)]}' { "log": [ { "commit": "b3fc9c6c58c986f7a24fd8b0794d1e0794935a28", "tree": "1556dbab26c37c438ea6bc980bc05143a10c4c86", "parents": [ "2dd0f69222c481574baf6a4affb9256a7c7410e7" ], "author": { "name": "Johannes Berg", "email": "johannes@sipsolutions.net", "time": "Sun Apr 13 10:12:47 2008 +0200" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Tue Apr 15 15:04:36 2008 -0400" }, "message": "mac80211: remove message on receiving unexpected unencrypted frames\n\nSome people are getting this message a lot, and we have traced it to\nbroken access points that much too often send completely empty frames\n(all bytes zeroed, which they shouldn\u0027t do at all.)\n\nSince we cannot do anything about such frames in any case except the\nspecial case where we\u0027re debugging an AP, just remove the message.\n\nSigned-off-by: Johannes Berg \u003cjohannes@sipsolutions.net\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "89796f64a20d31e74ee0051df2e26812c852e734", "tree": "5430ac55e047fa88c5cc11f1d1f1e8b95ceddf56", "parents": [ "cdbbe3d1f53086ece706674d3bf4f6d148083694" ], "author": { "name": "Carlos Corbacho", "email": "carlos@strangeworlds.co.uk", "time": "Sat Apr 12 16:39:47 2008 +0100" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Tue Apr 15 15:04:35 2008 -0400" }, "message": "rfkill: Fix device type check when toggling states\n\nrfkill_switch_all() is supposed to only switch all the interfaces of a\ngiven type, but does not actually do this; instead, it just switches\neverything currently in the same state.\n\nAdd the necessary type check in.\n\n(This fixes a bug I\u0027ve been seeing while developing an rfkill laptop\ndriver, with both bluetooth and wireless simultaneously changing state\nafter only pressing either KEY_WLAN or KEY_BLUETOOTH).\n\nSigned-off-by: Carlos Corbacho \u003ccarlos@strangeworlds.co.uk\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "b000cd3707e7b25d76745f9c0e261c23d21fa578", "tree": "d86dfd47954404e581f7f9d730ae3d3ce2532ad1", "parents": [ "066a3b5b2346febf9a655b444567b7138e3bb939" ], "author": { "name": "Vitaliy Gusev", "email": "vgusev@openvz.org", "time": "Tue Apr 15 00:33:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 15 00:33:38 2008 -0700" }, "message": "[TCP]: Fix never pruned tcp out-of-order queue.\n\ntcp_prune_queue() doesn\u0027t prune an out-of-order queue at all.\nTherefore sk_rmem_schedule() can fail but the out-of-order queue isn\u0027t\npruned . This can lead to tcp deadlock state if the next two\nconditions are held:\n\n1. There are a sequence hole between last received in\n order segment and segments enqueued to the out-of-order queue.\n\n2. Size of all segments in the out-of-order queue is more than tcp_mem[2].\n\nSigned-off-by: Vitaliy Gusev \u003cvgusev@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "066a3b5b2346febf9a655b444567b7138e3bb939", "tree": "84d824c9b5e31bfaa8a4b06ba3a96f5ac68e14c5", "parents": [ "159d83363b629c91d020734207c1bc788b96af5a" ], "author": { "name": "Jarek Poplawski", "email": "jarkao2@gmail.com", "time": "Mon Apr 14 15:10:42 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 14 15:10:42 2008 -0700" }, "message": "[NET_SCHED] sch_api: fix qdisc_tree_decrease_qlen() loop\n\nTC_H_MAJ(parentid) for root classes is the same as for ingress, and if\ningress qdisc is created qdisc_lookup() returns its pointer (without\ningress NULL is returned). After this all qdisc_lookups give the same,\nand we get endless loop. (I don\u0027t know how this could hide for so long\n- it should trigger with every leaf class deleted if it\u0027s qdisc isn\u0027t\nempty.)\n\nAfter this fix qdisc_lookup() is omitted both for ingress and root\nparents, but looking for root is only wasting a little time here...\nMany thanks to Enrico Demarin for finding a test for catching this\nbug, which probably bothered quite a lot of admins.\n \nReported-by: Enrico Demarin \u003cenrico@superclick.com\u003e,\nSigned-off-by: Jarek Poplawski \u003cjarkao2@gmail.com\u003e\nAcked-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "159d83363b629c91d020734207c1bc788b96af5a", "tree": "9e2bd418f94bed1dc0f62eb6a1f7671055c473b9", "parents": [ "4dee959723e2bf3a0f9343a46841cd2f0029d424" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Mon Apr 14 00:46:01 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 14 00:46:01 2008 -0700" }, "message": "[BRIDGE]: Fix crash in __ip_route_output_key with bridge netfilter\n\nThe bridge netfilter code attaches a fake dst_entry with a pointer to a\nfake net_device structure to skbs it passes up to IPv4 netfilter. This\nleads to crashes when the skb is passed to __ip_route_output_key when\ndereferencing the namespace pointer.\n\nSince bridging can currently only operate in the init_net namespace,\nthe easiest fix for now is to initialize the nd_net pointer of the\nfake net_device struct to \u0026init_net.\n\nShould fix bugzilla 10323: http://bugzilla.kernel.org/show_bug.cgi?id\u003d10323\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4dee959723e2bf3a0f9343a46841cd2f0029d424", "tree": "a7ae0cedd9dccee806b6a7ba6a7e017a271850ca", "parents": [ "9625ed72e8bd619c3984f3024bd37143b7f0c7b0" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Mon Apr 14 00:44:52 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 14 00:44:52 2008 -0700" }, "message": "[NETFILTER]: ipt_CLUSTERIP: fix race between clusterip_config_find_get and _entry_put\n\nConsider we are putting a clusterip_config entry with the \"entries\"\ncount \u003d\u003d 1, and on the other CPU there\u0027s a clusterip_config_find_get\nin progress:\n\nCPU1:\t\t\t\t\t\t\tCPU2:\nclusterip_config_entry_put:\t\t\t\tclusterip_config_find_get:\nif (atomic_dec_and_test(\u0026c-\u003eentries)) {\n\t/* true */\n\t\t\t\t\t\t\tread_lock_bh(\u0026clusterip_lock);\n\t\t\t\t\t\t\tc \u003d __clusterip_config_find(clusterip);\n\t\t\t\t\t\t\t/* found - it\u0027s still in list */\n\t\t\t\t\t\t\t...\n\t\t\t\t\t\t\tatomic_inc(\u0026c-\u003eentries);\n\t\t\t\t\t\t\tread_unlock_bh(\u0026clusterip_lock);\n\n\twrite_lock_bh(\u0026clusterip_lock);\n\tlist_del(\u0026c-\u003elist);\n\twrite_unlock_bh(\u0026clusterip_lock);\n\t...\n\tdev_put(c-\u003edev);\n\nOops! We have an entry returned by the clusterip_config_find_get,\nwhich is a) not in list b) has a stale dev pointer.\n\nThe problems will happen when the CPU2 will release the entry - it\nwill remove it from the list for the 2nd time, thus spoiling it, and\nwill put a stale dev pointer.\n\nThe fix is to make atomic_dec_and_test under the clusterip_lock.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\n" }, { "commit": "9625ed72e8bd619c3984f3024bd37143b7f0c7b0", "tree": "c3952ee1350f84cce42ab7e235c1ce33a587acd6", "parents": [ "b077d7ababdb5433aef18c62bf1f785e8729f49a" ], "author": { "name": "YOSHIFUJI Hideaki", "email": "yoshfuji@linux-ipv6.org", "time": "Sun Apr 13 23:47:11 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 23:47:11 2008 -0700" }, "message": "[IPV6] ADDRCONF: Don\u0027t generate temporary address for ip6-ip6 interface.\n\nAs far as I can remember, I was going to disable privacy extensions\non all \"tunnel\" interfaces. Disable it on ip6-ip6 interface as well.\n\nAlso, just remove ifdefs for SIT for simplicity.\n\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b077d7ababdb5433aef18c62bf1f785e8729f49a", "tree": "0fa561ab57495cad85084329d61a96e466d11854", "parents": [ "e9df2e8fd8fbc95c57dbd1d33dada66c4627b44c" ], "author": { "name": "YOSHIFUJI Hideaki", "email": "yoshfuji@linux-ipv6.org", "time": "Sun Apr 13 23:42:18 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 23:42:18 2008 -0700" }, "message": "[IPV6] ADDRCONF: Ensure disabling multicast RS even if privacy extensions are disabled.\n\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b45e9189c058bfa495073951ff461ee0eea968be", "tree": "a620ecd6443b456bdc4b2f0d144d11ee953b166d", "parents": [ "2ed9926e16094ad143b96b09c64cba8bcba05ee1" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 23:14:15 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 23:14:15 2008 -0700" }, "message": "[IPV6]: Fix ipv6 address fetching in raw6_icmp_error().\n\nFixes kernel bugzilla 10437\n\nBased almost entirely upon a patch by Dmitry Butskoy.\n\nWhen deciding what raw sockets to deliver the ICMPv6\nto, we should use the addresses in the ICMPv6 quoted\nIPV6 header, not the top-level one.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2ed9926e16094ad143b96b09c64cba8bcba05ee1", "tree": "80694dbe8455fe47053709bd3c14a7edab1a8c24", "parents": [ "d2dcba612b357a4687843ad4f46629d219233fd7" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Sun Apr 13 22:45:40 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 22:45:40 2008 -0700" }, "message": "[NET]: Return more appropriate error from eth_validate_addr().\n\nPaul Bolle wrote:\n\u003e http://bugzilla.kernel.org/show_bug.cgi?id\u003d9923 would have been much easier to\n\u003e track down if eth_validate_addr() would somehow complain aloud if an address \n\u003e is invalid. Shouldn\u0027t it make at least some noise?\n\nI guess it should return -EADDRNOTAVAIL similar to eth_mac_addr()\nwhen validation fails.\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4c821d753d5c097babd6609bcd85f08e254a3505", "tree": "597d86c6b81340e66adae5e127b27a428473cdc2", "parents": [ "f37f0afb2916ccf287428983026261db78c7661a" ], "author": { "name": "Ben Hutchings", "email": "bhutchings@solarflare.com", "time": "Sun Apr 13 21:52:48 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 21:52:48 2008 -0700" }, "message": "[NET]: Fix kernel-doc for skb_segment\n\nThe kernel-doc comment for skb_segment is clearly wrong. This states\nwhat it actually does.\n\nSigned-off-by: Ben Hutchings \u003cbhutchings@solarflare.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f37f0afb2916ccf287428983026261db78c7661a", "tree": "0c86b039a03b166dc168b8ef74f6026ffecc81c9", "parents": [ "e01bf1c83332c4653ffd30eed20a94a9c83d82b2" ], "author": { "name": "Eric Dumazet", "email": "dada1@cosmosbay.com", "time": "Sun Apr 13 21:39:26 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 13 21:39:26 2008 -0700" }, "message": "[SOCK] sk_stamp: should be initialized to ktime_set(-1L, 0)\n\nProblem spotted by Andrew Brampton\n\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a40a7d15ba602b547f56b7b19e0282fe4fc3dee3", "tree": "81eede5e751b3e5600547018285c739825a08d48", "parents": [ "ab38fb04c9f8928cfaf6f4966633d783419906a1" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Sat Apr 12 18:40:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:40:38 2008 -0700" }, "message": "[SCTP]: IPv4 vs IPv6 addresses mess in sctp_inet[6]addr_event.\n\nAll IP addresses that are present in a system are duplicated on\nstruct sctp_sockaddr_entry. They are linked in the global list\ncalled sctp_local_addr_list. And this struct unions IPv4 and IPv6\naddresses.\n\nSo, there can be rare case, when a sockaddr_in.sin_addr coincides\nwith the corresponding part of the sockaddr_in6 and the notifier\nfor IPv4 will carry away an IPv6 entry.\n\nThe fix is to check the family before comparing the addresses.