cuda.bindings.cufile.buf_register#

cuda.bindings.cufile.buf_register(intptr_t buf_ptr_base, size_t length, int flags)#

register an existing cudaMalloced memory with cuFile to pin for GPUDirect Storage access or register host allocated memory with cuFile.

Parameters:

buf_ptr_base (intptr_t) – buffer pointer allocated.
length (size_t) – size of memory region from the above specified bufPtr.
flags (int) – CU_FILE_RDMA_REGISTER.

Note

This memory will be use to perform GPU direct DMA from the supported storage.

Note

This API is intended for usecases where the memory is used as streaming buffer that is reused across multiple cuFile IO operations before calling cuFileBufDeregister.