openssl/ssl/record/record.h

264 lines
9.5 KiB
C
Raw Normal View History

/*
* Copyright 1995-2016 The OpenSSL Project Authors. All Rights Reserved.
*
* Licensed under the OpenSSL license (the "License"). You may not use
* this file except in compliance with the License. You can obtain a copy
* in the file LICENSE in the source distribution or at
* https://www.openssl.org/source/license.html
*/
/*****************************************************************************
* *
* These structures should be considered PRIVATE to the record layer. No *
* non-record layer code should be using these structures in any way. *
* *
*****************************************************************************/
typedef struct ssl3_buffer_st {
/* at least SSL3_RT_MAX_PACKET_SIZE bytes, see ssl3_setup_buffers() */
unsigned char *buf;
/* default buffer size (or 0 if no default set) */
size_t default_len;
/* buffer size */
size_t len;
/* where to 'copy from' */
int offset;
/* how many bytes left */
int left;
} SSL3_BUFFER;
#define SEQ_NUM_SIZE 8
typedef struct ssl3_record_st {
/* Record layer version */
/* r */
int rec_version;
/* type of record */
/* r */
int type;
/* How many bytes available */
/* rw */
unsigned int length;
/*
* How many bytes were available before padding was removed? This is used
* to implement the MAC check in constant time for CBC records.
*/
/* rw */
unsigned int orig_len;
/* read/write offset into 'buf' */
/* r */
unsigned int off;
/* pointer to the record data */
/* rw */
unsigned char *data;
/* where the decode bytes are */
/* rw */
unsigned char *input;
/* only used with decompression - malloc()ed */
/* r */
unsigned char *comp;
/* Whether the data from this record has already been read or not */
/* r */
unsigned int read;
/* epoch number, needed by DTLS1 */
/* r */
unsigned long epoch;
/* sequence number, needed by DTLS1 */
/* r */
unsigned char seq_num[SEQ_NUM_SIZE];
} SSL3_RECORD;
typedef struct dtls1_bitmap_st {
/* Track 32 packets on 32-bit systems and 64 - on 64-bit systems */
unsigned long map;
/* Max record number seen so far, 64-bit value in big-endian encoding */
unsigned char max_seq_num[SEQ_NUM_SIZE];
} DTLS1_BITMAP;
typedef struct record_pqueue_st {
unsigned short epoch;
struct pqueue_st *q;
} record_pqueue;
typedef struct dtls1_record_data_st {
unsigned char *packet;
unsigned int packet_length;
SSL3_BUFFER rbuf;
SSL3_RECORD rrec;
# ifndef OPENSSL_NO_SCTP
struct bio_dgram_sctp_rcvinfo recordinfo;
# endif
} DTLS1_RECORD_DATA;
typedef struct dtls_record_layer_st {
/*
* The current data and handshake epoch. This is initially
* undefined, and starts at zero once the initial handshake is
* completed
*/
unsigned short r_epoch;
unsigned short w_epoch;
/* records being received in the current epoch */
DTLS1_BITMAP bitmap;
/* renegotiation starts a new set of sequence numbers */
DTLS1_BITMAP next_bitmap;
/* Received handshake records (processed and unprocessed) */
record_pqueue unprocessed_rcds;
record_pqueue processed_rcds;
/*
* Buffered application records. Only for records between CCS and
* Finished to prevent either protocol violation or unnecessary message
* loss.