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Vlad Yasevich \u003cvladislav.yasevich@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ab38fb04c9f8928cfaf6f4966633d783419906a1", "tree": "e7025f9cc3979a509081eb7ad93f5f6a39610c71", "parents": [ "f4ad85ca3ef8a1ede76c5020a28a8f4057b4d24f" ], "author": { "name": "Vlad Yasevich", "email": "vladislav.yasevich@hp.com", "time": "Sat Apr 12 18:40:06 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:40:06 2008 -0700" }, "message": "[SCTP]: Fix compiler warning about const qualifiers\n\nFix 3 warnings about discarding const qualifiers:\n\nnet/sctp/ulpevent.c:862: warning: passing argument 1 of \u0027sctp_event2skb\u0027 discards qualifiers from pointer target type\nnet/sctp/sm_statefuns.c:4393: warning: passing argument 1 of \u0027SCTP_ASOC\u0027 discards qualifiers from pointer target type\nnet/sctp/socket.c:5874: warning: passing argument 1 of \u0027cmsg_nxthdr\u0027 discards qualifiers from pointer target type\n\nSigned-off-by: Vlad Yasevich \u003cvladislav.yasevich@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f4ad85ca3ef8a1ede76c5020a28a8f4057b4d24f", "tree": "cfab9dcd7b29eb150bd6168cf40c375bd9006bbe", "parents": [ "72da7b3860cabf427590b4982bc880bafab4d5c8" ], "author": { "name": "Gui Jianfeng", "email": "guijianfeng@cn.fujitsu.com", "time": "Sat Apr 12 18:39:34 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:39:34 2008 -0700" }, "message": "[SCTP]: Fix protocol violation when receiving an error lenght INIT-ACK\n\nWhen receiving an error length INIT-ACK during COOKIE-WAIT,\na 0-vtag ABORT will be responsed. This action violates the\nprotocol apparently. This patch achieves the following things.\n1 If the INIT-ACK contains all the fixed parameters, use init-tag\n recorded from INIT-ACK as vtag.\n2 If the INIT-ACK doesn\u0027t contain all the fixed parameters,\n just reflect its vtag.\n\nSigned-off-by: Gui Jianfeng \u003cguijianfeng@cn.fujitsu.com\u003e\nSigned-off-by: Vlad Yasevich \u003cvladislav.yasevich@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "72da7b3860cabf427590b4982bc880bafab4d5c8", "tree": "7a5084f9eac6c5556d792d0503158860b046201f", "parents": [ "e56cfad132f2ae269082359d279c17230c987e74" ], "author": { "name": "Wei Yongjun", "email": "yjwei@cn.fujitsu.com", "time": "Sat Apr 12 18:39:19 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:39:19 2008 -0700" }, "message": "[SCTP]: Add check for hmac_algo parameter in sctp_verify_param()\n\nRFC 4890 has the following text:\n\n The HMAC algorithm based on SHA-1 MUST be supported and\n included in the HMAC-ALGO parameter.\n\nAs a result, we need to check in sctp_verify_param() that HMAC_SHA1 is\npresent in the list. If not, we should probably treat this as a\nprotocol violation.\n\nIt should also be a protocol violation if the HMAC parameter is empty.\n\nSigned-off-by: Wei Yongjun \u003cyjwei@cn.fujitsu.com\u003e\nSigned-off-by: Vlad Yasevich \u003cvladislav.yasevich@hp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e56cfad132f2ae269082359d279c17230c987e74", "tree": "6867ab5499fe617a16b15f1572b5d22d863d2524", "parents": [ "028b027524b162eef90839a92ba4b8bddf23e06c" ], "author": { "name": "Jarek Poplawski", "email": "jarkao2@gmail.com", "time": "Sat Apr 12 18:37:13 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:37:13 2008 -0700" }, "message": "[NET_SCHED] cls_u32: refcounting fix for u32_delete()\n\nDeleting of nonroot hnodes mostly doesn\u0027t work in u32_delete():\nrefcnt \u003d\u003d 1 is expected, but such hnodes\u0027 refcnts are initialized\nwith 0 and charged only with \"link\" nodes. Now they\u0027ll start with\n1 like usual. Thanks to Patrick McHardy for an improving suggestion.\n\nSigned-off-by: Jarek Poplawski \u003cjarkao2@gmail.com\u003e\nAcked-by: Jamal Hadi Salim \u003chadi@cyberus.ca\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "028b027524b162eef90839a92ba4b8bddf23e06c", "tree": "5c128717f73ff6fb93c361cfae6961267ed52082", "parents": [ "ae1b6a31b1f9ef2c7ba5ef89799f210a9ba6937c" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Sat Apr 12 18:35:41 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:35:41 2008 -0700" }, "message": "[DCCP]: Fix skb-\u003ecb conflicts with IP\n\ndev_queue_xmit() and the other IP output functions expect to get a skb\nwith clear or properly initialized skb-\u003ecb. Unlike TCP and UDP, the\ndccp_skb_cb doesn\u0027t contain a struct inet_skb_parm at the beginning,\nso the DCCP-specific data is interpreted by the IP output functions.\nThis can cause false negatives for the conditional POST_ROUTING hook\ninvocation, making the packet bypass the hook.\n\nAdd a inet_skb_parm/inet6_skb_parm union to the beginning of\ndccp_skb_cb to avoid clashes. Also add a BUILD_BUG_ON to make\nsure it fits in the cb.\n\n[ Combined with patch from Gerrit Renker to remove two now unnecessary\n memsets of IPCB(skb)-\u003eopt ]\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nAcked-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ae1b6a31b1f9ef2c7ba5ef89799f210a9ba6937c", "tree": "76f26c291c480345de8d70a6c37143c43ac5f749", "parents": [ "2de58e30db3f7e83592e22fb459dd6695b648889" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Sat Apr 12 18:33:50 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Apr 12 18:33:50 2008 -0700" }, "message": "[AX25]: Potential ax25_uid_assoc-s leaks on module unload.\n\nThe ax25_uid_free call walks the ax25_uid_list and releases entries\nfrom it. The problem is that after the fisrt call to hlist_del_init\nthe hlist_for_each_entry (which hides behind the ax25_uid_for_each)\nwill consider the current position to be the last and will return.\n\nThus, the whole list will be left not freed.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7951f0b03a63d657c72c7d54d306ef3357e7e604", "tree": "4c0faf167b81118be69b7c924024da58e5174518", "parents": [ "951e07c930f5f66b676eaa4c32a1b0d8e2d7d06a" ], "author": { "name": "Daniel Lezcano", "email": "dlezcano@fr.ibm.com", "time": "Thu Apr 10 20:53:10 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 10 20:53:10 2008 -0700" }, "message": "[NETNS][IPV6] tcp - assign the netns for timewait sockets\n\nCopy the network namespace from the socket to the timewait socket.\n\nSigned-off-by: Daniel Lezcano \u003cdlezcano@fr.ibm.com\u003e\nAcked-by: Mark Lord \u003cmlord@pobox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "951e07c930f5f66b676eaa4c32a1b0d8e2d7d06a", "tree": "dbbecc7be4c616d47586ef0ff6a552c110bcb8d5", "parents": [ "619c714c1d6e4dff00ddde582d78492fd95452d6" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 10 01:29:36 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 10 01:29:36 2008 -0700" }, "message": "[IPV4]: Fix byte value boundary check in do_ip_getsockopt().\n\nThis fixes kernel bugzilla 10371.\n\nAs reported by M.Piechaczek@osmosys.tv, if we try to grab a\nchar sized socket option value, as in:\n\n unsigned char ttl \u003d 255;\n socklen_t len \u003d sizeof(ttl);\n setsockopt(socket, IPPROTO_IP, IP_MULTICAST_TTL, \u0026ttl, \u0026len);\n\n getsockopt(socket, IPPROTO_IP, IP_MULTICAST_TTL, \u0026ttl, \u0026len);\n\nThe ttl returned will be wrong on big-endian, and on both little-\nendian and big-endian the next three bytes in userspace are written\nwith garbage.\n\nIt\u0027s because of this test in do_ip_getsockopt():\n\n\tif (len \u003c sizeof(int) \u0026\u0026 len \u003e 0 \u0026\u0026 val\u003e\u003d0 \u0026\u0026 val\u003c255) {\n\nIt should allow a \u0027val\u0027 of 255 to pass here, but it doesn\u0027t so it\ncopies a full \u0027int\u0027 back to userspace.\n\nOn little-endian that will write the correct value into the location\nbut it spams on the next three bytes in userspace. On big endian it\nwrites the wrong value into the location and spams the next three\nbytes.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "475959d4773e53a2700e523dd30acebbd47556a5", "tree": "67f1af04a4dd1f47f7002262b0230e5da3c97745", "parents": [ "1b9b70ea2ebaab26c3e4fed385dfab6fc16359ed" ], "author": { "name": "Jan Engelhardt", "email": "jengelh@computergmbh.de", "time": "Wed Apr 09 15:14:58 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 09 15:14:58 2008 -0700" }, "message": "[NETFILTER]: nf_nat: autoload IPv4 connection tracking\n\nWithout this patch, the generic L3 tracker would kick in\nif nf_conntrack_ipv4 was not loaded before nf_nat, which\nwould lead to translation problems with ICMP errors.\n\nNAT does not make sense without IPv4 connection tracking\nanyway, so just add a call to need_ipv4_conntrack().\n\nSigned-off-by: Jan Engelhardt \u003cjengelh@computergmbh.de\u003e\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1b9b70ea2ebaab26c3e4fed385dfab6fc16359ed", "tree": "dad450564144193ea3591a491cc24665f4b22224", "parents": [ "b41f5bfff73f244101b34f3603974ef7aeadf545" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Wed Apr 09 15:14:18 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 09 15:14:18 2008 -0700" }, "message": "[NETFILTER]: xt_hashlimit: fix mask calculation\n\nShifts larger than the data type are undefined, don\u0027t try to shift\nan u32 by 32. Also remove some special-casing of bitmasks divisible\nby 32.\n\nBased on patch by Jan Engelhardt \u003cjengelh@computergmbh.de\u003e.\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b41f5bfff73f244101b34f3603974ef7aeadf545", "tree": "5c7fc7de15a020cde0f99fac3e61d981a1d59022", "parents": [ "bcf0dda8d2408fe1c1040cdec5a98e5fcad2ac72", "e91e9d490d9ae382003ef9d05fd50238db54c35b" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 09 15:10:14 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 09 15:10:14 2008 -0700" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6\n" }, { "commit": "bcf0dda8d2408fe1c1040cdec5a98e5fcad2ac72", "tree": "33117ad780d212f8a83c007ccc1b95bc0debbdbf", "parents": [ "21f644f3eabde637f255f75ad05d0821a7a36b7f" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Wed Apr 09 15:08:24 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 09 15:08:24 2008 -0700" }, "message": "[XFRM]: xfrm_user: fix selector family initialization\n\nCommit df9dcb45 ([IPSEC]: Fix inter address family IPsec tunnel handling)\nbroke openswan by removing the selector initialization for tunnel mode\nin case it is uninitialized.\n\nThis patch restores the initialization, fixing openswan, but probably\nbreaking inter-family tunnels again (unknown since the patch author\ndisappeared). The correct thing for inter-family tunnels is probably\nto simply initialize the selector family explicitly.\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "216bce90b811a35eb5cd2ed8216bdbb1753e9b2b", "tree": "1e0b8fd87ca2e36bdf668afe264f15555bf65b5a", "parents": [ "16f2e85d3151efa643879fa5aa87c9d77d60f57e" ], "author": { "name": "Vladimir Koutny", "email": "vlado@work.ksp.sk", "time": "Mon Mar 31 17:05:10 2008 +0200" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Wed Apr 09 10:33:49 2008 -0400" }, "message": "mac80211: use short_preamble mode from capability if ERP IE not present\n\nWhen associating to a b-only AP where there is no ERP IE, short preamble\nmode is left at previous state (probably also protection mode). In this\ncase, disable protection and use short preamble mode as specified in\ncapability field. The same is done if capability field is changed on-the-fly.\n\nSigned-off-by: Vladimir Koutny \u003cvlado@ksp.sk\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "21f644f3eabde637f255f75ad05d0821a7a36b7f", "tree": "2dc70aade59bf77c2d156b8e2544906128368f87", "parents": [ "6adb4f733e9996b4fd68a6db50dd51bd2463ccac" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 08 16:50:44 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 08 16:50:44 2008 -0700" }, "message": "[NET]: Undo code bloat in hot paths due to print_mac().\n\nIf print_mac() is used inside of a pr_debug() the compiler\ncan\u0027t see that the call is redundant so still performs it\neven of pr_debug() ends up being a nop.\n\nSo don\u0027t use print_mac() in such cases in hot code paths,\nuse MAC_FMT et al. instead.