[PATCH 00/31] net/tcp: Add TCP-AO support

From: Dmitry Safonov
Date: Thu Aug 18 2022 - 13:01:19 EST

Next message: Dmitry Safonov: "[PATCH 05/31] net/tcp: Use crypto_pool for TCP-MD5"
Previous message: Darrick J. Wong: "[PATCH] xfs: on memory failure, only shut down fs after scanning all mappings"
Next in thread: Dmitry Safonov: "[PATCH 04/31] net/tcp: Disable TCP-MD5 static key on tcp_md5sig_info destruction"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This patchset implements the TCP-AO option as described in RFC5925. There
is a request from industry to move away from TCP-MD5SIG and it seems the time
is right to have a TCP-AO upstreamed. This TCP option is meant to replace
the TCP MD5 option and address its shortcomings. Specifically, it provides
more secure hashing, key rotation and support for long-lived connections
(see the summary of TCP-AO advantages over TCP-MD5 in (1.3) of RFC5925).
The patch series starts with six patches that are not specific to TCP-AO
but implement a general crypto facility that we thought is useful
to eliminate code duplication between TCP-MD5SIG and TCP-AO as well as other
crypto users. These six patches are being submitted separately in
a different patchset [1]. Including them here will show better the gain
in code sharing. Next are 18 patches that implement the actual TCP-AO option,
followed by patches implementing selftests.

The patch set was written as a collaboration of three authors (in alphabetical
order): Dmitry Safonov, Francesco Ruggeri and Salam Noureddine. Additional
credits should be given to Prasad Koya, who was involved in early prototyping
a few years back. There is also a separate submission done by Leonard Crestez
whom we thank for his efforts getting an implementation of RFC5925 submitted
for review upstream [2]. This is an independent implementation that makes
different design decisions.

For example, we chose a similar design to the TCP-MD5SIG implementation and
used setsockopt()s to program per-socket keys, avoiding the extra complexity
of managing a centralized key database in the kernel. A centralized database
in the kernel has dubious benefits since it doesn’t eliminate per-socket
setsockopts needed to specify which sockets need TCP-AO and what are the
currently preferred keys. It also complicates traffic key caching and
preventing deletion of in-use keys.

In this implementation, a centralized database of keys can be thought of
as living in user space and user applications would have to program those
keys on matching sockets. On the server side, the user application programs
keys (MKTS in TCP-AO nomenclature) on the listening socket for all peers that
are expected to connect. Prefix matching on the peer address is supported.
When a peer issues a successful connect, all the MKTs matching the IP address
of the peer are copied to the newly created socket. On the active side,
when a connect() is issued all MKTs that do not match the peer are deleted
from the socket since they will never match the peer. This implementation
uses three setsockopt()s for adding, deleting and modifying keys on a socket.
All three setsockopt()s have extensive sanity checks that prevent
inconsistencies in the keys on a given socket. A getsockopt() is provided
to get key information from any given socket.

Few things to note about this implementation:
- Traffic keys are cached for established connections avoiding the cost of
such calculation for each packet received or sent.
- Great care has been taken to avoid deleting in-use MKTs
as required by the RFC.
- Any crypto algorithm supported by the Linux kernel can be used
to calculate packet hashes.
- Fastopen works with TCP-AO but hasn’t been tested extensively.
- Tested for interop with other major networking vendors (on linux-4.19),
including testing for key rotation and long lived connections.

There are a couple of limitations that we’re aware of, including (but not
limited to) the following:
- setsockopt(TCP_REPAIR) not supported yet
- IPv4-mapped-IPv6 addresses not tested
- static key not implemented yet
- CONFIG_TCP_AO depends on CONFIG_TCP_MD5SIG
- A small window for a race condition exists between accept and key
adding/deletion on a listening socket but can be easily overcome by using
the getsockopt() to make sure the right keys are there on a newly accepted
connection
- Key matching by TCP port numbers, peer ranges, asterisks is unsupported
as it’s unlikely to be useful

