Intel® C++ Compiler
Community support and assistance for creating C++ code that runs on platforms based on Intel® processors.

Missing several Intrinsics for Miscellaneous Operations in User and Reference Guide for C++

Nathan_Weeks
Beginner
334 Views

In the User and Reference Guide for the Intel® C++ Compiler 15.0, the "Intrinsics for Miscellaneous Operations" (https://software.intel.com/en-us/node/524118) is missing the following intrinsics:

  • _mm256_set_epi8
  • _mm256_set_epi16
  • _mm256_set_epi64x
  • _mm256_setr_epi8
  • _mm256_setr_epi16
  • _mm256_setr_epi64x
  • _mm256_set1_epi8
  • _mm256_set1_epi16
  • _mm256_set1_epi64x

--
Nathan Weeks
Systems Analyst
Iowa State University -- Department of Mathematics
http://weeks.public.iastate.edu/

0 Kudos
3 Replies
KitturGanesh
Employee
334 Views

Thanks Nathan, yet again I'll escalate this to the doc team - appreciate much.

_Kittur

0 Kudos
Nathan_Weeks
Beginner
334 Views

Hi Kittur,

I noticed additional documentation discrepancies for the instructions corresponding to AVX Miscellaneous Intrinsics in the Intel® 64 and IA-32 Architectures Software Developer’s Manual:

1. Vol. 2B 4-513 (PDF p. 1609): In the "Intel C/C++ Compiler Intrinsic Equivalent" section, INSERTF128 should be VINSERTF128.

2. Vol. 2A 3-453 (PDF p. 943): In the "Intel C/C++ Compiler Intrinsic Equivalent" section, 

    LDDQU: __m256i _mm256_lddqu_si256 (__m256i * p);

should be

    VLDDQU: __m256i _mm256_lddqu_si256 (__m256i * p);

3. Vol. 2A 3-547 (PDF p. 1037): The "Intel C/C++ Compiler Intrinsic Equivalent" section should contain:

    VMOVMSKPD:     _mm256_movemask_pd(__m256d a)

--
Nathan Weeks
Systems Analyst
Iowa State University -- Department of Mathematics
http://weeks.public.iastate.edu/

Vol. 2A 3-453 (PDF p. 943):
Vol. 2A 3-453 (PDF p. 943):
0 Kudos
KitturGanesh
Employee
334 Views

Thanks Nathan. I've updated the issue I'd filed with this new issue you noticed as well, with our doc team. Again, appreciate much for catching this.

_Regards, Kittur

0 Kudos
Reply