\n\nAs noted by Joe Perches, pr_debug() could be modified to\nhandle this better, but that is a change to an interface\nused by the entire kernel and thus needs to be validated\ncarefully. This here is thus the less risky fix for\n2.6.25\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6adb4f733e9996b4fd68a6db50dd51bd2463ccac", "tree": "2ed50e47632902fe93f601d79fbb0ee8cb43c11c", "parents": [ "882bebaaca4bb1484078d44ef011f918c0e1e14e" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Mon Apr 07 22:33:57 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 07 22:33:57 2008 -0700" }, "message": "[TCP]: Don\u0027t allow FRTO to take place while MTU is being probed\n\nMTU probe can cause some remedies for FRTO because the normal\npacket ordering may be violated allowing FRTO to make a wrong\ndecision (it might not be that serious threat for anything\nthough). Thus it\u0027s safer to not run FRTO while MTU probe is\nunderway.\n\nIt seems that the basic FRTO variant should also look for an\nskb at probe_seq.start to check if that\u0027s retransmitted one\nbut I didn\u0027t implement it now (plain seqno in window check\nisn\u0027t robust against wraparounds).\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "882bebaaca4bb1484078d44ef011f918c0e1e14e", "tree": "d6542dd9fc3d941ecc5f418b66ea09e91f47b71f", "parents": [ "c137f3dda04b0aee1bc6889cdc69185f53df8a82" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Mon Apr 07 22:33:07 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 07 22:33:07 2008 -0700" }, "message": "[TCP]: tcp_simple_retransmit can cause S+L\n\nThis fixes Bugzilla #10384\n\ntcp_simple_retransmit does L increment without any checking\nwhatsoever for overflowing S+L when Reno is in use.\n\nThe simplest scenario I can currently think of is rather\ncomplex in practice (there might be some more straightforward\ncases though). Ie., if mss is reduced during mtu probing, it\nmay end up marking everything lost and if some duplicate ACKs\narrived prior to that sacked_out will be non-zero as well,\nleading to S+L \u003e packets_out, tcp_clean_rtx_queue on the next\ncumulative ACK or tcp_fastretrans_alert on the next duplicate\nACK will fix the S counter.\n\nMore straightforward (but questionable) solution would be to\njust call tcp_reset_reno_sack() in tcp_simple_retransmit but\nit would negatively impact the probe\u0027s retransmission, ie.,\nthe retransmissions would not occur if some duplicate ACKs\nhad arrived.\n\nSo I had to add reno sacked_out reseting to CA_Loss state\nwhen the first cumulative ACK arrives (this stale sacked_out\nmight actually be the explanation for the reports of left_out\noverflows in kernel prior to 2.6.23 and S+L overflow reports\nof 2.6.24). However, this alone won\u0027t be enough to fix kernel\nbefore 2.6.24 because it is building on top of the commit\n1b6d427bb7e ([TCP]: Reduce sacked_out with reno when purging\nwrite_queue) to keep the sacked_out from overflowing.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nReported-by: Alessandro Suardi \u003calessandro.suardi@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c137f3dda04b0aee1bc6889cdc69185f53df8a82", "tree": "75031e829df932aae697b92ee1ef490bdf71c762", "parents": [ "1b69d745397eac12b3f8a2eb6b799cd476aef282" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Mon Apr 07 22:32:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 07 22:32:38 2008 -0700" }, "message": "[TCP]: Fix NewReno\u0027s fast rexmit/recovery problems with GSOed skb\n\nFixes a long-standing bug which makes NewReno recovery crippled.\nWith GSO the whole head skb was marked as LOST which is in\nviolation of NewReno procedure that only wants to mark one packet\nand ended up breaking our TCP code by causing counter overflow\nbecause our code was built on top of assumption about valid\nNewReno procedure. This manifested as triggering a WARN_ON for\nthe overflow in a number of places.\n\nIt seems relatively safe alternative to just do nothing if\ntcp_fragment fails due to oom because another duplicate ACK is\nlikely to be received soon and the fragmentation will be retried.\n\nSpecial thanks goes to Soeren Sonnenburg \u003ckernel@nn7.de\u003e who was\nlucky enough to be able to reproduce this so that the warning\nfor the overflow was hit. It\u0027s not as easy task as it seems even\nif this bug happens quite often because the amount of outstanding\ndata is pretty significant for the mismarkings to lead to an\noverflow.\n\nBecause it\u0027s very late in 2.6.25-rc cycle (if this even makes in\ntime), I didn\u0027t want to touch anything with SACK enabled here.\nFragmenting might be useful for it as well but it\u0027s more or less\na policy decision rather than mandatory fix. Thus there\u0027s no need\nto rush and we can postpone considering tcp_fragment with SACK\nfor 2.6.26.\n\nIn 2.6.24 and earlier, this very same bug existed but the effect\nis slightly different because of a small changes in the if\nconditions that fit to the patch\u0027s context. With them nothing\ngot lost marker and thus no retransmissions happened.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1b69d745397eac12b3f8a2eb6b799cd476aef282", "tree": "5cf3ab06c6c6ebf1adc55d4a3aae17f6fd110521", "parents": [ "bfe87dbc7b4da5b05a1a78480e996787a500cc6f" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Mon Apr 07 22:31:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 07 22:31:38 2008 -0700" }, "message": "[TCP]: Restore 2.6.24 mark_head_lost behavior for newreno/fack\n\nThe fast retransmission can be forced locally to the rfc3517\nbranch in tcp_update_scoreboard instead of making such fragile\nconstructs deeper in tcp_mark_head_lost.\n\nThis is necessary for the next patch which must not have\nloopholes for cnt \u003e packets check. As one can notice,\nreadability got some improvements too because of this :-).\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "16f2e85d3151efa643879fa5aa87c9d77d60f57e", "tree": "10490ca8550850a80c5f9c27cb9bd0b9f4b1a228", "parents": [ "881400a20c3551e90eed1062cf0387fa686a2fd0" ], "author": { "name": "Johannes Berg", "email": "johannes@sipsolutions.net", "time": "Mon Apr 07 14:35:46 2008 +0200" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Mon Apr 07 22:19:30 2008 -0400" }, "message": "nl80211: fix STA AID bug\n\nThis fixes the STA AID setting and actually makes hostapd/mac80211\nwork properly in presence of power-saving stations.\n\nSigned-off-by: Johannes Berg \u003cjohannes@sipsolutions.net\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "23556323b22fef35bdc36465b7e7439ba3748c9f", "tree": "a82c7bfe8716a04b183bc9ac0689040be74229bc", "parents": [ "b2a5c19ca0315723cecb9489ff8b67c4f17367b4" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Fri Apr 04 12:45:12 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Apr 04 12:45:12 2008 -0700" }, "message": "[VLAN]: Fix egress priority mappings leak.\n\nThese entries are allocated in vlan_dev_set_egress_priority, \nbut are never released and leaks on vlan device removal.\n\nDrop these in vlan\u0027s -\u003euninit callback - after the device is \nbrought down and everyone is notified about it is going to\nbe unregistered.\n\nFound during testing vlan netnsization patchset.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nAcked-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "84f59370c519449c70dcc813b050f5cbbf0098e7", "tree": "247917a098519394b08c52334d6eacc949f10848", "parents": [ "eb867579311a9c1e998d6911af056772c400122a" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Thu Apr 03 13:33:00 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 03 13:33:00 2008 -0700" }, "message": "[IPV6]: Fix refcounting for anycast dst entries.\n\nAnycast DST entries allocated inside ipv6_dev_ac_inc are leaked when\nnetwork device is stopped without removing IPv6 addresses from it. The\nbug has been observed in the reality on 2.6.18-rhel5 kernel.\n\nIn the above case addrconf_ifdown marks all entries as obsolete and\nip6_del_rt called from __ipv6_dev_ac_dec returns ENOENT. The\nreferrence is not dropped.\n\nThe fix is simple. DST entry should not keep referrence when stored in\nthe FIB6 tree.\n\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eb867579311a9c1e998d6911af056772c400122a", "tree": "8f0ba1a08b34472dc8bbf1355de54e92414c5ebd", "parents": [ "439e23857a21c3a953826eed23c818697a97de1a" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Thu Apr 03 13:31:53 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 03 13:31:53 2008 -0700" }, "message": "[IPV6]: inet6_dev on loopback should be kept until namespace stop.\n\nIn the other case it will be destroyed when last address will be removed\nfrom lo inside a namespace. This will break IPv6 in several places. The\nmost obvious one is ip6_dst_ifdown.\n\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "439e23857a21c3a953826eed23c818697a97de1a", "tree": "72d751492b9bafb55da3c6a35cea9e17573f938b", "parents": [ "af2681828af5f2b42e12e8b16ba0cf113cf486c8" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Thu Apr 03 13:30:17 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 03 13:30:17 2008 -0700" }, "message": "[IPV6]: Event type in addrconf_ifdown is mis-used.\n\naddrconf_ifdown is broken in respect to the usage of how\nparameter. This function is called with (event !\u003d NETDEV_DOWN) and (2)\non the IPv6 stop. It the latter case inet6_dev from loopback device\nshould be destroyed.\n\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "af2681828af5f2b42e12e8b16ba0cf113cf486c8", "tree": "ed25908930ea98782ec7a13fd985c9825317a8b2", "parents": [ "9597362d354f8655ece324b01d0c640a0e99c077" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Thu Apr 03 12:52:19 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Apr 03 12:52:19 2008 -0700" }, "message": "[ICMP]: Ensure that ICMP relookup maintains status quo\n\nThe ICMP relookup path is only meant to modify behaviour when\nappropriate IPsec policies are in place and marked as requiring\nrelookups. It is certainly not meant to modify behaviour when\nIPsec policies don\u0027t exist at all.\n\nHowever, due to an oversight on the error paths existing behaviour\nmay in fact change should one of the relookup steps fail.\n\nThis patch corrects this by redirecting all errors on relookup\nfailures to the previous code path. That is, if the initial\nxfrm_lookup let the packet pass, we will stand by that decision\nshould the relookup fail due to an error.\n\nThis should be safe from a security point-of-view because compliant\nsystems must install a default deny policy so the packet would\u0027nt\nhave passed in that case.\n\nMany thanks to Julian Anastasov for pointing out this error.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2f819ae8816990aaad46dd6d1748a096d136df68", "tree": "65891bf45d2d0e4e7884c4cc69beb9c27b67be95", "parents": [ "d40c2f29f3b0f5937bace575722598dac474c478", "802fb176d8c635ae42da31b80841c26e8c7338a0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 07:46:18 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 07:46:18 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (45 commits)\n [VLAN]: Proc entry is not renamed when vlan device name changes.\n [IPV6]: Fix ICMP relookup error path dst leak\n [ATM] drivers/atm/iphase.c: compilation warning fix\n IPv6: do not create temporary adresses with too short preferred lifetime\n IPv6: only update the lifetime of the relevant temporary address\n bluetooth : __rfcomm_dlc_close lock fix\n bluetooth : use lockdep sub-classes for diffrent bluetooth protocol\n [ROSE/AX25] af_rose: rose_release() fix\n mac80211: correct use_short_preamble handling\n b43: Fix PCMCIA IRQ routing\n b43: Add DMA mapping failure messages\n mac80211: trigger ieee80211_sta_work after opening interface\n [LLC]: skb allocation size for responses\n [IP] UDP: Use SEQ_START_TOKEN.\n [NET]: Remove Documentation/networking/sk98lin.txt\n [ATM] atm/idt77252.c: Make 2 functions static\n [ATM]: Make atm/he.c:read_prom_byte() static\n [IPV6] MCAST: Ensure to check multicast listener(s).\n [LLC]: Kill llc_station_mac_sa symbol export.