Re: [PATCH v2 0/7] x86: further template folding and misc improvements

["H.J. Lu" <hjl dot tools at gmail dot com>]

On Thu, Sep 13, 2018 at 5:55 AM, Jan Beulich <JBeulich@suse.com> wrote:
>>>> On 13.09.18 at 14:48, <hjl.tools@gmail.com> wrote:
>> On Thu, Sep 13, 2018 at 4:58 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>> On Thu, Sep 13, 2018 at 1:59 AM, Jan Beulich <JBeulich@suse.com> wrote:
>>>>>>> On 12.09.18 at 23:19, <hjl.tools@gmail.com> wrote:
>>>>> On Wed, Sep 5, 2018 at 4:47 AM, Jan Beulich <JBeulich@suse.com> wrote:
>>>>>> 1: add code comment on deprecated status of pseudo-suffixes
>>>>>> 2: improve operand reversal
>>>>>> 3: fold ILP32 output of "opts" tests
>>>>>> 4: extra operand reversal "opts" tests
>>>>>> 5: use D attribute also for SIMD templates
>>>>>> 6: also allow D on 3-operand insns
>>>>>> 7: drop unnecessary {,No}Rex64
>>>>>>
>>>>>> Besides the new last patch the main change here compared to v1 is
>>>>>> the splitting up of what is now patch 2. Patches 3 and 4 are therefore
>>>>>> optional now. Patches 5 and 6 have been approved already, but
>>>>>> can't go in without what is now patch 2.
>>>>>
>>>>> All looks good, except for
>>>>>
>>>>> https://sourceware.org/ml/binutils/2018-09/msg00026.html
>>>>>
>>>>> Please don't add new tests with the .s suffix.  Please use {load} or
>>>>> {store} instead.
>>>>
>>>> That's done in the earlier patch. I did submit the split parts in order
>>>> to have everything available, but I'll simply omit that patch when
>>>> committing (I take your response as "okay except for this one patch").
>>>
>>> That is OK.
>>>
>>> Thanks.
>>>
>>> --
>>> H.J.
>>
>> But I see:
>>
>> FAIL: i386 arch 10
>> FAIL: i386 arch 10 (lzcnt)
>> FAIL: i386 arch 10 (prefetchw)
>> FAIL: i386 arch 10 (bdver1)
>> FAIL: i386 arch 10 (bdver2)
>> FAIL: i386 arch 10 (bdver3)
>> FAIL: i386 arch 10 (bdver4)
>> FAIL: i386 arch 10 (btver1)
>> FAIL: i386 arch 10 (btver2)
>> FAIL: i386 noavx-1
>> FAIL: i386 noavx-3
>> FAIL: i386 AVX
>> FAIL: i386 AVX (Intel disassembly)
>> FAIL: x86-64 arch 2
>> FAIL: x86-64 arch 2 (lzcnt)
>> FAIL: x86-64 arch 2 (prefetchw)
>> FAIL: x86-64 arch 2 (bdver1)
>> FAIL: x86-64 arch 2 (bdver2)
>> FAIL: x86-64 arch 2 (bdver3)
>> FAIL: x86-64 arch 2 (bdver4)
>> FAIL: x86-64 arch 2 (btver1)
>> FAIL: x86-64 arch 2 (btver2)
>> FAIL: x86-64 AVX
>> FAIL: x86-64 AVX (Intel mode)
>> FAIL: x86-64 (ILP32) arch 2
>> FAIL: x86-64 (ILP32) AVX (Intel mode)
>> FAIL: x86-64 (ILP32) AVX
>
> For which target(s)? The testsuite certainly ran without any regressions
> for me, at every patch boundary.

it failed on "zveroall".  I checked in a fix.


-- 
H.J.