[1]: https://lore.kernel.org/all/20220726201600.1715505-1-dima@xxxxxxxxxx/
[2]: https://lore.kernel.org/all/cover.1658815925.git.cdleonard@xxxxxxxxx/

Cc: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Cc: Ard Biesheuvel <ardb@xxxxxxxxxx>
Cc: Bob Gilligan <gilligan@xxxxxxxxxx>
Cc: David Ahern <dsahern@xxxxxxxxxx>
Cc: "David S. Miller" <davem@xxxxxxxxxxxxx>
Cc: Dmitry Safonov <0x7f454c46@xxxxxxxxx>
Cc: Eric Biggers <ebiggers@xxxxxxxxxx>
Cc: Eric Dumazet <edumazet@xxxxxxxxxx>
Cc: Francesco Ruggeri <fruggeri@xxxxxxxxxx>
Cc: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Cc: Hideaki YOSHIFUJI <yoshfuji@xxxxxxxxxxxxxx>
Cc: Ivan Delalande <colona@xxxxxxxxxx>
Cc: Jakub Kicinski <kuba@xxxxxxxxxx>
Cc: Leonard Crestez <cdleonard@xxxxxxxxx>
Cc: Paolo Abeni <pabeni@xxxxxxxxxx>
Cc: Salam Noureddine <noureddine@xxxxxxxxxx>
Cc: Shuah Khan <shuah@xxxxxxxxxx>
Cc: netdev@xxxxxxxxxxxxxxx
Cc: linux-crypto@xxxxxxxxxxxxxxx
Cc: linux-kernel@xxxxxxxxxxxxxxx

Dmitry Safonov (31):
crypto: Introduce crypto_pool
crypto_pool: Add crypto_pool_reserve_scratch()
net/tcp: Separate tcp_md5sig_info allocation into tcp_md5sig_info_add()
net/tcp: Disable TCP-MD5 static key on tcp_md5sig_info destruction
net/tcp: Use crypto_pool for TCP-MD5
net/ipv6: sr: Switch to using crypto_pool
tcp: Add TCP-AO config and structures
net/tcp: Introduce TCP_AO setsockopt()s
net/tcp: Prevent TCP-MD5 with TCP-AO being set
net/tcp: Calculate TCP-AO traffic keys
net/tcp: Add TCP-AO sign to outgoing packets
net/tcp: Add tcp_parse_auth_options()
net/tcp: Add AO sign to RST packets
net/tcp: Add TCP-AO sign to twsk
net/tcp: Wire TCP-AO to request sockets
net/tcp: Sign SYN-ACK segments with TCP-AO
net/tcp: Verify inbound TCP-AO signed segments
net/tcp: Add TCP-AO segments counters
net/tcp: Add TCP-AO SNE support
net/tcp: Add tcp_hash_fail() ratelimited logs
net/tcp: Ignore specific ICMPs for TCP-AO connections
net/tcp: Add option for TCP-AO to (not) hash header
net/tcp: Add getsockopt(TCP_AO_GET)
net/tcp: Allow asynchronous delete for TCP-AO keys (MKTs)
selftests/net: Add TCP-AO library
selftests/net: Verify that TCP-AO complies with ignoring ICMPs
selftest/net: Add TCP-AO ICMPs accept test
selftest/tcp-ao: Add a test for MKT matching
selftest/tcp-ao: Add test for TCP-AO add setsockopt() command
selftests/tcp-ao: Add TCP-AO + TCP-MD5 + no sign listen socket tests
selftests/aolib: Add test/benchmark for removing MKTs