\n forcedeth: fix locking bug with netconsole\n ...\n" }, { "commit": "802fb176d8c635ae42da31b80841c26e8c7338a0", "tree": "3f0a1283ca650868fd41942fb13e5ada129788c7", "parents": [ "f32c5f2c3866bf4d932d2bc42216dafb90a50ab7" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Apr 02 00:08:01 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 02 00:08:01 2008 -0700" }, "message": "[VLAN]: Proc entry is not renamed when vlan device name changes.\n\nThis may lead to situations, when each of two proc entries produce\ndata for the other\u0027s device.\n\nLooks like a BUG, so this patch is for net-2.6. It will not apply to\nnet-2.6.26 since dev-\u003end_net access is replaced with dev_net(dev)\none.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f32c5f2c3866bf4d932d2bc42216dafb90a50ab7", "tree": "fd2d7da5651b76027977dc35af353a45ca348a45", "parents": [ "a22eb6faae4ec64d59cc206671eb0bfe46c5ffd2" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Wed Apr 02 00:06:09 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 02 00:06:09 2008 -0700" }, "message": "[IPV6]: Fix ICMP relookup error path dst leak\n\nWhen we encounter an error while looking up the dst the second\ntime we need to drop the first dst. This patch is pretty much\nthe same as the one for IPv4.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eac55bf97094f6b64116426864cf4666ef7587bc", "tree": "467776de93a7fdbb75904dbd702f7c417e18aa69", "parents": [ "c6fbfac2e61c9a8617f64b93e8c990b8d864bce5" ], "author": { "name": "Benoit Boissinot", "email": "benoit.boissinot@ens-lyon.org", "time": "Wed Apr 02 00:01:35 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 02 00:01:35 2008 -0700" }, "message": "IPv6: do not create temporary adresses with too short preferred lifetime\n\nFrom RFC341:\nA temporary address is created only if this calculated Preferred\nLifetime is greater than REGEN_ADVANCE time units. In particular, an\nimplementation must not create a temporary address with a zero\nPreferred Lifetime.\n\nSigned-off-by: Benoit Boissinot \u003cbenoit.boissinot@ens-lyon.org\u003e\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c6fbfac2e61c9a8617f64b93e8c990b8d864bce5", "tree": "54fd0c1543069a617e3b315346e3593e0b3be52c", "parents": [ "1905f6c736cb618e07eca0c96e60e3c024023428" ], "author": { "name": "Benoit Boissinot", "email": "benoit.boissinot@ens-lyon.org", "time": "Wed Apr 02 00:00:58 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Apr 02 00:00:58 2008 -0700" }, "message": "IPv6: only update the lifetime of the relevant temporary address\n\nWhen receiving a prefix information from a routeur, only update the\nlifetimes of the temporary address associated with that prefix.\n\nOtherwise if one deprecated prefix is advertized, all your temporary\naddresses will become deprecated.\n\nSigned-off-by: Benoit Boissinot \u003cbenoit.boissinot@ens-lyon.org\u003e\nAcked-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1905f6c736cb618e07eca0c96e60e3c024023428", "tree": "321a987569bf380fe1960879d03dd5e98fab92cc", "parents": [ "68845cb2c82275efd7390026bba70c320ca6ef86" ], "author": { "name": "Dave Young", "email": "hidave.darkstar@gmail.com", "time": "Tue Apr 01 23:59:06 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 01 23:59:06 2008 -0700" }, "message": "bluetooth : __rfcomm_dlc_close lock fix\n\nLockdep warning will be trigged while rfcomm connection closing.\n\nThe locks taken in rfcomm_dev_add:\nrfcomm_dev_lock --\u003e d-\u003elock\n\nIn __rfcomm_dlc_close:\nd-\u003elock --\u003e rfcomm_dev_lock (in rfcomm_dev_state_change)\n\nThere\u0027s two way to fix it, one is in rfcomm_dev_add we first locking\nd-\u003elock then the rfcomm_dev_lock\n\nThe other (in this patch), remove the locking of d-\u003elock for\nrfcomm_dev_state_change because just locking \"d-\u003estate \u003d BT_CLOSED;\"\nis enough.\n\n[ 295.002046] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ 295.002046] [ INFO: possible circular locking dependency detected ]\n[ 295.002046] 2.6.25-rc7 #1\n[ 295.002046] -------------------------------------------------------\n[ 295.002046] krfcommd/2705 is trying to acquire lock:\n[ 295.002046] (rfcomm_dev_lock){-.--}, at: [\u003cf89a090a\u003e] rfcomm_dev_state_change+0x6a/0xd0 [rfcomm]\n[ 295.002046] \n[ 295.002046] but task is already holding lock:\n[ 295.002046] (\u0026d-\u003elock){--..}, at: [\u003cf899c533\u003e] __rfcomm_dlc_close+0x43/0xd0 [rfcomm]\n[ 295.002046] \n[ 295.002046] which lock already depends on the new lock.\n[ 295.002046] \n[ 295.002046] \n[ 295.002046] the existing dependency chain (in reverse order) is:\n[ 295.002046] \n[ 295.002046] -\u003e #1 (\u0026d-\u003elock){--..}:\n[ 295.002046] [\u003cc0149b23\u003e] check_prev_add+0xd3/0x200\n[ 295.002046] [\u003cc0149ce5\u003e] check_prevs_add+0x95/0xe0\n[ 295.002046] [\u003cc0149f6f\u003e] validate_chain+0x23f/0x320\n[ 295.002046] [\u003cc014b7b1\u003e] __lock_acquire+0x1c1/0x760\n[ 295.002046] [\u003cc014c349\u003e] lock_acquire+0x79/0xb0\n[ 295.002046] [\u003cc03d6b99\u003e] _spin_lock+0x39/0x80\n[ 295.002046] [\u003cf89a01c0\u003e] rfcomm_dev_add+0x240/0x360 [rfcomm]\n[ 295.002046] [\u003cf89a047e\u003e] rfcomm_create_dev+0x6e/0xe0 [rfcomm]\n[ 295.002046] [\u003cf89a0823\u003e] rfcomm_dev_ioctl+0x33/0x60 [rfcomm]\n[ 295.002046] [\u003cf899facc\u003e] rfcomm_sock_ioctl+0x2c/0x50 [rfcomm]\n[ 295.002046] [\u003cc0363d38\u003e] sock_ioctl+0x118/0x240\n[ 295.002046] [\u003cc0194196\u003e] vfs_ioctl+0x76/0x90\n[ 295.002046] [\u003cc0194446\u003e] do_vfs_ioctl+0x56/0x140\n[ 295.002046] [\u003cc0194569\u003e] sys_ioctl+0x39/0x60\n[ 295.002046] [\u003cc0104faa\u003e] syscall_call+0x7/0xb\n[ 295.002046] [\u003cffffffff\u003e] 0xffffffff\n[ 295.002046] \n[ 295.002046] -\u003e #0 (rfcomm_dev_lock){-.--}:\n[ 295.002046] [\u003cc0149a84\u003e] check_prev_add+0x34/0x200\n[ 295.002046] [\u003cc0149ce5\u003e] check_prevs_add+0x95/0xe0\n[ 295.002046] [\u003cc0149f6f\u003e] validate_chain+0x23f/0x320\n[ 295.002046] [\u003cc014b7b1\u003e] __lock_acquire+0x1c1/0x760\n[ 295.002046] [\u003cc014c349\u003e] lock_acquire+0x79/0xb0\n[ 295.002046] [\u003cc03d6639\u003e] _read_lock+0x39/0x80\n[ 295.002046] [\u003cf89a090a\u003e] rfcomm_dev_state_change+0x6a/0xd0 [rfcomm]\n[ 295.002046] [\u003cf899c548\u003e] __rfcomm_dlc_close+0x58/0xd0 [rfcomm]\n[ 295.002046] [\u003cf899d44f\u003e] rfcomm_recv_ua+0x6f/0x120 [rfcomm]\n[ 295.002046] [\u003cf899e061\u003e] rfcomm_recv_frame+0x171/0x1e0 [rfcomm]\n[ 295.002046] [\u003cf899e357\u003e] rfcomm_run+0xe7/0x550 [rfcomm]\n[ 295.002046] [\u003cc013c18c\u003e] kthread+0x5c/0xa0\n[ 295.002046] [\u003cc0105c07\u003e] kernel_thread_helper+0x7/0x10\n[ 295.002046] [\u003cffffffff\u003e] 0xffffffff\n[ 295.002046] \n[ 295.002046] other info that might help us debug this:\n[ 295.002046] \n[ 295.002046] 2 locks held by krfcommd/2705:\n[ 295.002046] #0: (rfcomm_mutex){--..}, at: [\u003cf899e2eb\u003e] rfcomm_run+0x7b/0x550 [rfcomm]\n[ 295.002046] #1: (\u0026d-\u003elock){--..}, at: [\u003cf899c533\u003e] __rfcomm_dlc_close+0x43/0xd0 [rfcomm]\n[ 295.002046] \n[ 295.002046] stack backtrace:\n[ 295.002046] Pid: 2705, comm: krfcommd Not tainted 2.6.25-rc7 #1\n[ 295.002046] [\u003cc0128a38\u003e] ? printk+0x18/0x20\n[ 295.002046] [\u003cc014927f\u003e] print_circular_bug_tail+0x6f/0x80\n[ 295.002046] [\u003cc0149a84\u003e] check_prev_add+0x34/0x200\n[ 295.002046] [\u003cc0149ce5\u003e] check_prevs_add+0x95/0xe0\n[ 295.002046] [\u003cc0149f6f\u003e] validate_chain+0x23f/0x320\n[ 295.002046] [\u003cc014b7b1\u003e] __lock_acquire+0x1c1/0x760\n[ 295.002046] [\u003cc014c349\u003e] lock_acquire+0x79/0xb0\n[ 295.002046] [\u003cf89a090a\u003e] ? rfcomm_dev_state_change+0x6a/0xd0 [rfcomm]\n[ 295.002046] [\u003cc03d6639\u003e] _read_lock+0x39/0x80\n[ 295.002046] [\u003cf89a090a\u003e] ? rfcomm_dev_state_change+0x6a/0xd0 [rfcomm]\n[ 295.002046] [\u003cf89a090a\u003e] rfcomm_dev_state_change+0x6a/0xd0 [rfcomm]\n[ 295.002046] [\u003cf899c548\u003e] __rfcomm_dlc_close+0x58/0xd0 [rfcomm]\n[ 295.002046] [\u003cf899d44f\u003e] rfcomm_recv_ua+0x6f/0x120 [rfcomm]\n[ 295.002046] [\u003cf899e061\u003e] rfcomm_recv_frame+0x171/0x1e0 [rfcomm]\n[ 295.002046] [\u003cc014abd9\u003e] ? trace_hardirqs_on+0xb9/0x130\n[ 295.002046] [\u003cc03d6e89\u003e] ? _spin_unlock_irqrestore+0x39/0x70\n[ 295.002046] [\u003cf899e357\u003e] rfcomm_run+0xe7/0x550 [rfcomm]\n[ 295.002046] [\u003cc03d4559\u003e] ? __sched_text_start+0x229/0x4c0\n[ 295.002046] [\u003cc0120000\u003e] ? cpu_avg_load_per_task+0x20/0x30\n[ 295.002046] [\u003cf899e270\u003e] ? rfcomm_run+0x0/0x550 [rfcomm]\n[ 295.002046] [\u003cc013c18c\u003e] kthread+0x5c/0xa0\n[ 295.002046] [\u003cc013c130\u003e] ? kthread+0x0/0xa0\n[ 295.002046] [\u003cc0105c07\u003e] kernel_thread_helper+0x7/0x10\n[ 295.002046] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: Dave Young \u003chidave.darkstar@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "68845cb2c82275efd7390026bba70c320ca6ef86", "tree": "d28413ecb9e8695bc3292386d3cfacb76928a668", "parents": [ "4965291acf8cc2c31dcb2fc7d292a04ee08da2dd" ], "author": { "name": "Dave Young", "email": "hidave.darkstar@gmail.com", "time": "Tue Apr 01 23:58:35 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 01 23:58:35 2008 -0700" }, "message": "bluetooth : use lockdep sub-classes for diffrent bluetooth protocol\n\n\u0027rfcomm connect\u0027 will trigger lockdep warnings which is caused by\nlocking diffrent kinds of bluetooth sockets at the same time.\n\nSo using sub-classes per AF_BLUETOOTH sub-type for lockdep.\n\nThanks for the hints from dave jones.\n\n---\n\u003e From: Dave Jones \u003cdavej@codemonkey.org.uk\u003e\n\u003e Date: Thu, 27 Mar 2008 12:21:56 -0400\n\u003e\n\u003e \u003e Mar 27 08:10:57 localhost kernel: Pid: 3611, comm: obex-data-serve Not tainted 2.6.25-0.121.rc5.git4.fc9 #1\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [__lock_acquire+2287/3089] __lock_acquire+0x8ef/0xc11\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [sched_clock+8/11] ? sched_clock+0x8/0xb\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [lock_acquire+106/144] lock_acquire+0x6a/0x90\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [\u003cf8bd9321\u003e] ? l2cap_sock_bind+0x29/0x108 [l2cap]\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [lock_sock_nested+182/198] lock_sock_nested+0xb6/0xc6\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [\u003cf8bd9321\u003e] ? l2cap_sock_bind+0x29/0x108 [l2cap]\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [security_socket_post_create+22/27] ? security_socket_post_create+0x16/0x1b\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [__sock_create+388/472] ? __sock_create+0x184/0x1d8\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [\u003cf8bd9321\u003e] l2cap_sock_bind+0x29/0x108 [l2cap]\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [kernel_bind+10/13] kernel_bind+0xa/0xd\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [\u003cf8dad3d7\u003e] rfcomm_dlc_open+0xc8/0x294 [rfcomm]\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [lock_sock_nested+187/198] ? lock_sock_nested+0xbb/0xc6\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [\u003cf8dae18c\u003e] rfcomm_sock_connect+0x8b/0xc2 [rfcomm]\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [sys_connect+96/125] sys_connect+0x60/0x7d\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [__lock_acquire+1370/3089] ? __lock_acquire+0x55a/0xc11\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [sys_socketcall+140/392] sys_socketcall+0x8c/0x188\n\u003e \u003e Mar 27 08:10:57 localhost kernel: [syscall_call+7/11] syscall_call+0x7/0xb\n---\n\nSigned-off-by: Dave Young \u003chidave.darkstar@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4965291acf8cc2c31dcb2fc7d292a04ee08da2dd", "tree": "f38695b8fcdee173cf714222193c8f12274ba879", "parents": [ "d43c7b37ad787173d08683f05eadeea0398fefdf" ], "author": { "name": "Jarek Poplawski", "email": "jarkao2@gmail.com", "time": "Tue Apr 01 23:56:17 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 01 23:56:17 2008 -0700" }, "message": "[ROSE/AX25] af_rose: rose_release() fix\n\nrose_release() doesn\u0027t release sockets properly, e.g. it skips\nsock_orphan(), so OOPSes are triggered in sock_def_write_space(),\nwhich was observed especially while ROSE skbs were kfreed from\nax25_frames_acked(). There is also sock_hold() and lock_sock() added -\nsimilarly to ax25_release(). Thanks to Bernard Pidoux for substantial\nhelp in debugging this problem.