I am using vslsconvexec1d() to realize the convolution. However, I have several pairs of arrays, such as (x1, y1), (x2, y2)...... Before I use this function, I need to create a task descriptor. I am wondering whether I only create a task descriptor for all the array pairs in convolution. Thank you very much for your help.
For more complete information about compiler optimizations, see our Optimization Notice.