diff --git a/library/ssl_tls.c b/library/ssl_tls.c index 8bc79be05..fbf43ab8f 100644 --- a/library/ssl_tls.c +++ b/library/ssl_tls.c @@ -1445,7 +1445,7 @@ static void ssl_mac( mbedtls_md_context_t *md_ctx, defined(MBEDTLS_SSL_PROTO_TLS1_2) ) /* This function makes sure every byte in the memory region is accessed * (in ascending addresses order) */ -static void ssl_read_memory( unsigned char *p, size_t len ) +static void ssl_read_memory( const unsigned char *p, size_t len ) { unsigned char acc = 0; volatile unsigned char force; @@ -1819,12 +1819,83 @@ int mbedtls_ssl_cf_hmac( size_t min_data_len, size_t max_data_len, unsigned char *output ) { - /* WORK IN PROGRESS - THIS IS NOT CONSTANT FLOW AT ALL */ - (void) min_data_len; - (void) max_data_len; + /* WORK IN PROGRESS - THIS IS ONLY PSEUDO-CONTANT-TIME */ + + /* + * Process MAC and always update for padlen afterwards to make + * total time independent of padlen. + * + * Known timing attacks: + * - Lucky Thirteen (http://www.isg.rhul.ac.uk/tls/TLStiming.pdf) + * + * To compensate for different timings for the MAC calculation + * depending on how much padding was removed (which is determined + * by padlen), process extra_run more blocks through the hash + * function. + * + * The formula in the paper is + * extra_run = ceil( (L1-55) / 64 ) - ceil( (L2-55) / 64 ) + * where L1 is the size of the header plus the decrypted message + * plus CBC padding and L2 is the size of the header plus the + * decrypted message. This is for an underlying hash function + * with 64-byte blocks. + * We use ( (Lx+8) / 64 ) to handle 'negative Lx' values + * correctly. We round down instead of up, so -56 is the correct + * value for our calculations instead of -55. + * + * Repeat the formula rather than defining a block_size variable. + * This avoids requiring division by a variable at runtime + * (which would be marginally less efficient and would require + * linking an extra division function in some builds). + */ + size_t j, extra_run = 0; + /* This size is enough to server either as input to + * md_process() or as output to md_finish() */ + unsigned char tmp[128]; + + memset( tmp, 0, sizeof( tmp ) ); + + switch( mbedtls_md_get_type( ctx->md_info ) ) + { +#if defined(MBEDTLS_MD5_C) || defined(MBEDTLS_SHA1_C) || \ +defined(MBEDTLS_SHA256_C) + case MBEDTLS_MD_MD5: + case MBEDTLS_MD_SHA1: + case MBEDTLS_MD_SHA256: + /* 8 bytes of message size, 64-byte compression blocks */ + extra_run = ( add_data_len + max_data_len + 8 ) / 64 - + ( add_data_len + data_len_secret + 8 ) / 64; + break; +#endif +#if defined(MBEDTLS_SHA512_C) + case MBEDTLS_MD_SHA384: + /* 16 bytes of message size, 128-byte compression blocks */ + extra_run = ( add_data_len + max_data_len + 16 ) / 128 - + ( add_data_len + data_len_secret + 16 ) / 128; + break; +#endif + default: + return( MBEDTLS_ERR_SSL_INTERNAL_ERROR ); + } + mbedtls_md_hmac_update( ctx, add_data, add_data_len ); mbedtls_md_hmac_update( ctx, data, data_len_secret ); + /* Make sure we access everything even when padlen > 0. This + * makes the synchronisation requirements for just-in-time + * Prime+Probe attacks much tighter and hopefully impractical. */ + ssl_read_memory( data + min_data_len, max_data_len - min_data_len ); mbedtls_md_hmac_finish( ctx, output ); + + /* Dummy calls to compression function. + * Call mbedtls_md_process at least once due to cache attacks + * that observe whether md_process() was called of not. + * Respect the usual start-(process|update)-finish sequence for + * the sake of hardware accelerators that might require it. */ + mbedtls_md_starts( ctx ); + for( j = 0; j < extra_run + 1; j++ ) + mbedtls_md_process( ctx, tmp ); + mbedtls_md_finish( ctx, tmp ); + mbedtls_md_hmac_reset( ctx ); return( 0 ); @@ -2238,34 +2309,8 @@ static int ssl_decrypt_buf( mbedtls_ssl_context *ssl ) defined(MBEDTLS_SSL_PROTO_TLS1_2) if( ssl->minor_ver > MBEDTLS_SSL_MINOR_VERSION_0 ) { - /* - * Process MAC and always update for padlen afterwards to make - * total time independent of padlen. - * - * Known timing