\n\nSigned-off-by: Jarek Poplawski \u003cjarkao2@gmail.com\u003e\nReported-and-tested-by: Bernard Pidoux \u003cbpidoux@free.fr\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d43c7b37ad787173d08683f05eadeea0398fefdf", "tree": "3a618a512e0874876344300b5028b97c1dc61b7a", "parents": [ "e645890115004fa6c4dd1f9541bb834970c8bcae" ], "author": { "name": "Vladimir Koutny", "email": "vlado@work.ksp.sk", "time": "Mon Mar 31 17:05:03 2008 +0200" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Tue Apr 01 15:44:08 2008 -0400" }, "message": "mac80211: correct use_short_preamble handling\n\nERP IE bit for preamble mode is 0 for short and 1 for long, not the other\nway around. This fixes the value reported to the driver via\nbss_conf-\u003euse_short_preamble field.\n\nSigned-off-by: Vladimir Koutny \u003cvlado@ksp.sk\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "64f851e410ae37a0990212ae34f0c96b641478f7", "tree": "6ff8af643731d195b1d9439f77b4168f53fe0c62", "parents": [ "f83f1768f833cb45bc93429fdc552252a4f55ac3" ], "author": { "name": "Jan Niehusmann", "email": "jan@gondor.com", "time": "Sun Mar 23 20:23:56 2008 +0100" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Tue Apr 01 15:44:07 2008 -0400" }, "message": "mac80211: trigger ieee80211_sta_work after opening interface\n\nieee80211_sta_work is disabled while network interface\nis down. Therefore, if you configure wireless parameters\nbefore bringing the interface up, these configurations are\nnot yet effective and association fails.\n\nA workaround from userspace is calling a command like\n\u0027iwconfig wlan0 ap any\u0027 after the interface is brought up.\n\nTo fix this behaviour, trigger execution of ieee80211_sta_work from\nieee80211_open when in STA or IBSS mode.\n\nSigned-off-by: Jan Niehusmann \u003cjan@gondor.com\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "f83f1768f833cb45bc93429fdc552252a4f55ac3", "tree": "15de7d2df2fc3a35e0a6b933bb37aefcba2cc3ef", "parents": [ "b50660f1fe4ebd6129064e4fba0bd882b60c2425" ], "author": { "name": "Joonwoo Park", "email": "joonwpark81@gmail.com", "time": "Mon Mar 31 21:02:47 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 31 21:02:47 2008 -0700" }, "message": "[LLC]: skb allocation size for responses\n\nAllocate the skb for llc responses with the received packet size by\nusing the size adjustable llc_frame_alloc.\nDon\u0027t allocate useless extra payload.\nCleanup magic numbers.\n\nSo, this fixes oops.\nReported by Jim Westfall:\nkernel: skb_over_panic: text:c0541fc7 len:1000 put:997 head:c166ac00 data:c166ac2f tail:0xc166b017 end:0xc166ac80 dev:eth0\nkernel: ------------[ cut here ]------------\nkernel: kernel BUG at net/core/skbuff.c:95!\n\nSigned-off-by: Joonwoo Park \u003cjoonwpark81@gmail.com\u003e\nAcked-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b50660f1fe4ebd6129064e4fba0bd882b60c2425", "tree": "347215a85892028efa2b8b6d6abb5538f6505282", "parents": [ "5e0751f57ae1aa5cda285f8c565ce202d8d5e5a3" ], "author": { "name": "YOSHIFUJI Hideaki", "email": "yoshfuji@linux-ipv6.org", "time": "Mon Mar 31 19:38:15 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 31 19:38:15 2008 -0700" }, "message": "[IP] UDP: Use SEQ_START_TOKEN.\n\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4c7966b86b910d6d4869aba1d7417d053ac9682c", "tree": "ec65994e682b6c3e05533ddd1bb15986687f8ea1", "parents": [ "9f09243890a4e7d2e06d40b56f26a64f88c6ec8b" ], "author": { "name": "YOSHIFUJI Hideaki", "email": "yoshfuji@linux-ipv6.org", "time": "Mon Mar 31 19:30:45 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 31 19:30:45 2008 -0700" }, "message": "[IPV6] MCAST: Ensure to check multicast listener(s).\n\nIn ip6_mc_input(), we need to check whether we have listener(s) for\nthe packet.\n\nAfter commit ae7bf20a6316272acfcaef5d265b18aaa54b41e4, all packets\nfor multicast destinations are delivered to upper layer if\nIFF_PROMISC or IFF_ALLMULTI is set.\n\nIn fact, bug was rather ancient; the original (before the commit)\nintent of the dev-\u003eflags check was to skip the ipv6_chk_mcast_addr()\ncall, assuming L2 filters packets appropriately, but it was even not\ntrue.\n\nLet\u0027s explicitly check our multicast list.\n\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nAcked-by: David L Stevens \u003cdlstevens@us.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "91e916cffec7c0153c5cbaa447151862a7a9a047", "tree": "99688e2f9fba662029cba930b4eae286ef8326d0", "parents": [ "74dbf719ed3c49687dab507967ebab9189e91ab0" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Sat Mar 29 03:08:38 2008 +0000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 30 14:20:23 2008 -0700" }, "message": "net/rxrpc trivial annotations\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9f09243890a4e7d2e06d40b56f26a64f88c6ec8b", "tree": "3ac61960349da32c89b79c67eb957c7d914b72b0", "parents": [ "98846b5e2363e96a2b821e211bf0169afb38a54a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 19:51:40 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 19:51:40 2008 -0700" }, "message": "[LLC]: Kill llc_station_mac_sa symbol export.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e8e16b706e8406f1ab3bccab16932ebc513896d8", "tree": "75b52bb42601756995a1ffba206608f3f97b6df5", "parents": [ "0ef4730927c54cac5006759fa0cf50fbeff9d191" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 17:30:18 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 17:30:18 2008 -0700" }, "message": "[INET]: inet_frag_evictor() must run with BH disabled\n\nBased upon a lockdep trace from Dave Jones.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a5a04819c5740cb1aa217af2cc8f5ef26f33d744", "tree": "044d51abecaf80e0488d2f43df2c512952687bca", "parents": [ "27785d83e4256fedeff45256d4c827fdcb47f2ce" ], "author": { "name": "Joonwoo Park", "email": "joonwpark81@gmail.com", "time": "Fri Mar 28 16:28:36 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 16:28:36 2008 -0700" }, "message": "[LLC]: station source mac address\n\nkill unnecessary llc_station_mac_sa.\n\nSigned-off-by: Joonwoo Park \u003cjoonwpark81@gmail.com\u003e\nAcked-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "27785d83e4256fedeff45256d4c827fdcb47f2ce", "tree": "8b44dd3333aa0235774f0d227de7852f2b1472b2", "parents": [ "2ba2506ca7ca62c56edaa334b0fe61eb5eab6ab0" ], "author": { "name": "Joonwoo Park", "email": "joonwpark81@gmail.com", "time": "Fri Mar 28 16:27:33 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 16:27:33 2008 -0700" }, "message": "[LLC]: bogus llc packet length\n\ndiscard llc packet which has bogus packet length.\n\nSigned-off-by: Joonwoo Park \u003cjoonwpark81@gmail.com\u003e\nAcked-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2ba2506ca7ca62c56edaa334b0fe61eb5eab6ab0", "tree": "0a1b1f7046eff42e2323334b656543efdf85d38a", "parents": [ "32aced7509cb20ef3ec67c9b56f5b55c41dd4f8d" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Fri Mar 28 16:25:26 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 16:25:26 2008 -0700" }, "message": "[NET]: Add preemption point in qdisc_run\n\nThe qdisc_run loop is currently unbounded and runs entirely in a\nsoftirq. This is bad as it may create an unbounded softirq run.\n\nThis patch fixes this by calling need_resched and breaking out if\nnecessary.\n\nIt also adds a break out if the jiffies value changes since that would\nindicate we\u0027ve been transmitting for too long which starves other\nsoftirqs.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "32aced7509cb20ef3ec67c9b56f5b55c41dd4f8d", "tree": "a0efa0f46ec27fb45ea3c4caf55ccdc83ed019fb", "parents": [ "d41a95e04ae80b77ddc186d0d97e6b439684adb8" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Fri Mar 28 16:23:19 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 16:23:19 2008 -0700" }, "message": "[NET]: Don\u0027t send ICMP_FRAG_NEEDED for GSO packets\n\nCommit 9af3912ec9e30509b76cb376abb65a4d8af27df3 (\"[NET] Move DF check\nto ip_forward\") added a new check to send ICMP fragmentation needed\nfor large packets.\n\nUnlike the check in ip_finish_output(), it doesn\u0027t check for GSO.\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d5fb2962c6157495e1365e4f30568ed3830d35a7", "tree": "952baad67a821bf47b46f7cd4d748e0fa3144e90", "parents": [ "293a3839304cac32e58929db0c23e00a509fbfc5" ], "author": { "name": "Robert P. J. Day", "email": "rpjday@crashcourse.ca", "time": "Fri Mar 28 16:17:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 28 16:17:38 2008 -0700" }, "message": "bluetooth: replace deprecated RW_LOCK_UNLOCKED macros\n\nThe older RW_LOCK_UNLOCKED macros defeat lockdep state tracing so\nreplace them with the newer __RW_LOCK_UNLOCKED macros.\n\nSigned-off-by: Robert P. J. Day \u003crpjday@crashcourse.ca\u003e\nAcked-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3387b804d8850494bdf91d16800925a3fd46e37d", "tree": "d9ca604f5cad6d930ee85973672e4ee0d126eb67", "parents": [ "f706d5d22c35e18ed13a4b2b4991aac75bf39df5" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Fri Mar 28 14:15:57 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 28 14:45:22 2008 -0700" }, "message": "net/9p/trans_fd.c:p9_trans_fd_init(): module_init functions should return 0 on success\n\nMar 23 09:06:31 opensuse103 kernel: Installing 9P2000 support\nMar 23 09:06:31 opensuse103 kernel: sys_init_module: \u00279pnet_fd\u0027-\u003einit suspiciously returned 1, it should follow 0/-E convention\nMar 23 09:06:31 opensuse103 kernel: sys_init_module: loading module anyway...\nMar 23 09:06:31 opensuse103 kernel: Pid: 5323, comm: modprobe Not tainted 2.6.25-rc6-git7-default #1\nMar 23 09:06:31 opensuse103 kernel: [\u003cc013c253\u003e] sys_init_module+0x172b/0x17c9\nMar 23 09:06:31 opensuse103 kernel: [\u003cc0108a6a\u003e] sys_mmap2+0x62/0x77\nMar 23 09:06:31 opensuse103 kernel: [\u003cc01059c4\u003e] sysenter_past_esp+0x6d/0xa9\nMar 23 09:06:31 opensuse103 kernel: \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nCc: Latchesar Ionkov \u003clucho@ionkov.net\u003e\nCc: Eric Van Hensbergen \u003cericvh@opteron.(none)\u003e\nCc: David S. Miller \u003cdavem@davemloft.net\u003e\nCc: \"Rafael J. Wysocki\" \u003crjw@sisk.pl\u003e\nCc: \u003cdevzero@web.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3480c63bdf008e9289aab94418f43b9592978fff", "tree": "14a355d650dc467e105fe00bc62a3599e890bc50", "parents": [ "349fb2d6492e7d644bc8e670ba11d764a3e2901b" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Thu Mar 27 20:28:10 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 27 20:28:10 2008 -0700" }, "message": "[LLC]: Restrict LLC sockets to root\n\nLLC currently allows users to inject raw frames, including IP packets\nencapsulated in SNAP. While Linux doesn\u0027t handle IP over SNAP, other\nsystems do. Restrict LLC sockets to root similar to packet sockets.\n\n[ Modified Patrick\u0027s patch to use CAP_NEW_RAW --DaveM ]\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8eeee8b152ae6bbe181518efaf62ba8e9c613693", "tree": "b5900762ddf2a5d03ed31da629b726c6fd6a710f", "parents": [ "920fc941a9617f95ccb283037fe6f8a38d95bb69" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Thu Mar 27 16:55:53 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 27 16:55:53 2008 -0700" }, "message": "[NETFILTER]: Replate direct proc_fops assignment with proc_create call.