| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <1455123041-22638-1-git-send-email-kraigatgoog@gmail.com> Date: Wed, 10 Feb 2016 11:50:34 -0500 From: Craig Gallek <kraigatgoog@...il.com> To: netdev@...r.kernel.org, David Miller <davem@...emloft.net> Subject: [PATCH net-next v4 0/7] Faster SO_REUSEPORT for TCP From: Craig Gallek <kraig@...gle.com> This patch series complements an earlier series (6a5ef90c58da) which added faster SO_REUSEPORT lookup for UDP sockets by extending the feature to TCP sockets. It uses the same array-based data structure which allows for socket selection after finding the first listening socket that matches an incoming packet. Prior to this feature, every socket in the reuseport group needed to be found and examined before a selection could be made. With this series the SO_ATTACH_REUSEPORT_CBPF and SO_ATTACH_REUSEPORT_EBPF socket options now work for TCP sockets as well. The test at the end of the series includes an example of how to use these options to select a reuseport socket based on the cpu core id handling the incoming packet. There are several refactoring patches that precede the feature implementation. Only the last two patches in this series should result in any behavioral changes. v4 - Fix build issue when compiling IPv6 as a module. This required moving the ipv6_rcv_saddr_equal into an object that is included as a built-in object. I included this change in the second patch which adds inet6_hash since that is where ipv6_rcv_saddr_equal will later be called from non-module code. v3: - Another warning in the first patch caught by a build bot. Return 0 in the no-op UDP hash function. v2: - In the first patched I missed a couple of hash functions that should now be returning int instead of void. I missed these the first time through as it only generated a warning and not an error :\ Craig Gallek (7): sock: struct proto hash function may error inet: create IPv6-equivalent inet_hash function tcp: __tcp_hdrlen() helper inet: refactor inet[6]_lookup functions to take skb soreuseport: Prep for fast reuseport TCP socket selection soreuseport: fast reuseport TCP socket selection soreuseport: BPF selection functional test for TCP include/linux/tcp.h | 7 +- include/net/addrconf.h | 2 + include/net/inet6_hashtables.h | 13 +- include/net/inet_hashtables.h | 25 ++- include/net/phonet/phonet.h | 2 +- include/net/ping.h | 2 +- include/net/raw.h | 2 +- include/net/sock.h | 6 +- include/net/udp.h | 3 +- net/core/filter.c | 2 +- net/core/sock.c | 1 + net/dccp/ipv4.c | 2 +- net/dccp/ipv6.c | 4 +- net/ieee802154/socket.c | 17 +- net/ipv4/af_inet.c | 9 +- net/ipv4/inet_connection_sock.c | 22 +- net/ipv4/inet_diag.c | 6 +- net/ipv4/inet_hashtables.c | 67 +++++- net/ipv4/ping.c | 4 +- net/ipv4/raw.c | 4 +- net/ipv4/tcp_ipv4.c | 10 +- net/ipv4/udp.c | 4 +- net/ipv6/af_inet6.c | 6 +- net/ipv6/inet6_connection_sock.c | 2 + net/ipv6/inet6_hashtables.c | 78 ++++++- net/ipv6/tcp_ipv6.c | 10 +- net/ipv6/udp.c | 44 +--- net/l2tp/l2tp_ip6.c | 3 +- net/netfilter/xt_TPROXY.c | 31 ++- net/netfilter/xt_socket.c | 28 ++- net/phonet/socket.c | 6 +- net/sctp/socket.c | 3 +- tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/Makefile | 2 +- tools/testing/selftests/net/reuseport_bpf.c | 117 ++++++++++- tools/testing/selftests/net/reuseport_bpf_cpu.c | 258 ++++++++++++++++++++++++ 36 files changed, 670 insertions(+), 133 deletions(-) create mode 100644 tools/testing/selftests/net/reuseport_bpf_cpu.c -- 2.7.0.rc3.207.g0ac5344
Powered by blists - more mailing lists