attacks: - * - Lucky Thirteen (http://www.isg.rhul.ac.uk/tls/TLStiming.pdf) - * - * To compensate for different timings for the MAC calculation - * depending on how much padding was removed (which is determined - * by padlen), process extra_run more blocks through the hash - * function. - * - * The formula in the paper is - * extra_run = ceil( (L1-55) / 64 ) - ceil( (L2-55) / 64 ) - * where L1 is the size of the header plus the decrypted message - * plus CBC padding and L2 is the size of the header plus the - * decrypted message. This is for an underlying hash function - * with 64-byte blocks. - * We use ( (Lx+8) / 64 ) to handle 'negative Lx' values - * correctly. We round down instead of up, so -56 is the correct - * value for our calculations instead of -55. - * - * Repeat the formula rather than defining a block_size variable. - * This avoids requiring division by a variable at runtime - * (which would be marginally less efficient and would require - * linking an extra division function in some builds). - */ - size_t j, extra_run = 0; + int ret; + unsigned char add_data[13]; /* * The next two sizes are the minimum and maximum values of @@ -2280,60 +2325,21 @@ static int ssl_decrypt_buf( mbedtls_ssl_context *ssl ) const size_t max_len = ssl->in_msglen + padlen; const size_t min_len = ( max_len > 256 ) ? max_len - 256 : 0; - switch( ssl->transform_in->ciphersuite_info->mac ) + memcpy( add_data + 0, ssl->in_ctr, 8 ); + memcpy( add_data + 8, ssl->in_hdr, 3 ); + memcpy( add_data + 11, ssl->in_len, 2 ); + + ret = mbedtls_ssl_cf_hmac( &ssl->transform_in->md_ctx_dec, + add_data, sizeof( add_data ), + ssl->in_msg, ssl->in_msglen, + min_len, max_len, + mac_expect ); + if( ret != 0 ) { -#if defined(MBEDTLS_MD5_C) || defined(MBEDTLS_SHA1_C) || \ - defined(MBEDTLS_SHA256_C) - case MBEDTLS_MD_MD5: - case MBEDTLS_MD_SHA1: - case MBEDTLS_MD_SHA256: - /* 8 bytes of message size, 64-byte compression blocks */ - extra_run = ( 13 + ssl->in_msglen + padlen + 8 ) / 64 - - ( 13 + ssl->in_msglen + 8 ) / 64; - break; -#endif -#if defined(MBEDTLS_SHA512_C) - case MBEDTLS_MD_SHA384: - /* 16 bytes of message size, 128-byte compression blocks */ - extra_run = ( 13 + ssl->in_msglen + padlen + 16 ) / 128 - - ( 13 + ssl->in_msglen + 16 ) / 128; - break; -#endif - default: - MBEDTLS_SSL_DEBUG_MSG( 1, ( "should never happen" ) ); - return( MBEDTLS_ERR_SSL_INTERNAL_ERROR ); + MBEDTLS_SSL_DEBUG_RET( 1, "mbedtls_ssl_cf_hmac", ret ); + return( ret ); } - extra_run &= correct * 0xFF; - - mbedtls_md_hmac_update( &ssl->transform_in->md_ctx_dec, ssl->in_ctr, 8 ); - mbedtls_md_hmac_update( &ssl->transform_in->md_ctx_dec, ssl->in_hdr, 3 ); - mbedtls_md_hmac_update( &ssl->transform_in->md_ctx_dec, ssl->in_len, 2 ); - mbedtls_md_hmac_update( &ssl->transform_in->md_ctx_dec, ssl->in_msg, - ssl->in_msglen ); - /* Make sure we access everything even when padlen > 0. This - * makes the synchronisation requirements for just-in-time - * Prime+Probe attacks much tighter and hopefully impractical. */ - ssl_read_memory( ssl->in_msg + ssl->in_msglen, padlen ); - mbedtls_md_hmac_finish( &ssl->transform_in->md_ctx_dec, mac_expect ); - - /* Dummy calls to compression function. - * Call mbedtls_md_process at least once due to cache attacks - * that observe whether md_process() was called of not. - * Respect the usual start-(process|update)-finish sequence for - * the sake of hardware accelerators that might require it. */ - mbedtls_md_starts( &ssl->transform_in->md_ctx_dec ); - for( j = 0; j < extra_run + 1; j++ ) - mbedtls_md_process( &ssl->transform_in->md_ctx_dec, ssl->in_msg ); - { - /* The switch statement above already checks that we're using - * one of MD-5, SHA-1, SHA-256 or SHA-384. */ - unsigned char tmp[384 / 8]; - mbedtls_md_finish( &ssl->transform_in->md_ctx_dec, tmp ); - } - - mbedtls_md_hmac_reset( &ssl->transform_in->md_ctx_dec ); - /* Make sure we access all the memory that could contain the MAC, * before we check it in the next code block. This makes the * synchronisation requirements for just-in-time Prime+Probe