\n\nThis elliminates infamous race during module loading when one could lookup\nproc entry without proc_fops assigned.\n\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "920fc941a9617f95ccb283037fe6f8a38d95bb69", "tree": "1de9a439eec25d514f25254320ded076f5fd8144", "parents": [ "732c8bd590625e8bc0b88313b82930e336b2bec4" ], "author": { "name": "Thomas Graf", "email": "tgraf@suug.ch", "time": "Thu Mar 27 16:08:03 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 27 16:08:03 2008 -0700" }, "message": "[ESP]: Ensure IV is in linear part of the skb to avoid BUG() due to OOB access\n\nESP does not account for the IV size when calling pskb_may_pull() to\nensure everything it accesses directly is within the linear part of a\npotential fragment. This results in a BUG() being triggered when the\nboth the IPv4 and IPv6 ESP stack is fed with an skb where the first\nfragment ends between the end of the esp header and the end of the IV.\n\nThis bug was found by Dirk Nehring \u003cdnehring@gmx.net\u003e .\n\nSigned-off-by: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ee20a0dd5400e771ff93eb7f2bcc731eab1f377e", "tree": "43743bd300a5b1ed94888881689e1d6a8ecac1e1", "parents": [ "d55a4528f7f607ca2872fec18574bc8cec060f05", "732c8bd590625e8bc0b88313b82930e336b2bec4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 26 18:35:50 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 26 18:35:50 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (43 commits)\n [IPSEC]: Fix BEET output\n [ICMP]: Dst entry leak in icmp_send host re-lookup code (v2).\n [AX25]: Remove obsolete references to BKL from TODO file.\n [NET]: Fix multicast device ioctl checks\n [IRDA]: Store irnet_socket termios properly.\n [UML]: uml-net: don\u0027t set IFF_ALLMULTI in set_multicast_list\n [VLAN]: Don\u0027t copy ALLMULTI/PROMISC flags from underlying device\n netxen, phy/marvell, skge: minor checkpatch fixes\n S2io: Handle TX completions on the same CPU as the sender for MIS-X interrupts\n b44: Truncate PHY address\n skge napi-\u003epoll() locking bug\n rndis_host: fix oops when query for OID_GEN_PHYSICAL_MEDIUM fails\n cxgb3: Fix lockdep problems with sge.reg_lock\n ehea: Fix IPv6 support\n dm9000: Support promisc and all-multi modes\n dm9601: configure MAC to drop invalid (crc/length) packets\n dm9601: add Hirose USB-100 device ID\n Marvell PHY m88e1111 driver fix\n netxen: fix rx dropped stats\n netxen: remove low level tx lock\n ...\n" }, { "commit": "732c8bd590625e8bc0b88313b82930e336b2bec4", "tree": "1cf3f6b9db7b886d35bbf38bab2ca6ff82d206d8", "parents": [ "0e5606e4f46b7cf52dd445af01e71ed9dbb7f735" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Wed Mar 26 16:51:09 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 16:51:09 2008 -0700" }, "message": "[IPSEC]: Fix BEET output\n\nThe IPv6 BEET output function is incorrectly including the inner\nheader in the payload to be protected. This causes a crash as\nthe packet doesn\u0027t actually have that many bytes for a second\nheader.\n\nThe IPv4 BEET output on the other hand is broken when it comes\nto handling an inner IPv6 header since it always assumes an\ninner IPv4 header.\n\nThis patch fixes both by making sure that neither BEET output\nfunction touches the inner header at all. All access is now\ndone through the protocol-independent cb structure. Two new\nattributes are added to make this work, the IP header length\nand the IPv4 option length. They\u0027re filled in by the inner\nmode\u0027s output function.\n\nThanks to Joakim Koskela for finding this problem.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c8237a5fcea9d49a73275b4c8f541dd42f8da1a4", "tree": "b93287513360a027baf3a8936c457ab2e21aa532", "parents": [ "12c22d6ef299ccf0955e5756eb57d90d7577ac68" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Tue Mar 25 22:27:19 2008 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 26 11:24:19 2008 -0700" }, "message": "SVCRDMA: Check num_sge when setting LAST_CTXT bit\n\nThe RDMACTXT_F_LAST_CTXT bit was getting set incorrectly\nwhen the last chunk in the read-list spanned multiple pages. This\nresulted in a kernel panic when the wrong context was used to\nbuild the RPC iovec page list.\n\nRDMA_READ is used to fetch RPC data from the client for\nNFS_WRITE requests. A scatter-gather is used to map the\nadvertised client side buffer to the server-side iovec and\nassociated page list.\n\nWR contexts are used to convey which scatter-gather entries are\nhandled by each WR. When the write data is large, a single RPC may\nrequire multiple RDMA_READ requests so the contexts for a single RPC\nare chained together in a linked list. The last context in this list\nis marked with a bit RDMACTXT_F_LAST_CTXT so that when this WR completes,\nthe CQ handler code can enqueue the RPC for processing.\n\nThe code in rdma_read_xdr was setting this bit on the last two\ncontexts on this list when the last read-list chunk spanned multiple\npages. This caused the svc_rdma_recvfrom logic to incorrectly build\nthe RPC and caused the kernel to crash because the second-to-last\ncontext doesn\u0027t contain the iovec page list.\n\nModified the condition that sets this bit so that it correctly detects\nthe last context for the RPC.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\nTested-by: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7c0ecc4c4f8fd90988aab8a95297b9c0038b6160", "tree": "3c937731c0bf2c3ddfd94e2141d83af596f98d9e", "parents": [ "5c2e2e239ebe44e3fdc5f2ae270d96c4ceee4e9a" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Wed Mar 26 02:27:09 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 02:27:09 2008 -0700" }, "message": "[ICMP]: Dst entry leak in icmp_send host re-lookup code (v2).\n\nCommit 8b7817f3a959ed99d7443afc12f78a7e1fcc2063 ([IPSEC]: Add ICMP host\nrelookup support) introduced some dst leaks on error paths: the rt\npointer can be forgotten to be put. Fix it bu going to a proper label.\n\nFound after net namespace\u0027s lo refused to unregister :) Many thanks to \nDen for valuable help during debugging.\n\nHerbert pointed out, that xfrm_lookup() will put the rtable in case\nof error itself, so the first goto fix is redundant.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5c2e2e239ebe44e3fdc5f2ae270d96c4ceee4e9a", "tree": "ea33a8e870904d3eccc273dc39b612f8ce3aa543", "parents": [ "61ee6bd487b9cc160e533034eb338f2085dc7922" ], "author": { "name": "Robert P. J. Day", "email": "rpjday@crashcourse.ca", "time": "Wed Mar 26 02:14:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 02:14:38 2008 -0700" }, "message": "[AX25]: Remove obsolete references to BKL from TODO file.\n\nGiven that there are no apparent calls to lock_kernel() or\nunlock_kernel() under net/ax25, delete the TODO reference related to\nthat.\n\nSigned-off-by: Robert P. J. Day \u003crpjday@crashcourse.ca\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "61ee6bd487b9cc160e533034eb338f2085dc7922", "tree": "33502a9da58f26f3b948f18288c0669b898db030", "parents": [ "8c7230f781749cd7261b504c0bfa188bb96e77ee" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Wed Mar 26 02:12:11 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 02:12:11 2008 -0700" }, "message": "[NET]: Fix multicast device ioctl checks\n\nSIOCADDMULTI/SIOCDELMULTI check whether the driver has a set_multicast_list\nmethod to determine whether it supports multicast. Drivers implementing\nsecondary unicast support use set_rx_mode however.\n\nCheck for both dev-\u003eset_multicast_mode and dev-\u003eset_rx_mode to determine\nmulticast capabilities.\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8c7230f781749cd7261b504c0bfa188bb96e77ee", "tree": "b62a6657e147284af9ec2a1210d7d5964c6bdc27", "parents": [ "f70c176619dde7f32283fc5c94df8e9aaac12729" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 00:55:50 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 00:55:50 2008 -0700" }, "message": "[IRDA]: Store irnet_socket termios properly.\n\nIt should be a \"struct ktermios\" not a \"struct termios\".\n\nBased upon a build warning reported by Stephen Rothwell.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0ed21b321a13421e2dfeaa70a6c324e05e3e91e6", "tree": "edaabd8f35f77f6ef4e39c659621d3017845d944", "parents": [ "f0c88f9c45f39acd017328515890481adcb32607" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Wed Mar 26 00:15:17 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 26 00:15:17 2008 -0700" }, "message": "[VLAN]: Don\u0027t copy ALLMULTI/PROMISC flags from underlying device\n\nChanging these flags requires to use dev_set_allmulti/dev_set_promiscuity\nor dev_change_flags. Setting it directly causes two unwanted effects:\n\n- the next dev_change_flags call will notice a difference between\n dev-\u003egflags and the actual flags, enable promisc/allmulti\n mode and incorrectly update dev-\u003egflags\n\n- this keeps the underlying device in promisc/allmulti mode until\n the VLAN device is deleted\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "df9dcb4588aca9cc243cf1f3f454361a84e1cbdb", "tree": "53dabed7cffee752109808cbea2f812e0a6d7faf", "parents": [ "fa86d322d89995fef1bfb5cc768b89d8c22ea0d9" ], "author": { "name": "Kazunori MIYAZAWA", "email": "kazunori@miyazawa.org", "time": "Mon Mar 24 14:51:51 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 24 14:51:51 2008 -0700" }, "message": "[IPSEC]: Fix inter address family IPsec tunnel handling.\n\nSigned-off-by: Kazunori MIYAZAWA \u003ckazunori@miyazawa.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fa86d322d89995fef1bfb5cc768b89d8c22ea0d9", "tree": "e657b8adc9ccd2e13b2e2276fab4733a273ded09", "parents": [ "8f3ea33a5078a09eba12bfe57424507809367756" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Mon Mar 24 14:48:59 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 24 14:48:59 2008 -0700" }, "message": "[NEIGH]: Fix race between pneigh deletion and ipv6\u0027s ndisc_recv_ns (v3).\n\nProxy neighbors do not have any reference counting, so any caller\nof pneigh_lookup (unless it\u0027s a netlink triggered add/del routine)\nshould _not_ perform any actions on the found proxy entry. \n\nThere\u0027s one exception from this rule - the ipv6\u0027s ndisc_recv_ns() \nuses found entry to check the flags for NTF_ROUTER.\n\nThis creates a race between the ndisc and pneigh_delete - after \nthe pneigh is returned to the caller, the nd_tbl.lock is dropped \nand the deleting procedure may proceed.\n\nOne of the fixes would be to add a reference counting, but this\nproblem exists for ndisc only. Besides such a patch would be too \nbig for -rc4.\n\nSo I propose to introduce a __pneigh_lookup() which is supposed\nto be called with the lock held and use it in ndisc code to check\nthe flags on alive pneigh entry.\n\n\nChanges from v2:\nAs David noticed, Exported the __pneigh_lookup() to ipv6 module. \nThe checkpatch generates a warning on it, since the EXPORT_SYMBOL \ndoes not follow the symbol itself, but in this file all the \nexports come at the end, so I decided no to break this harmony.\n\nChanges from v1:\nFixed comments from YOSHIFUJI - indentation of prototype in header\nand the pndisc_check_router() name - and a compilation fix, pointed\nby Daniel - the is_routed was (falsely) considered as uninitialized\nby gcc.