mirrors/tcpdump
0
0
Fork 0
mirror of https://github.com/the-tcpdump-group/tcpdump.git synced 2024-12-18 06:14:06 +08:00
Code Issues Packages Projects Releases Wiki Activity
4ac279241d
tcpdump/print-mptcp.c

429 lines
13 KiB
C
Raw Normal View History

Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
/**
* Copyright (c) 2012
*
* Gregory Detal <gregory.detal@uclouvain.be>
* Christoph Paasch <christoph.paasch@uclouvain.be>
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* 3. Neither the name of the University nor of the Laboratory may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include <tcpdump-stdinc.h>
#include "interface.h"
#include "extract.h"
#include "addrtoname.h"
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
#include "tcp.h"
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
#define MPTCP_SUB_CAPABLE 0x0
#define MPTCP_SUB_JOIN 0x1
#define MPTCP_SUB_DSS 0x2
#define MPTCP_SUB_ADD_ADDR 0x3
#define MPTCP_SUB_REMOVE_ADDR 0x4
#define MPTCP_SUB_PRIO 0x5
#define MPTCP_SUB_FAIL 0x6
#define MPTCP_SUB_FCLOSE 0x7
struct mptcp_option {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub_etc; /* subtype upper 4 bits, other stuff lower 4 bits */
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
#define MPTCP_OPT_SUBTYPE(sub_etc) (((sub_etc) >> 4) & 0xF)
struct mp_capable {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub_ver;
uint8_t flags;
uint8_t sender_key[8];
uint8_t receiver_key[8];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
#define MP_CAPABLE_OPT_VERSION(sub_ver) (((sub_ver) >> 0) & 0xF)
#define MP_CAPABLE_C 0x80
#define MP_CAPABLE_S 0x01
struct mp_join {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub_b;
uint8_t addr_id;
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
union {
struct {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t token[4];
uint8_t nonce[4];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
} syn;
struct {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t mac[8];
uint8_t nonce[4];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
} synack;
struct {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t mac[20];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
} ack;
} u;
};
#define MP_JOIN_B 0x01
struct mp_dss {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub;
uint8_t flags;
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
#define MP_DSS_F 0x10
#define MP_DSS_m 0x08
#define MP_DSS_M 0x04
#define MP_DSS_a 0x02
#define MP_DSS_A 0x01
struct mp_add_addr {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub_ipver;
uint8_t addr_id;
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
union {
struct {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t addr[4];
uint8_t port[2];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
} v4;
struct {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t addr[16];
uint8_t port[2];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
} v6;
} u;
};
#define MP_ADD_ADDR_IPVER(sub_ipver) (((sub_ipver) >> 0) & 0xF)
struct mp_remove_addr {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub;
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
/* list of addr_id */
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t addrs_id;
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
struct mp_fail {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub;
uint8_t resv;
uint8_t data_seq[8];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
struct mp_close {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub;
uint8_t rsv;
uint8_t key[8];
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
struct mp_prio {
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t kind;
uint8_t len;
uint8_t sub_b;
uint8_t addr_id;
Pull a bunch of headers into the only source file that includes them. For headers included in only one source file, put the header contents in the source file in question, and get rid of a bunch of stuff from the header not used in the source file.
2013-12-31 14:52:15 +08:00
};
#define MP_PRIO_B 0x01
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
dummy_print(netdissect_options *ndo _U_,
const u_char *opt _U_, u_int opt_len _U_, u_char flags _U_)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
return 1;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
mp_capable_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
struct mp_capable *mpc = (struct mp_capable *) opt;
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (!(opt_len == 12 && flags & TH_SYN) &&
!(opt_len == 20 && (flags & (TH_SYN | TH_ACK)) == TH_ACK))
return 0;
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (MP_CAPABLE_OPT_VERSION(mpc->sub_ver) != 0) {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " Unknown Version (%d)", MP_CAPABLE_OPT_VERSION(mpc->sub_ver)));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
return 1;
}
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mpc->flags & MP_CAPABLE_C)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " csum"));
ND_PRINT((ndo, " {0x%" PRIx64, EXTRACT_64BITS(mpc->sender_key)));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (opt_len == 20) /* ACK */
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, ",0x%" PRIx64, EXTRACT_64BITS(mpc->receiver_key)));
ND_PRINT((ndo, "}"));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
return 1;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
mp_join_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
struct mp_join *mpj = (struct mp_join *) opt;
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (!(opt_len == 12 && flags & TH_SYN) &&
!(opt_len == 16 && (flags & (TH_SYN | TH_ACK)) == (TH_SYN | TH_ACK)) &&
!(opt_len == 24 && flags & TH_ACK))
return 0;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (opt_len != 24) {
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mpj->sub_b & MP_JOIN_B)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " backup"));
ND_PRINT((ndo, " id %u", mpj->addr_id));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
}
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
switch (opt_len) {
case 12: /* SYN */
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " token 0x%x" " nonce 0x%x",
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
EXTRACT_32BITS(mpj->u.syn.token),
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
EXTRACT_32BITS(mpj->u.syn.nonce)));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
break;
case 16: /* SYN/ACK */
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " hmac 0x%" PRIx64 " nonce 0x%x",
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
EXTRACT_64BITS(mpj->u.synack.mac),
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
EXTRACT_32BITS(mpj->u.synack.nonce)));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
break;
case 24: {/* ACK */
Fix some compiler warnings from the previous checkin.