*/
record_pqueue buffered_app_data;
/*
* storage for Alert/Handshake protocol data received but not yet
* processed by ssl3_read_bytes:
*/
unsigned char alert_fragment[DTLS1_AL_HEADER_LENGTH];
unsigned int alert_fragment_len;
unsigned char handshake_fragment[DTLS1_HM_HEADER_LENGTH];
unsigned int handshake_fragment_len;
/* save last and current sequence numbers for retransmissions */
unsigned char last_write_sequence[8];
unsigned char curr_write_sequence[8];
} DTLS_RECORD_LAYER;
/*****************************************************************************
* *
* This structure should be considered "opaque" to anything outside of the *
* record layer. No non-record layer code should be accessing the members of *
* this structure. *
* *
*****************************************************************************/
typedef struct record_layer_st {
/* The parent SSL structure */
SSL *s;
/*
* Read as many input bytes as possible (for
* non-blocking reads)
*/
int read_ahead;
/* where we are when reading */
int rstate;
/* How many pipelines can be used to read data */
unsigned int numrpipes;
/* How many pipelines can be used to write data */
unsigned int numwpipes;
/* read IO goes into here */
SSL3_BUFFER rbuf;
/* write IO goes into here */
SSL3_BUFFER wbuf[SSL_MAX_PIPELINES];
/* each decoded record goes in here */
SSL3_RECORD rrec[SSL_MAX_PIPELINES];
/* used internally to point at a raw packet */
unsigned char *packet;
unsigned int packet_length;
/* number of bytes sent so far */
unsigned int wnum;
/*
* storage for Alert/Handshake protocol data received but not yet
* processed by ssl3_read_bytes:
*/
unsigned char alert_fragment[2];
unsigned int alert_fragment_len;
unsigned char handshake_fragment[4];
unsigned int handshake_fragment_len;
/* The number of consecutive empty records we have received */
unsigned int empty_record_count;
/* partial write - check the numbers match */
/* number bytes written */
int wpend_tot;
int wpend_type;
/* number of bytes submitted */
int wpend_ret;
const unsigned char *wpend_buf;
DTLSv1_listen rewrite The existing implementation of DTLSv1_listen() is fundamentally flawed. This function is used in DTLS solutions to listen for new incoming connections from DTLS clients. A client will send an initial ClientHello. The server will respond with a HelloVerifyRequest containing a unique cookie. The client the responds with a second ClientHello - which this time contains the cookie. Once the cookie has been verified then DTLSv1_listen() returns to user code, which is typically expected to continue the handshake with a call to (for example) SSL_accept(). Whilst listening for incoming ClientHellos, the underlying BIO is usually in an unconnected state. Therefore ClientHellos can come in from *any* peer. The arrival of the first ClientHello without the cookie, and the second one with it, could be interspersed with other intervening messages from different clients. The whole purpose of this mechanism is as a defence against DoS attacks. The idea is to avoid allocating state on the server until the client has verified that it is capable of receiving messages at the address it claims to come from. However the existing DTLSv1_listen() implementation completely fails to do this. It attempts to super-impose itself on the standard state machine and reuses all of this code. However the standard state machine expects to operate in a stateful manner with a single client, and this can cause various problems. A second more minor issue is that the return codes from this function are quite confused, with no distinction made between fatal and non-fatal errors. Most user code treats all errors as non-fatal, and simply retries the call to DTLSv1_listen(). This commit completely rewrites the implementation of DTLSv1_listen() and provides a stand alone implementation that does not rely on the existing state machine. It also provides more consistent return codes. Reviewed-by: Andy Polyakov <appro@openssl.org>
2015-09-14 21:49:35 +00:00
unsigned char read_sequence[SEQ_NUM_SIZE];
unsigned char write_sequence[SEQ_NUM_SIZE];
DTLS_RECORD_LAYER *d;
} RECORD_LAYER;
/*****************************************************************************
* *
* The following macros/functions represent the libssl internal API to the *
* record layer. Any libssl code may call these functions/macros *
* *
*****************************************************************************/
#define MIN_SSL2_RECORD_LEN 9
#define RECORD_LAYER_set_read_ahead(rl, ra) ((rl)->read_ahead = (ra))
