What is the purpose of pointerE array in the CSR sparse matrix format, if pointerB already contains enough information about start and end of a particular row?
The purpose of having pointerE is to allow working with a submatrix of a given sparse matrix. In other words, it allows the user to have a sparse matrix but use only a part of it in computations, essentially by skipping elements from the end of each row (being pointed to by pointerE[i], ... pointerB[row+1] in the column indices and values arrays).
By having also a pointerB different from some general rowStart of a given matrix, the user can also skip some elememnts from the beginning of each row.
Let us know if you have further questions.
This issue has been resolved and we will no longer respond to this thread. If you require additional assistance from Intel, please start a new thread. Any further interaction in this thread will be considered community only