Geek asks and answers

Incx in BLAS procedures

There are certain BLAS procedures that take as an argument the increment of the vector X, namely incX. I can’t find what the increment is and how it affects the result of the calculation.

Can someone provide some example or any other information?

Update:

I found the best information here: Intel HPC mkl manual

+5
source share
1 answer

It is pretty simple.

Let's look at an example axpy(n,a,*x,incx,*y,incy)that computes:y = ax + y

If, for example, you need to calculate:

y[0] = ax[0] + y[0]; y[1] = ax[2] + y[1]; y[2] = ax[4] + y[2]

Then your call: axpy(3,a,x,2,y,1)

But usually for basic operations you just need to specify incx = incy = 1

+9
source

More articles:


All Articles