2013-05-06 09:31:48 +08:00
size_t i;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " hmac 0x"));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
for (i = 0; i < sizeof(mpj->u.ack.mac); ++i)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, "%02x", mpj->u.ack.mac[i]));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
}
default:
break;
}
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
return 1;
}
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
static u_int mp_dss_len(struct mp_dss *m, int csum)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
u_int len;
len = 4;
if (m->flags & MP_DSS_A) {
/* Ack present - 4 or 8 octets */
len += (m->flags & MP_DSS_a) ? 8 : 4;
}
if (m->flags & MP_DSS_M) {
/*
* Data Sequence Number (DSN), Subflow Sequence Number (SSN),
* Data-Level Length present, and Checksum possibly present.
* All but the Checksum are 10 bytes if the m flag is
* clear (4-byte DSN) and 14 bytes if the m flag is set
* (8-byte DSN).
*/
len += (m->flags & MP_DSS_m) ? 14 : 10;
/*
* The Checksum is present only if negotiated.
*/
if (csum)
len += 2;
}
return len;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
mp_dss_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
struct mp_dss *mdss = (struct mp_dss *) opt;
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if ((opt_len != mp_dss_len(mdss, 1) &&
opt_len != mp_dss_len(mdss, 0)) || flags & TH_SYN)
return 0;
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mdss->flags & MP_DSS_F)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " fin"));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
opt += 4;
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mdss->flags & MP_DSS_A) {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " ack "));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mdss->flags & MP_DSS_a) {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, "%" PRIu64, EXTRACT_64BITS(opt)));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
opt += 8;
} else {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, "%u", EXTRACT_32BITS(opt)));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
opt += 4;
}
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
}
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mdss->flags & MP_DSS_M) {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " seq "));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mdss->flags & MP_DSS_m) {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, "%" PRIu64, EXTRACT_64BITS(opt)));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
opt += 8;
} else {
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, "%u", EXTRACT_32BITS(opt)));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
opt += 4;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " subseq %u", EXTRACT_32BITS(opt)));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
opt += 4;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " len %u", EXTRACT_16BITS(opt)));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
opt += 2;
if (opt_len == mp_dss_len(mdss, 1))
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " csum 0x%x", EXTRACT_16BITS(opt)));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
}
return 1;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
add_addr_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags _U_)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
struct mp_add_addr *add_addr = (struct mp_add_addr *) opt;
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
u_int ipver = MP_ADD_ADDR_IPVER(add_addr->sub_ipver);
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (!((opt_len == 8 || opt_len == 10) && ipver == 4) &&
!((opt_len == 20 || opt_len == 22) && ipver == 6))
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
return 0;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " id %u", add_addr->addr_id));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
switch (ipver) {
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
case 4:
Netdissectify the to-name resolution routines. Have them take a netdissect_options * argument, and get the "no name resolution" flag from it. Move the declaration of dnaddr_string to addrtoname.h, along with the other XXX-to-string routines.
2014-04-04 15:43:46 +08:00
ND_PRINT((ndo, " %s", ipaddr_string(ndo, add_addr->u.v4.addr)));
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (opt_len == 10)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, ":%u", EXTRACT_16BITS(add_addr->u.v4.port)));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
break;
case 6:
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
#ifdef INET6
Netdissectify the to-name resolution routines. Have them take a netdissect_options * argument, and get the "no name resolution" flag from it. Move the declaration of dnaddr_string to addrtoname.h, along with the other XXX-to-string routines.
2014-04-04 15:43:46 +08:00
ND_PRINT((ndo, " %s", ip6addr_string(ndo, add_addr->u.v6.addr)));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
#endif
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (opt_len == 22)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, ":%u", EXTRACT_16BITS(add_addr->u.v6.port)));
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
break;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
default:
return 0;
}
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
return 1;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
remove_addr_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags _U_)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
Rename a variable to attempt to fix an AIX compile error. On AIX 5.1, at least when compiling with xlc, the statement struct mp_remove_addr *rem_addr = (struct mp_remove_addr *) opt; gets a complaint "Syntax error: possible missing '{'?" "/usr/include/sys/xmem.h" #defines rem_addr; I've no idea whether we're indirectly including that, but maybe we are and maybe that's causing the problem.
