Displaying 5 results from an estimated 5 matches for "vcvtdq2pd".
2018 Jun 29
2
[RFC][VECLIB] how should we legalize VECLIB calls?
...gt; __svml_sin8(<8 x double>) after the vectorizer.
This is 8-element SVML sin() called with 8-element argument. On the surface, this looks very good.
Later on, standard vector type legalization kicks-in but only the argument and return data are legalized.
vmovaps %ymm0, %ymm1
vcvtdq2pd %xmm1, %ymm0
vextractf128 $1, %ymm1, %xmm1
vcvtdq2pd %xmm1, %ymm1
callq __svml_sin8
vmovups %ymm1, 32(%r15,%r12,8)
vmovups %ymm0, (%r15,%r12,8)
Unfortunately, __svml_sin8() doesn't use this form of input/output. It takes zmm0 and returns...
2018 Jun 29
2
[RFC][VECLIB] how should we legalize VECLIB calls?
...gt; __svml_sin8(<8 x double>) after the vectorizer.
This is 8-element SVML sin() called with 8-element argument. On the surface, this looks very good.
Later on, standard vector type legalization kicks-in but only the argument and return data are legalized.
vmovaps %ymm0, %ymm1
vcvtdq2pd %xmm1, %ymm0
vextractf128 $1, %ymm1, %xmm1
vcvtdq2pd %xmm1, %ymm1
callq __svml_sin8
vmovups %ymm1, 32(%r15,%r12,8)
vmovups %ymm0, (%r15,%r12,8)
Unfortunately, __svml_sin8() doesn't use this form of input/output. It takes zmm0 and returns...
2018 Jul 02
2
[RFC][VECLIB] how should we legalize VECLIB calls?
...ent SVML sin() called with 8-element argument. On the
>> surface, this looks very good.
>>
>> Later on, standard vector type legalization kicks-in but only the
>> argument and return data are legalized.
>>
>> vmovaps %ymm0, %ymm1
>>
>> vcvtdq2pd %xmm1, %ymm0
>>
>> vextractf128 $1, %ymm1, %xmm1
>>
>> vcvtdq2pd %xmm1, %ymm1
>>
>> callq __svml_sin8
>>
>> vmovups %ymm1, 32(%r15,%r12,8)
>>
>> vmovups %ymm0, (%r15,%r12,8)
>>...
2018 Jul 02
2
[RFC][VECLIB] how should we legalize VECLIB calls?
...orizer.
>
> This is 8-element SVML sin() called with 8-element argument. On the
> surface, this looks very good.
>
> Later on, standard vector type legalization kicks-in but only the argument
> and return data are legalized.
>
> vmovaps %ymm0, %ymm1
>
> vcvtdq2pd %xmm1, %ymm0
>
> vextractf128 $1, %ymm1, %xmm1
>
> vcvtdq2pd %xmm1, %ymm1
>
> callq __svml_sin8
>
> vmovups %ymm1, 32(%r15,%r12,8)
>
> vmovups %ymm0, (%r15,%r12,8)
>
> Unfortunately, __svml_sin8() doesn’t u...
2018 Jul 02
8
[RFC][VECLIB] how should we legalize VECLIB calls?
...-element
> argument. On the surface, this looks very good.
>
> Later on, standard vector type legalization kicks-in but
> only the argument and return data are legalized.
>
> vmovaps %ymm0, %ymm1
>
> vcvtdq2pd %xmm1, %ymm0
>
> vextractf128 $1, %ymm1, %xmm1
>
> vcvtdq2pd %xmm1, %ymm1
>
> callq __svml_sin8
>
> vmovups %ymm1, 32(%r15,%r12,8)
>
> vmovups %ymm0, (%...