Support for CUDA pitched memory allocation and access
Submitted by Charles Durham
Assigned to Nobody
Link to original bugzilla bug (#1649)
Platform: GPU (CUDA)
Description
Would be nice to have functionality to allocate matricies with cudaMallocPitch and copied with cudaMemcpy2D, with overloaded access of "[idx]" to be pitch aligned for Matrices passed into a cuda kernel.
Edited by Eigen Bugzilla