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ca1a6ba57c5fca755b4ac7a13395bca2e2e371b1", "tree": "1dd8f5090238d644668818c54f9f91b1c5665991", "parents": [ "d3073779f8362d64b804882f5f41c208c4a5e11e", "8f3ea33a5078a09eba12bfe57424507809367756" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 24 13:07:24 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 24 13:07:24 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:\n sch_htb: fix \"too many events\" situation\n connector: convert to single-threaded workqueue\n [ATM]: When proc_create() fails, do some error handling work and return -ENOMEM.\n [SUNGEM]: Fix NAPI assertion failure.\n BNX2X: prevent ethtool from setting port type\n [9P] net/9p/trans_fd.c: remove unused variable\n [IPV6] net/ipv6/ndisc.c: remove unused variable\n [IPV4] fib_trie: fix warning from rcu_assign_poinger\n [TCP]: Let skbs grow over a page on fast peers\n [DLCI]: Fix tiny race between module unload and sock_ioctl.\n [SCTP]: Fix build warnings with IPV6 disabled.\n [IPV4]: Fix null dereference in ip_defrag\n" }, { "commit": "d3073779f8362d64b804882f5f41c208c4a5e11e", "tree": "6aa6a950cc9bed39deafd73bca1e99e108d31b51", "parents": [ "b9e76a00749521f2b080fa8a4fb15f66538ab756" ], "author": { "name": "Roland Dreier", "email": "rdreier@cisco.com", "time": "Mon Mar 24 12:03:03 2008 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 24 11:25:25 2008 -0700" }, "message": "SVCRDMA: Use only 1 RDMA read scatter entry for iWARP adapters\n\nThe iWARP protocol limits RDMA read requests to a single scatter\nentry. NFS/RDMA has code in rdma_read_max_sge() that is supposed to\nlimit the sge_count for RDMA read requests to 1, but the code to do\nthat is inside an #ifdef RDMA_TRANSPORT_IWARP block. In the mainline\nkernel at least, RDMA_TRANSPORT_IWARP is an enum and not a\npreprocessor #define, so the #ifdef\u0027ed code is never compiled.\n\nIn my test of a kernel build with -j8 on an NFS/RDMA mount, this\nproblem eventually leads to trouble starting with:\n\n svcrdma: Error posting send \u003d -22\n svcrdma : RDMA_READ error \u003d -22\n\nand things go downhill from there.\n\nThe trivial fix is to delete the #ifdef guard. The check seems to be\na remnant of when the NFS/RDMA code was not merged and needed to\ncompile against multiple kernel versions, although I don\u0027t think it\never worked as intended. In any case now that the code is upstream\nthere\u0027s no need to test whether the RDMA_TRANSPORT_IWARP constant is\ndefined or not.\n\nWithout this patch, my kernel build on an NFS/RDMA mount using NetEffect\nadapters quickly and 100% reproducibly failed with an error like:\n\n ld: final link failed: Software caused connection abort\n\nWith the patch applied I was able to complete a kernel build on the\nsame setup.\n\n(Tom Tucker says this is \"actually an _ancient_ remnant when it had to\ncompile against iWARP vs. non-iWARP enabled OFA trees.\")\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nAcked-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8f3ea33a5078a09eba12bfe57424507809367756", "tree": "4795705a232810311c72bc6dda939399653806f1", "parents": [ "4b1b366721101f2f0d2350fbdccb679f7909cf57" ], "author": { "name": "Martin Devera", "email": "devik@cdi.cz", "time": "Sun Mar 23 22:00:38 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Mar 23 22:00:38 2008 -0700" }, "message": "sch_htb: fix \"too many events\" situation\n\nHTB is event driven algorithm and part of its work is to apply\nscheduled events at proper times. It tried to defend itself from\nlivelock by processing only limited number of events per dequeue.\nBecause of faster computers some users already hit this hardcoded\nlimit.\n\nThis patch limits processing up to 2 jiffies (why not 1 jiffie ?\nbecause it might stop prematurely when only fraction of jiffie\nremains).\n\nSigned-off-by: Martin Devera \u003cdevik@cdi.cz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "dbee0d3f4603b9d0e56234a0743321fe4dad31ca", "tree": "d13b829d46ae603b8e7772ecae72e9be1fdfb7af", "parents": [ "da990a2402aeaee84837f29054c4628eb02f7493" ], "author": { "name": "Wang Chen", "email": "wangchen@cn.fujitsu.com", "time": "Sun Mar 23 21:45:36 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Mar 23 21:45:36 2008 -0700" }, "message": "[ATM]: When proc_create() fails, do some error handling work and return -ENOMEM.\n\nSigned-off-by: Wang Chen \u003cwangchen@cn.fujitsu.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "53a6201fdfa04accc91ea1a7accce8e8bc37ef8e", "tree": "6b94d41162db9cf344116ad157a5bfd0e628377b", "parents": [ "421f099bc555c5f1516fdf5060de1d6bb5f51002" ], "author": { "name": "Julia Lawall", "email": "julia@diku.dk", "time": "Sat Mar 22 18:05:33 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 22 18:05:33 2008 -0700" }, "message": "[9P] net/9p/trans_fd.c: remove unused variable\n\nThe variable cb is initialized but never used otherwise.\n\nThe semantic patch that makes this change is as follows:\n(http://www.emn.fr/x-info/coccinelle/)\n\n// \u003csmpl\u003e\n@@\ntype T;\nidentifier i;\nconstant C;\n@@\n\n(\nextern T i;\n|\n- T i;\n \u003c+... when !\u003d i\n- i \u003d C;\n ...+\u003e\n)\n// \u003c/smpl\u003e\n\nSigned-off-by: Julia Lawall \u003cjulia@diku.dk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "421f099bc555c5f1516fdf5060de1d6bb5f51002", "tree": "e6175f1bd48428f9159eb91a81ba1cbb0418026d", "parents": [ "6440cc9e0f48ade57af7be28008cbfa6a991f287" ], "author": { "name": "Julia Lawall", "email": "julia@diku.dk", "time": "Sat Mar 22 18:04:16 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 22 18:04:16 2008 -0700" }, "message": "[IPV6] net/ipv6/ndisc.c: remove unused variable\n\nThe variable hlen is initialized but never used otherwise.\n\nThe semantic patch that makes this change is as follows:\n(http://www.emn.fr/x-info/coccinelle/)\n\n// \u003csmpl\u003e\n@@\ntype T;\nidentifier i;\nconstant C;\n@@\n\n(\nextern T i;\n|\n- T i;\n \u003c+... when !\u003d i\n- i \u003d C;\n ...+\u003e\n)\n// \u003c/smpl\u003e\n\nSigned-off-by: Julia Lawall \u003cjulia@diku.dk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6440cc9e0f48ade57af7be28008cbfa6a991f287", "tree": "233300e665ed862dab958087b8afa31baaf0ef5f", "parents": [ "69d1506731168d6845a76a303b2c45f7c05f3f2c" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@vyatta.com", "time": "Sat Mar 22 17:59:58 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 22 17:59:58 2008 -0700" }, "message": "[IPV4] fib_trie: fix warning from rcu_assign_poinger\n\nThis gets rid of a warning caused by the test in rcu_assign_pointer.\nI tried to fix rcu_assign_pointer, but that devolved into a long set\nof discussions about doing it right that came to no real solution.\nSince the test in rcu_assign_pointer for constant NULL would never\nsucceed in fib_trie, just open code instead.\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nAcked-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "69d1506731168d6845a76a303b2c45f7c05f3f2c", "tree": "3bedf2680b30c09b0375616a1c2b0d291a9f376f", "parents": [ "7512cbf6efc97644812f137527a54b8e92b6a90a" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Sat Mar 22 15:47:05 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 22 15:47:05 2008 -0700" }, "message": "[TCP]: Let skbs grow over a page on fast peers\n\nWhile testing the virtio-net driver on KVM with TSO I noticed\nthat TSO performance with a 1500 MTU is significantly worse\ncompared to the performance of non-TSO with a 16436 MTU. The\npacket dump shows that most of the packets sent are smaller\nthan a page.\n\nLooking at the code this actually is quite obvious as it always\nstop extending the packet if it\u0027s the first packet yet to be\nsent and if it\u0027s larger than the MSS. Since each extension is\nbound by the page size, this means that (given a 1500 MTU) we\u0027re\nvery unlikely to construct packets greater than a page, provided\nthat the receiver and the path is fast enough so that packets can\nalways be sent immediately.\n\nThe fix is also quite obvious. The push calls inside the loop\nis just an optimisation so that we don\u0027t end up doing all the\nsending at the end of the loop. Therefore there is no specific\nreason why it has to do so at MSS boundaries. For TSO, the\nmost natural extension of this optimisation is to do the pushing\nonce the skb exceeds the TSO size goal.\n\nThis is what the patch does and testing with KVM shows that the\nTSO performance with a 1500 MTU easily surpasses that of a 16436\nMTU and indeed the packet sizes sent are generally larger than\n16436.\n\nI don\u0027t see any obvious downsides for slower peers or connections,\nbut it would be prudent to test this extensively to ensure that\nthose cases don\u0027t regress.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7512cbf6efc97644812f137527a54b8e92b6a90a", "tree": "bf603e7ede2af1163c866bc04e540807d1a298c1", "parents": [ "1233823b0847190976d69a86d7bb1287992ba2c7" ], "author": { "name": "Pavel Emelyanov", "email": "xemul@openvz.org", "time": "Fri Mar 21 15:58:52 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 21 15:58:52 2008 -0700" }, "message": "[DLCI]: Fix tiny race between module unload and sock_ioctl.\n\nThis is a narrow pedantry :) but the dlci_ioctl_hook check and call\nshould not be parted with the mutex lock.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "12b101555f4a67db67a66966a516075bd477741f", "tree": "940f5aab5d774a7265ffa47977ab1cad8e3bcc6b", "parents": [ "94833dfb8c98ed4ca1944dd2c1339d88a2d1c758" ], "author": { "name": "Phil Oester", "email": "kernel@linuxace.com", "time": "Fri Mar 21 15:01:50 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Mar 21 15:01:50 2008 -0700" }, "message": "[IPV4]: Fix null dereference in ip_defrag\n\nBeen seeing occasional panics in my testing of 2.6.25-rc in ip_defrag.\nOffending line in ip_defrag is here:\n\n\tnet \u003d skb-\u003edev-\u003end_net\n\nwhere dev is NULL. Bisected the problem down to commit\nac18e7509e7df327e30d6e073a787d922eaf211d ([NETNS][FRAGS]: Make the\ninet_frag_queue lookup work in namespaces). \n\nBelow patch (idea from Patrick McHardy) fixes the problem for me.\n\nSigned-off-by: Phil Oester \u003ckernel@linuxace.com\u003e\nAcked-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7d3628b230ecbdc29566c18bc7800ff8ed66a71f", "tree": "eac1bff40bdb655fdfdeaf1e22ce12a81296f1fb", "parents": [ "2c7871982cf27caaddbaeb7e2121ce1374b520ff", "94833dfb8c98ed4ca1944dd2c1339d88a2d1c758" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 21 07:57:45 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 21 07:57:45 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (46 commits)\n [NET] ifb: set separate lockdep classes for queue locks\n [IPV6] KCONFIG: Fix description about IPV6_TUNNEL.\n [TCP]: Fix shrinking windows with window scaling\n netpoll: zap_completion_queue: adjust skb-\u003eusers counter\n bridge: use time_before() in br_fdb_cleanup()\n [TG3]: Fix build warning on sparc32.\n MAINTAINERS: bluez-devel is subscribers-only\n audit: netlink socket can be auto-bound to pid other than current-\u003epid (v2)\n [NET]: Fix permissions of /proc/net\n [SCTP]: Fix a race between module load and protosw access\n [NETFILTER]: ipt_recent: sanity check hit count\n [NETFILTER]: nf_conntrack_h323: logical-bitwise \u0026 confusion in process_setup()\n [RT2X00] drivers/net/wireless/rt2x00/rt2x00dev.c: remove dead code, fix warning\n [IPV4]: esp_output() misannotations\n [8021Q]: vlan_dev misannotations\n xfrm: -\u003eeth_proto is __be16\n [IPV4]: ipv4_is_lbcast() misannotations\n [SUNRPC]: net/* NULL noise\n [SCTP]: fix misannotated __sctp_rcv_asconf_lookup()\n [PKT_SCHED]: annotate cls_u32\n ...\n" }, { "commit": "38fe999e2286139cccdaa500a81bd49a16a81158", "tree": "317cb01faf3c2fd08a908e694a61df27006f3fea", "parents": [ "607bfbf2d55dd1cfe5368b41c2a81a8c9ccf4723" ], "author": { "name": "YOSHIFUJI Hideaki", "email": "yoshfuji@linux-ipv6.org", "time": "Thu Mar 20 16:13:58 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 16:13:58 2008 -0700" }, "message": "[IPV6] KCONFIG: Fix description about IPV6_TUNNEL.