECTOR_BCD_ENH, "vxbeh", STFL, 152, "Vector-Packed-Decimal-Enhancement Facility")
>>>>> +DEF_FEAT(VECTOR_BCD_ENH, "vxp", STFL, 152, "Vector-Packed-Decimal-Enhancement Facility")
>>>>
>>>> We already have:
>>>>
>>>> DEF_FEAT(VECTOR_PACKED_DECIMAL, "vxpd", STFL, 134, "Vector packed decimal facility")
>>>>
>>>> ... so I rather expected something like "vxpde" here instead? Or is there a reason
>>>>
>>> for just using "vxp"?
>>>
>>> Matching what the Linux kernel has.
>>>
>>> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/arch/s390/kernel/processor.c?id=a8fd61688dfad6fdce95fa64cacd8a66595697b8
>>>>
>>
>> Since we differ from the kernel in other places as well we might use something else, of course.
>>
> 
> We also have
> 
> sortl vs. sort
> vxe vs. vxeh
> vxe2 vs. vxeh2
> 
> So I tend to prefer "vxpde", or rather "vxpdeh".
> 
> (all other enhancement facilities have "eh", so we should actually use
> "vxpdeh")

Fine with me. Conny, shall I resend or can you fixup everything?