* Sebastian Siewior | 2007-12-07 19:24:47 [+0100]: >This is required for one or two crypto users in tree which >chain the IV. Snippets from tcrypt mode=200, before the patch: | testing speed of cbc(aes) encryption | test 5 (192 bit key, 16 byte blocks): 1 operation in 3109 cycles (16 bytes) | test 9 (192 bit key, 8192 byte blocks): 1 operation in 680945 cycles (8192 bytes) | test 10 (256 bit key, 16 byte blocks): 1 operation in 2143 cycles (16 bytes) | test 14 (256 bit key, 8192 byte blocks): 1 operation in 757506 cycles (8192 bytes) | | testing speed of cbc(aes) decryption | test 5 (192 bit key, 16 byte blocks): 1 operation in 3222 cycles (16 bytes) | test 9 (192 bit key, 8192 byte blocks): 1 operation in 706132 cycles (8192 bytes) | test 10 (256 bit key, 16 byte blocks): 1 operation in 2279 cycles (16 bytes) | test 14 (256 bit key, 8192 byte blocks): 1 operation in 781974 cycles (8192 bytes) and after: | testing speed of cbc(aes) encryption | test 5 (192 bit key, 16 byte blocks): 1 operation in 3401 cycles (16 bytes) | test 9 (192 bit key, 8192 byte blocks): 1 operation in 680604 cycles (8192 bytes) | test 10 (256 bit key, 16 byte blocks): 1 operation in 2177 cycles (16 bytes) | test 14 (256 bit key, 8192 byte blocks): 1 operation in 756906 cycles (8192 bytes) | | testing speed of cbc(aes) decryption | test 5 (192 bit key, 16 byte blocks): 1 operation in 3567 cycles (16 bytes) | test 9 (192 bit key, 8192 byte blocks): 1 operation in 705889 cycles (8192 bytes) | test 10 (256 bit key, 16 byte blocks): 1 operation in 2317 cycles (16 bytes) | test 14 (256 bit key, 8192 byte blocks): 1 operation in 781241 cycles (8192 bytes) it is about 9% slower for test 5, the others are less than three percent. Sebastian - To unsubscribe from this list: send the line "unsubscribe linux-crypto" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
