Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6404 Discussions

How to use the shave to get the square root of a 32-bit float ?

idata
Employee
557 Views

I see this in the documentation, but there are no samples or documentation for how to use it:

 

/* 3.3.13 SAU.SQT _/

 

/_ Square root function */

 

half __builtin_shave_sau_sqt_f16_l_r ( half ); /* SAU.SQT.f16_l _/

 

half __builtin_shave_sau_sqt_f16_h_r ( half ); /_ SAU.SQT.f16_h */

 

We're using this in an embedded product. I can use inline assembly or the C++ moviCompile compiler.

 

The objective is to do a square root of a 32-bit float, on the shave, as fast as possible.

 

Since this has a 'high' and 'low', is that for two bytes of the float at a time? How is it used?

 

There are a dozen other functions like this that I'd like to use as well, but can't because of poor documentation.

 

Thank You,

 

Jeff
0 Kudos
1 Reply
idata
Employee
347 Views

where did you get the moviCompiler ?

0 Kudos
Reply