crypto/Kconfig | 12 +
crypto/Makefile | 1 +
crypto/crypto_pool.c | 323 +++
include/crypto/pool.h | 33 +
include/linux/sockptr.h | 23 +
include/linux/tcp.h | 24 +
include/net/dropreason.h | 25 +
include/net/seg6_hmac.h | 7 -
include/net/tcp.h | 193 +-
include/net/tcp_ao.h | 283 +++
include/uapi/linux/snmp.h | 5 +
include/uapi/linux/tcp.h | 62 +
net/ipv4/Kconfig | 15 +-
net/ipv4/Makefile | 1 +
net/ipv4/proc.c | 5 +
net/ipv4/tcp.c | 191 +-
net/ipv4/tcp_ao.c | 1939 +++++++++++++++++
net/ipv4/tcp_input.c | 94 +-
net/ipv4/tcp_ipv4.c | 385 +++-
net/ipv4/tcp_minisocks.c | 37 +-
net/ipv4/tcp_output.c | 188 +-
net/ipv6/Kconfig | 2 +-
net/ipv6/Makefile | 1 +
net/ipv6/seg6.c | 3 -
net/ipv6/seg6_hmac.c | 204 +-
net/ipv6/tcp_ao.c | 151 ++
net/ipv6/tcp_ipv6.c | 327 ++-
tools/testing/selftests/Makefile | 1 +
tools/testing/selftests/net/tcp_ao/.gitignore | 2 +
tools/testing/selftests/net/tcp_ao/Makefile | 50 +
.../selftests/net/tcp_ao/bench-lookups.c | 403 ++++
.../selftests/net/tcp_ao/connect-deny.c | 217 ++
tools/testing/selftests/net/tcp_ao/connect.c | 81 +
.../selftests/net/tcp_ao/icmps-accept.c | 1 +
.../selftests/net/tcp_ao/icmps-discard.c | 447 ++++
.../testing/selftests/net/tcp_ao/lib/aolib.h | 333 +++
.../selftests/net/tcp_ao/lib/netlink.c | 341 +++
tools/testing/selftests/net/tcp_ao/lib/proc.c | 267 +++
.../testing/selftests/net/tcp_ao/lib/setup.c | 297 +++
tools/testing/selftests/net/tcp_ao/lib/sock.c | 294 +++
.../testing/selftests/net/tcp_ao/lib/utils.c | 30 +
.../selftests/net/tcp_ao/setsockopt-closed.c | 191 ++
.../selftests/net/tcp_ao/unsigned-md5.c | 483 ++++
43 files changed, 7516 insertions(+), 456 deletions(-)
create mode 100644 crypto/crypto_pool.c
create mode 100644 include/crypto/pool.h
create mode 100644 include/net/tcp_ao.h
create mode 100644 net/ipv4/tcp_ao.c
create mode 100644 net/ipv6/tcp_ao.c
create mode 100644 tools/testing/selftests/net/tcp_ao/.gitignore
create mode 100644 tools/testing/selftests/net/tcp_ao/Makefile
create mode 100644 tools/testing/selftests/net/tcp_ao/bench-lookups.c
create mode 100644 tools/testing/selftests/net/tcp_ao/connect-deny.c
create mode 100644 tools/testing/selftests/net/tcp_ao/connect.c
create mode 120000 tools/testing/selftests/net/tcp_ao/icmps-accept.c
create mode 100644 tools/testing/selftests/net/tcp_ao/icmps-discard.c
create mode 100644 tools/testing/selftests/net/tcp_ao/lib/aolib.h
create mode 100644 tools/testing/selftests/net/tcp_ao/lib/netlink.c
create mode 100644 tools/testing/selftests/net/tcp_ao/lib/proc.c
create mode 100644 tools/testing/selftests/net/tcp_ao/lib/setup.c
create mode 100644 tools/testing/selftests/net/tcp_ao/lib/sock.c
create mode 100644 tools/testing/selftests/net/tcp_ao/lib/utils.c
create mode 100644 tools/testing/selftests/net/tcp_ao/setsockopt-closed.c
create mode 100644 tools/testing/selftests/net/tcp_ao/unsigned-md5.c

base-commit: e34cfee65ec891a319ce79797dda18083af33a76
--
2.37.2

Next message: Dmitry Safonov: "[PATCH 05/31] net/tcp: Use crypto_pool for TCP-MD5"
Previous message: Darrick J. Wong: "[PATCH] xfs: on memory failure, only shut down fs after scanning all mappings"
Next in thread: Dmitry Safonov: "[PATCH 04/31] net/tcp: Disable TCP-MD5 static key on tcp_md5sig_info destruction"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]