On 17/11/2011 16:08, Ivan Voras wrote: > On 17/11/2011 15:21, Jakob Bohm wrote: >> On 11/17/2011 3:05 PM, Ivan Voras wrote: >>> Hello, >>> >>> I'm having trouble using AES_cfb128_encrypt() - I've isolated it to the >>> simple example below. The problem is that the decryption doesn't work - >>> the output (in buf3) is binary gibberish. >>> >>> SHA256(key, key_size, enc_key); >>> AES_set_encrypt_key(enc_key, 256,&aes_enc_key); >>> AES_set_decrypt_key(enc_key, 256,&aes_dec_key); >>> >>> memcpy(buf1, str, str_size); >>> dump_data(buf1, str_size, "1"); >>> >>> num = 0; >>> memset(iv, 0, 16); >>> AES_cfb128_encrypt(buf1, buf2, str_size,&aes_enc_key, iv,&num, >>> AES_ENCRYPT); >>> dump_data(buf2, str_size, "2"); >>> >>> num = 0; >>> memset(iv, 0, 16); >>> AES_cfb128_encrypt(buf2, buf3, str_size,&aes_dec_key, iv,&num, >>> AES_DECRYPT); >> Shouldn't this be AES_cfb128_decrypt() ? > > No, there is only AES_cfb128_encrypt() and it receives an argument > telling it what to do (AES_ENCRYPT or AES_DECRYPT).
But your question was in the right direction :) Actually it was obvious - since CFB generates a keystream, the same key must be used in both direction (i.e. no separate key schedule for decryption). ______________________________________________________________________ OpenSSL Project http://www.openssl.org User Support Mailing List openssl-users@openssl.org Automated List Manager majord...@openssl.org
