)]}' { "log": [ { "commit": "645ca708f936b2fbeb79e52d7823e3eb2c0905f8", "tree": "b384696994ee3cb04759a7bfffc29a48e4bf40f6", "parents": [ "b189db5d299c6824780af5590564ff608adb3dea" ], "author": { "name": "Eric Dumazet", "email": "dada1@cosmosbay.com", "time": "Wed Oct 29 01:41:45 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 29 01:41:45 2008 -0700" }, "message": "udp: introduce struct udp_table and multiple spinlocks\n\nUDP sockets are hashed in a 128 slots hash table.\n\nThis hash table is protected by *one* rwlock.\n\nThis rwlock is readlocked each time an incoming UDP message is handled.\n\nThis rwlock is writelocked each time a socket must be inserted in\nhash table (bind time), or deleted from this table (close time)\n\nThis is not scalable on SMP machines :\n\n1) Even in read mode, lock() and unlock() are atomic operations and\n must dirty a contended cache line, shared by all cpus.\n\n2) A writer might be starved if many readers are \u0027in flight\u0027. This can\n happen on a machine with some NIC receiving many UDP messages. User\n process can be delayed a long time at socket creation/dismantle time.\n\nThis patch prepares RCU migration, by introducing \u0027struct udp_table\nand struct udp_hslot\u0027, and using one spinlock per chain, to reduce\ncontention on central rwlock.\n\nIntroducing one spinlock per chain reduces latencies, for port\nrandomization on heavily loaded UDP servers. This also speedup\nbindings to specific ports.\n\nudp_lib_unhash() was uninlined, becoming to big.\n\nSome cleanups were done to ease review of following patch\n(RCUification of UDP Unicast lookups)\n\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1781f7f5804e52ee2d35328b129602146a8d8254", "tree": "dc67102f480e59d87859c50392452be4f41487bd", "parents": [ "27ab2568649d5ba6c5a20212079b7c4f6da4ca0d" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Tue Dec 11 11:30:32 2007 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 28 14:56:33 2008 -0800" }, "message": "[UDP]: Restore missing inDatagrams increments\n\nThe previous move of the the UDP inDatagrams counter caused the\ncounting of encapsulated packets, SUNRPC data (as opposed to call)\npackets and RXRPC packets to go missing.\n\nThis patch restores all of these.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "df2bc459a3ad71f8b44c358bf7169acf9caf4acd", "tree": "143206b16d59c723be3c2cf2375a33b2a9306280", "parents": [ "3c0d2f3780fc94746c4842e965bd2570e2119bb6" ], "author": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Tue Jun 05 15:18:43 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Thu Jun 07 13:40:50 2007 -0700" }, "message": "[UDP]: Revert 2-pass hashing changes.\n\nThis reverts changesets:\n\n6aaf47fa48d3c44280810b1b470261d340e4ed87\nb7b5f487ab39bc10ed0694af35651a03d9cb97ff\nde34ed91c4ffa4727964a832c46e624dd1495cf5\nfc038410b4b1643766f8033f4940bcdb1dace633\n\nThere are still some correctness issues recently\ndiscovered which do not have a known fix that doesn\u0027t\ninvolve doing a full hash table scan on port bind.\n\nSo revert for now.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fc038410b4b1643766f8033f4940bcdb1dace633", "tree": "3ee59190ecaa77061a9b64cdc09fcce6b6efc389", "parents": [ "a2af421f1819946556c6f467b1efdd0dc84af4d5" ], "author": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed May 09 16:42:20 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Thu May 10 23:47:22 2007 -0700" }, "message": "[UDP]: Fix AF-specific references in AF-agnostic code.\n\n__udp_lib_port_inuse() cannot make direct references to\ninet_sk(sk)-\u003ercv_saddr as that is ipv4 specific state and\nthis code is used by ipv6 too.\n\nUse an operations vector to solve this, and this also paves\nthe way for ipv6 support for non-wild saddr hashing in UDP.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4bedb45203eab92a87b4c863fe2d0cded633427f", "tree": "d46e43d0a08253cf85ee2db64df6aee5e4ff5a07", "parents": [ "d9edf9e2be0f7661558984c32bd53867a7037fd3" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Tue Mar 13 14:28:48 2007 -0300" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Apr 25 22:25:22 2007 -0700" }, "message": "[SK_BUFF]: Introduce udp_hdr(), remove skb-\u003eh.uh\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ea2ae17d6443abddc79480dc9f7af8feacabddc4", "tree": "2d6f48a5e4a40f761b5b510af9aac1fca55004cb", "parents": [ "badff6d01a8589a1c828b0bf118903ca38627f4e" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Wed Apr 25 17:55:53 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Apr 25 22:25:16 2007 -0700" }, "message": "[SK_BUFF]: Introduce skb_transport_offset()\n\nFor the quite common \u0027skb-\u003eh.raw - skb-\u003edata\u0027 sequence.