2013-05-06 10:52:14 +08:00
struct mp_remove_addr *remove_addr = (struct mp_remove_addr *) opt;
u_intN_t is dead, long live uintN_t. And, as we require at least autoconf 2.61, and as autoconf 2.61 and later have AC_TYPE_UINTn_T and AC_TYPE_INTn_T macros, we use them to define the uintN_t and intN_t macros if the system doesn't define them for us. This lets us get rid of bitypes.h as well.
2014-04-23 15:20:40 +08:00
uint8_t *addr_id = &remove_addr->addrs_id;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (opt_len < 4)
return 0;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
opt_len -= 3;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " id"));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
while (opt_len--)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " %u", *addr_id++));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
return 1;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
mp_prio_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags _U_)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
struct mp_prio *mpp = (struct mp_prio *) opt;
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (opt_len != 3 && opt_len != 4)
return 0;
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
if (mpp->sub_b & MP_PRIO_B)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " backup"));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
else
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " non-backup"));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
if (opt_len == 4)
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " id %u", mpp->addr_id));
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
return 1;
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
mp_fail_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags _U_)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (opt_len != 12)
return 0;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " seq %" PRIu64, EXTRACT_64BITS(opt + 4)));
Fix some compiler warnings. Some of those warnings are real bugs - some routines whose callers expected them to return values weren't returning values.
2013-04-15 08:21:19 +08:00
return 1;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
}
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
static int
mp_fast_close_print(netdissect_options *ndo,
const u_char *opt, u_int opt_len, u_char flags _U_)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
if (opt_len != 12)
return 0;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " key 0x%" PRIx64, EXTRACT_64BITS(opt + 4)));
Fix some compiler warnings. Some of those warnings are real bugs - some routines whose callers expected them to return values weren't returning values.
2013-04-15 08:21:19 +08:00
return 1;
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
}
make more array declarations static/const
2013-12-19 18:05:59 +08:00
static const struct {
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
const char *name;
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
int (*print)(netdissect_options *, const u_char *, u_int, u_char);
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
} mptcp_options[] = {
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
{ "capable", mp_capable_print},
{ "join", mp_join_print },
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{ "dss", mp_dss_print },
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
{ "add-addr", add_addr_print },
{ "rem-addr", remove_addr_print },
{ "prio", mp_prio_print },
{ "fail", mp_fail_print },
{ "fast-close", mp_fast_close_print },
{ "unknown", dummy_print },
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
};
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
int
mptcp_print(netdissect_options *ndo,
const u_char *cp, u_int len, u_char flags)
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
{
Fix MPTCP support Added: * option length check * option type and TCP flags check * more information printed Signed-off-by: Gregory Detal <gregory.detal@uclouvain.be>
2013-04-17 22:05:21 +08:00
struct mptcp_option *opt;
u_int subtype;
if (len < 3)
return 0;
opt = (struct mptcp_option *) cp;
Avoid bitfields, unaligned accesses, packed structures, and PRI[ux]{16,32}. Bitfields are not one of C's shining points. There is *NO* guarantee in what order bitfields are put within a structure - it's *NOT* necessarily the same as the byte order of the machine, and it's *ESPECIALLY* not guaranteed to be correlated with the value of the LBL_ALIGN definition (that definition has to do with whether unaligned accesses are supported by the hardware). In addition, even if they're declared as unsigned, that doesn't mean they're guaranteed to *be* unsigned. Don't use them. Unaligned accesses are not guaranteed to work, and fields in packets are not guaranteed to be naturally aligned. Use the EXTRACT_nBITS() macros. __attribute((packed))__ is a GCCism, and is not guaranteed to be supported by all compilers with which tcpdump can be compiled. Make integral fields > 1 byte arrays of u_int8_t's (which also lets us avoid the & in the EXTRACT_nBITS() macros). Some systems don't define the PRI[doux]16 and PRI[doux]32 macros, and others define them infelicitously (i.e., for PRI[doux]32, with an "l"; our 32-bit integer types are *not* longs, as we don't care about 16-bit-"int" platforms).
2013-05-06 09:52:54 +08:00
subtype = min(MPTCP_OPT_SUBTYPE(opt->sub_etc), MPTCP_SUB_FCLOSE + 1);
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
NDOize ATM, MPTCP, NTP, VTP & Whiteboard decoders
2014-03-20 18:39:44 +08:00
ND_PRINT((ndo, " %s", mptcp_options[subtype].name));
return mptcp_options[subtype].print(ndo, cp, len, flags);
Multipath TCP (RFC 6824) support This commit adds the support of Multipath TCP (MPTCP). MPTCP is a new extension to TCP standardized at the IETF. MPTCP allows to use several IP addresses at the same time by distributing data across several subflows (TCP connections) while still presenting the standard TCP socket API to the application. Its benefits are better resource utilization, better throughput and smoother reaction to failures.
2013-02-26 00:00:02 +08:00
}
Reference in New Issue Copy Permalink