#define RECORD_LAYER_get_read_ahead(rl) ((rl)->read_ahead)
#define RECORD_LAYER_get_packet(rl) ((rl)->packet)
#define RECORD_LAYER_get_packet_length(rl) ((rl)->packet_length)
#define RECORD_LAYER_add_packet_length(rl, inc) ((rl)->packet_length += (inc))
#define DTLS_RECORD_LAYER_get_w_epoch(rl) ((rl)->d->w_epoch)
#define DTLS_RECORD_LAYER_get_processed_rcds(rl) \
((rl)->d->processed_rcds)
#define DTLS_RECORD_LAYER_get_unprocessed_rcds(rl) \
((rl)->d->unprocessed_rcds)
void RECORD_LAYER_init(RECORD_LAYER *rl, SSL *s);
void RECORD_LAYER_clear(RECORD_LAYER *rl);
void RECORD_LAYER_release(RECORD_LAYER *rl);
int RECORD_LAYER_read_pending(const RECORD_LAYER *rl);
int RECORD_LAYER_write_pending(const RECORD_LAYER *rl);
int RECORD_LAYER_set_data(RECORD_LAYER *rl, const unsigned char *buf, int len);
void RECORD_LAYER_reset_read_sequence(RECORD_LAYER *rl);
void RECORD_LAYER_reset_write_sequence(RECORD_LAYER *rl);
int RECORD_LAYER_is_sslv2_record(RECORD_LAYER *rl);
unsigned int RECORD_LAYER_get_rrec_length(RECORD_LAYER *rl);
__owur int ssl3_pending(const SSL *s);
__owur int ssl3_write_bytes(SSL *s, int type, const void *buf, int len);
__owur int do_ssl3_write(SSL *s, int type, const unsigned char *buf,
unsigned int *pipelens, unsigned int numpipes,
int create_empty_fragment);
__owur int ssl3_read_bytes(SSL *s, int type, int *recvd_type,
unsigned char *buf, int len, int peek);
__owur int ssl3_setup_buffers(SSL *s);
__owur int ssl3_enc(SSL *s, SSL3_RECORD *inrecs, unsigned int n_recs, int send);
__owur int n_ssl3_mac(SSL *ssl, SSL3_RECORD *rec, unsigned char *md, int send);
__owur int ssl3_write_pending(SSL *s, int type, const unsigned char *buf,
unsigned int len);
__owur int tls1_enc(SSL *s, SSL3_RECORD *recs, unsigned int n_recs, int send);
__owur int tls1_mac(SSL *ssl, SSL3_RECORD *rec, unsigned char *md, int send);
int DTLS_RECORD_LAYER_new(RECORD_LAYER *rl);
void DTLS_RECORD_LAYER_free(RECORD_LAYER *rl);
void DTLS_RECORD_LAYER_clear(RECORD_LAYER *rl);
void DTLS_RECORD_LAYER_set_saved_w_epoch(RECORD_LAYER *rl, unsigned short e);
void DTLS_RECORD_LAYER_clear(RECORD_LAYER *rl);
void DTLS_RECORD_LAYER_resync_write(RECORD_LAYER *rl);
DTLSv1_listen rewrite The existing implementation of DTLSv1_listen() is fundamentally flawed. This function is used in DTLS solutions to listen for new incoming connections from DTLS clients. A client will send an initial ClientHello. The server will respond with a HelloVerifyRequest containing a unique cookie. The client the responds with a second ClientHello - which this time contains the cookie. Once the cookie has been verified then DTLSv1_listen() returns to user code, which is typically expected to continue the handshake with a call to (for example) SSL_accept(). Whilst listening for incoming ClientHellos, the underlying BIO is usually in an unconnected state. Therefore ClientHellos can come in from *any* peer. The arrival of the first ClientHello without the cookie, and the second one with it, could be interspersed with other intervening messages from different clients. The whole purpose of this mechanism is as a defence against DoS attacks. The idea is to avoid allocating state on the server until the client has verified that it is capable of receiving messages at the address it claims to come from. However the existing DTLSv1_listen() implementation completely fails to do this. It attempts to super-impose itself on the standard state machine and reuses all of this code. However the standard state machine expects to operate in a stateful manner with a single client, and this can cause various problems. A second more minor issue is that the return codes from this function are quite confused, with no distinction made between fatal and non-fatal errors. Most user code treats all errors as non-fatal, and simply retries the call to DTLSv1_listen(). This commit completely rewrites the implementation of DTLSv1_listen() and provides a stand alone implementation that does not rely on the existing state machine. It also provides more consistent return codes. Reviewed-by: Andy Polyakov <appro@openssl.org>
2015-09-14 21:49:35 +00:00
void DTLS_RECORD_LAYER_set_write_sequence(RECORD_LAYER *rl, unsigned char *seq);
__owur int dtls1_read_bytes(SSL *s, int type, int *recvd_type,
unsigned char *buf, int len, int peek);
__owur int dtls1_write_bytes(SSL *s, int type, const void *buf, int len);
__owur int do_dtls1_write(SSL *s, int type, const unsigned char *buf,
unsigned int len, int create_empty_fragement);
void dtls1_reset_seq_numbers(SSL *s, int rw);