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "759e5d006462d53fb708daa8284b4ad909415da1", "tree": "edcc4e9d975199b3fe5e2aadc3d1e06824755e75", "parents": [ "1ab6eb62b02e0949a392fb19bf31ba59ae1022b1" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Sun Mar 25 20:10:56 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Apr 25 22:23:51 2007 -0700" }, "message": "[UDP]: Clean up UDP-Lite receive checksum\n\nThis patch eliminates some duplicate code for the verification of\nreceive checksums between UDP-Lite and UDP. It does this by\nintroducing __skb_checksum_complete_head which is identical to\n__skb_checksum_complete_head apart from the fact that it takes\na length parameter rather than computing the first skb-\u003elen bytes.\n\nAs a result UDP-Lite will be able to use hardware checksum offload\nfor packets which do not use partial coverage checksums. It also\nmeans that UDP-Lite loopback no longer does unnecessary checksum\nverification.\n\nIf any NICs start support UDP-Lite this would also start working\nautomatically.\n\nThis patch removes the assumption that msg_flags has MSG_TRUNC clear\nupon entry in recvmsg.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8e5200f54062b8af0ed1d186ea0f113854786d89", "tree": "71e00b21ae4f8410413f70141aa275529dbbfb84", "parents": [ "34bcca28335977e969338c98c6c43a1e08f592b2" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Nov 20 18:06:37 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:27:16 2006 -0800" }, "message": "[NET]: Fix assorted misannotations (from md5 and udplite merges).\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6bb100b9fc8f1ce330231b360028ab705a9f0378", "tree": "215b375c368d8ec38df89f97dc0a2e240cea2029", "parents": [ "078250d68d883495f31d0e6ce8321292f2dc1cfa" ], "author": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Nov 15 01:09:32 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:23:48 2006 -0800" }, "message": "[UDPLite]: udplite.h needs ip6_checksum.h\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "868c86bcb5bdea7ed8d45979b17bb919af9254db", "tree": "ebc232092d6a8a75261a9e3f698206cc09740fdf", "parents": [ "2bda2853150e20ca2a44627d00b5f85af0b24a42" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Nov 14 21:35:48 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:23:31 2006 -0800" }, "message": "[NET]: annotate csum_ipv6_magic() callers in net/*\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ba4e58eca8aa9473b44fdfd312f26c4a2e7798b3", "tree": "700f8f989f48da480beb83b983637cfd2b5a3f67", "parents": [ "6051e2f4fb68fc8e5343db58fa680ece376f405c" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 11:10:57 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:22:46 2006 -0800" }, "message": "[NET]: Supporting UDP-Lite (RFC 3828) in Linux\n\nThis is a revision of the previously submitted patch, which alters\nthe way files are organized and compiled in the following manner:\n\n\t* UDP and UDP-Lite now use separate object files\n\t* source file dependencies resolved via header files\n\t net/ipv{4,6}/udp_impl.h\n\t* order of inclusion files in udp.c/udplite.c adapted\n\t accordingly\n\n[NET/IPv4]: Support for the UDP-Lite protocol (RFC 3828)\n\nThis patch adds support for UDP-Lite to the IPv4 stack, provided as an\nextension to the existing UDPv4 code:\n * generic routines are all located in net/ipv4/udp.c\n * UDP-Lite specific routines are in net/ipv4/udplite.c\n * MIB/statistics support in /proc/net/snmp and /proc/net/udplite\n * shared API with extensions for partial checksum coverage\n\n[NET/IPv6]: Extension for UDP-Lite over IPv6\n\nIt extends the existing UDPv6 code base with support for UDP-Lite\nin the same manner as per UDPv4. In particular,\n * UDPv6 generic and shared code is in net/ipv6/udp.c\n * UDP-Litev6 specific extensions are in net/ipv6/udplite.c\n * MIB/statistics support in /proc/net/snmp6 and /proc/net/udplite6\n * support for IPV6_ADDRFORM\n * aligned the coding style of protocol initialisation with af_inet6.c\n * made the error handling in udpv6_queue_rcv_skb consistent;\n to return `-1\u0027 on error on all error cases\n * consolidation of shared code\n\n[NET]: UDP-Lite Documentation and basic XFRM/Netfilter support\n\nThe UDP-Lite patch further provides\n * API documentation for UDP-Lite\n * basic xfrm support\n * basic netfilter support for IPv4 and IPv6 (LOG target)\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" } ] }