show the ipp function list which is optimize by AVX or AVX2?

Fu__Yu — Wed, 27 Jun 2018 08:16:08 GMT

Currently I am doing some benchmark comparing the time usage between enable / disable the AVX / AVX2 / SSE. And I found there is no remarkable difference for some math function such as add & multiply. I switch the flag by manually change the cpu variable and call function: ippSetCpuFeatures. Is it reasonable or not?

status = ippGetCpuFeatures(&mask, 0);
  // chage instruction flag
  //ippCPUID_MMX
  printf("\nBefore set ippCPUID_MMX:      %c\t%c\t", (mask & ippCPUID_MMX) ? 'Y' : 'N', (emask & ippCPUID_MMX) ? 'Y' : 'N');
  if (a = (mask & ippCPUID_MMX))
  {
   emask = (emask - ippCPUID_MMX);
  }
  printf("\nAfter set ippCPUID_MMX:     %c\t%c\t", (mask & ippCPUID_MMX) ? 'Y' : 'N', (emask & ippCPUID_MMX) ? 'Y' : 'N');
  ...
  IppSts = ippSetCpuFeatures(emask);
  if (IppSts == ippStsFeatureNotSupported || IppSts == ippStsUnknownFeature) //ippStsNoErr
  {
   wStatus = IppSts;
   goto EXIT;
  }

More detailed pls. refer to below.

	Intel Xeon CPU E5-2650 V4 @ 2.2GHz
	Image sz: 2048 * 2048
	Unit of time: ms
	Ipp version	Instruction & Compiler setting	Add		Multiply		SobelX&Y
			Integer (unsigned char)	Float	Integer	Float	Integer
	Ipp New (2018)	AVX2 off + AVX off + SSSE3 off	0.55	5.96	0.67	5.69	14.84
31	Ipp New (2018)	AVX2 off + AVX off + SSSE3 on	0.56	5.99	0.62	5.68	4.29
	Ipp New (2018)	AVX2 off + AVX on	0.56	5.99	0.62	5.68	4.49
61439	Ipp New (2018)	AVX2 on	0.56	5.84	0.57	5.61	2.99
	Ipp New (2018)	AVX2 on + VS AVX Compiler on	0.54	5.91	0.55	5.72	3.06
	Ipp New (2018)	AVX2 on + VS AVX2 Compiler on	0.55	5.85	0.55	5.67	2.98

Thanks & regards,

Could you provide a table to

Fu__Yu — Wed, 27 Jun 2018 08:19:02 GMT

Could you provide a table to show the ipp function list which is optimize by AVX or AVX2?

topic Could you provide a table to in Intel® Integrated Performance Primitives

show the ipp function list which is optimize by AVX or AVX2?

Could you provide a table to

https://software.intel.com/en