CUDA Programming

Index

Note: Page numbers with “f” denote figures; “t” tables.

A

Advanced Encryption Standard (AES), 264–266

cipher, 232

vs. CPU

AES-NI code, 251

AES-NI sample library, 250

decode kernel, 252

__forceinline__ directive function, 258

global memory bandwidth, 259

gmul tables, 256

GTX470 GPUs, 251

memory bandwidth, 255–256, 256f

num_cipher_blocks, 251

PTX instructions, 257

Sandybridge-E system, 252

s_box function, 255

shared memory, 257, 258f

stack elimination, memory usage, 259, 259f

uint4 type, 257, 266

cudaMemcpyToSymbol function, 262

decryption, 233, 260

ECB, 232, 233f

EXTRACT macro, 239

GT9800 card, 262, 265–266

GTX260 card, 261, 265–266

GTX460 card, 260

__host__ and __device__ qualifiers, 239

initial kernel, 239

ITEF, 232

kernel performance ...