\n\nBased on notice from \"Colin\" \u003ccolins@sjtu.edu.cn\u003e.\n\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "607bfbf2d55dd1cfe5368b41c2a81a8c9ccf4723", "tree": "3e773dec77c677d01cfd1bce37ee9f292b19d6a3", "parents": [ "8a455b087c9629b3ae3b521b4f1ed16672f978cc" ], "author": { "name": "Patrick McHardy", "email": "kaber@trash.net", "time": "Thu Mar 20 16:11:27 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 16:11:27 2008 -0700" }, "message": "[TCP]: Fix shrinking windows with window scaling\n\nWhen selecting a new window, tcp_select_window() tries not to shrink\nthe offered window by using the maximum of the remaining offered window\nsize and the newly calculated window size. The newly calculated window\nsize is always a multiple of the window scaling factor, the remaining\nwindow size however might not be since it depends on rcv_wup/rcv_nxt.\nThis means we\u0027re effectively shrinking the window when scaling it down.\n\n\nThe dump below shows the problem (scaling factor 2^7):\n\n- Window size of 557 (71296) is advertised, up to 3111907257:\n\nIP 172.2.2.3.33000 \u003e 172.2.2.2.33000: . ack 3111835961 win 557 \u003c...\u003e\n\n- New window size of 514 (65792) is advertised, up to 3111907217, 40 bytes\n below the last end:\n\nIP 172.2.2.3.33000 \u003e 172.2.2.2.33000: . 3113575668:3113577116(1448) ack 3111841425 win 514 \u003c...\u003e\n\nThe number 40 results from downscaling the remaining window:\n\n3111907257 - 3111841425 \u003d 65832\n65832 / 2^7 \u003d 514\n65832 % 2^7 \u003d 40\n\nIf the sender uses up the entire window before it is shrunk, this can have\nchaotic effects on the connection. When sending ACKs, tcp_acceptable_seq()\nwill notice that the window has been shrunk since tcp_wnd_end() is before\ntp-\u003esnd_nxt, which makes it choose tcp_wnd_end() as sequence number.\nThis will fail the receivers checks in tcp_sequence() however since it\nis before it\u0027s tp-\u003ercv_wup, making it respond with a dupack.\n\nIf both sides are in this condition, this leads to a constant flood of\nACKs until the connection times out.\n\nMake sure the window is never shrunk by aligning the remaining window to\nthe window scaling factor.\n\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8a455b087c9629b3ae3b521b4f1ed16672f978cc", "tree": "db93f743a1d9add48876d9d0ccdbc2696c0f771f", "parents": [ "2bec008ca9fd009aa503b75344d1c22da9256141" ], "author": { "name": "Jarek Poplawski", "email": "jarkao2@gmail.com", "time": "Thu Mar 20 16:07:27 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 16:07:27 2008 -0700" }, "message": "netpoll: zap_completion_queue: adjust skb-\u003eusers counter\n\nzap_completion_queue() retrieves skbs from completion_queue where they have\nzero skb-\u003eusers counter. Before dev_kfree_skb_any() it should be non-zero\nyet, so it\u0027s increased now.\n\nReported-and-tested-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Jarek Poplawski \u003cjarkao2@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2bec008ca9fd009aa503b75344d1c22da9256141", "tree": "5c3e97da66e6d0346d9de30d2fd48764e6ac29fd", "parents": [ "7582a33557cc6dc42b4c6918c6e7f8e465b72a70" ], "author": { "name": "Fabio Checconi", "email": "fabio@gandalf.sssup.it", "time": "Thu Mar 20 15:54:58 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 15:54:58 2008 -0700" }, "message": "bridge: use time_before() in br_fdb_cleanup()\n\nIn br_fdb_cleanup() next_timer and this_timer are in jiffies, so they\nshould be compared using the time_after() macro.\n\nSigned-off-by: Fabio Checconi \u003cfabio@gandalf.sssup.it\u003e\nSigned-off-by: Stephen Hemminger \u003cstephen.hemminger@vyatta.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "270637abff0cdf848b910b9f96ad342e1da61c66", "tree": "c5335ff19071e083588240da49b2aac27a402d8b", "parents": [ "d0ebf133590abdc035af6e19a6568667af0ab3b0" ], "author": { "name": "Vlad Yasevich", "email": "vladislav.yasevich@hp.com", "time": "Thu Mar 20 15:17:14 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 15:17:14 2008 -0700" }, "message": "[SCTP]: Fix a race between module load and protosw access\n\nThere is a race is SCTP between the loading of the module\nand the access by the socket layer to the protocol functions.\nIn particular, a list of addresss that SCTP maintains is\nnot initialized prior to the registration with the protosw.\nThus it is possible for a user application to gain access\nto SCTP functions before everything has been initialized.\nThe problem shows up as odd crashes during connection\ninitializtion when we try to access the SCTP address list.\n\nThe solution is to refactor how we do registration and\ninitialize the lists prior to registering with the protosw.\nCare must be taken since the address list initialization\ndepends on some other pieces of SCTP initialization. Also\nthe clean-up in case of failure now also needs to be refactored.\n\nSigned-off-by: Vlad Yasevich \u003cvladislav.yasevich@hp.com\u003e\nAcked-by: Sridhar Samudrala \u003csri@us.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d0ebf133590abdc035af6e19a6568667af0ab3b0", "tree": "a17acfb490969d7a99c902feab1d29ac6acbde2f", "parents": [ "6aebb9b280e5662ece41cf570e25e61795443985" ], "author": { "name": "Daniel Hokka Zakrisson", "email": "daniel@hozac.com", "time": "Thu Mar 20 15:07:10 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 15:07:10 2008 -0700" }, "message": "[NETFILTER]: ipt_recent: sanity check hit count\n\nIf a rule using ipt_recent is created with a hit count greater than\nip_pkt_list_tot, the rule will never match as it cannot keep track\nof enough timestamps. This patch makes ipt_recent refuse to create such\nrules.\n\nWith ip_pkt_list_tot\u0027s default value of 20, the following can be used\nto reproduce the problem.\n\nnc -u -l 0.0.0.0 1234 \u0026\nfor i in `seq 1 100`; do echo $i | nc -w 1 -u 127.0.0.1 1234; done\n\nThis limits it to 20 packets:\niptables -A OUTPUT -p udp --dport 1234 -m recent --set --name test \\\n --rsource\niptables -A OUTPUT -p udp --dport 1234 -m recent --update --seconds \\\n 60 --hitcount 20 --name test --rsource -j DROP\n\nWhile this is unlimited:\niptables -A OUTPUT -p udp --dport 1234 -m recent --set --name test \\\n --rsource\niptables -A OUTPUT -p udp --dport 1234 -m recent --update --seconds \\\n 60 --hitcount 21 --name test --rsource -j DROP\n\nWith the patch the second rule-set will throw an EINVAL.\n\nReported-by: Sean Kennedy \u003cskennedy@vcn.com\u003e\nSigned-off-by: Daniel Hokka Zakrisson \u003cdaniel@hozac.com\u003e\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6aebb9b280e5662ece41cf570e25e61795443985", "tree": "97f654f3f7833f583e34e576e0c852a3000ff6f0", "parents": [ "247df4548fdbb0f263aaa0386bbaf52bc359a972" ], "author": { "name": "Roel Kluin", "email": "12o3l@tiscali.nl", "time": "Thu Mar 20 15:06:23 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 20 15:06:23 2008 -0700" }, "message": "[NETFILTER]: nf_conntrack_h323: logical-bitwise \u0026 confusion in process_setup()\n\nlogical-bitwise \u0026 confusion\n\nSigned-off-by: Roel Kluin \u003c12o3l@tiscali.nl\u003e\nSigned-off-by: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6f3d09291b4982991680b61763b2541e53e2a95f", "tree": "683ca6bd7c93635ba315fa2f939de74124a39196", "parents": [ "f48273860edfca2306236d0f0de609aab3f773d4" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 01:44:24 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Mar 19 04:27:53 2008 +0100" }, "message": "sched, net: socket wakeups are sync\n\n\u0027sync\u0027 wakeups are a hint towards the scheduler that (certain)\nnetworking related wakeups likely create coupling between tasks.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "2f633928cbba8a5858bb39b11e7219a41b0fbef5", "tree": "9a82f4b7f2c3afe4b0208d8e44ea61bae90a7d22", "parents": [ "5e226e4d9016daee170699f8a4188a5505021756", "bde4f8fa8db2abd5ac9c542d76012d0fedab050f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 23:44:31 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 23:44:31 2008 -0700" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6\n" }, { "commit": "5e226e4d9016daee170699f8a4188a5505021756", "tree": "10a4024a21caca45ae1f57997f45541e966a9492", "parents": [ "9534f035ee67630688447e8aac91b1d365a51913" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 17 22:50:23 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 22:50:23 2008 -0700" }, "message": "[IPV4]: esp_output() misannotations\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9534f035ee67630688447e8aac91b1d365a51913", "tree": "05af741aea9b709f6aa0227ed8598edcd60b5d73", "parents": [ "8e3d716cce0c186ae9c88b91cc5686b78e0fd2d4" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 17 22:49:48 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 22:49:48 2008 -0700" }, "message": "[8021Q]: vlan_dev misannotations\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "27724426a9000086993a8107a11cff276c4bd4d4", "tree": "bb875f3df5b0b64e3bde48093674d857b2874901", "parents": [ "bc92dd194d05e8334b210552fbc0ac5711d72ea9" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 17 22:48:03 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 22:48:03 2008 -0700" }, "message": "[SUNRPC]: net/* NULL noise\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bc92dd194d05e8334b210552fbc0ac5711d72ea9", "tree": "58fb30bceeec669f3e8723c131ded67f7f9a2c10", "parents": [ "0382b9c35469be273ed10fa374496a924055a3c8" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 17 22:47:32 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 22:47:32 2008 -0700" }, "message": "[SCTP]: fix misannotated __sctp_rcv_asconf_lookup()\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0382b9c35469be273ed10fa374496a924055a3c8", "tree": "a71ff964e57ab8308bca073505aacfd7a3b4a764", "parents": [ "e6f1cebf71c4e7aae7dfa43414ce2631291def9f" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 17 22:46:46 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 22:46:46 2008 -0700" }, "message": "[PKT_SCHED]: annotate cls_u32\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e6f1cebf71c4e7aae7dfa43414ce2631291def9f", "tree": "a3abf6bb3e63f2ecea1a4e05bf8d288d920acb17", "parents": [ "fc4d30a6fc88121081916faef930d8c3e4ed6036" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 17 22:44:53 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 17 22:44:53 2008 -0700" }, "message": "[NET] endianness noise: INADDR_ANY\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "609eb39c8d8a8d2930780428f6cbe2f63eb84734", "tree": "0cf74a0e6e64531b31c5a1c9d45fbcb88f5e4d1f", "parents": [ "123d43acd2e55cd7db792d17c7e906db42cada42", "22626216c46f2ec86287e75ea86dd9ac3df54265" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 12 13:08:09 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 12 13:08:09 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (47 commits)\n [SCTP]: Fix local_addr deletions during list traversals.\n net: fix build with CONFIG_NET\u003dn\n [TCP]: Prevent sending past receiver window with TSO (at last skb)\n rt2x00: Add new D-Link USB ID\n rt2x00: never disable multicast because it disables broadcast too\n libertas: fix the \u0027compare command with itself\u0027 properly\n drivers/net/Kconfig: fix whitespace for GELIC_WIRELESS entry\n [NETFILTER]: nf_queue: don\u0027t return error when unregistering a non-existant handler\n [NETFILTER]: nfnetlink_queue: fix EPERM when binding/unbinding and instance 0 exists\n [NETFILTER]: nfnetlink_log: fix EPERM when binding/unbinding and instance 0 exists\n [NETFILTER]: nf_conntrack: replace horrible hack with ksize()\n [NETFILTER]: nf_conntrack: add \\n to \"expectation table full\" message\n [NETFILTER]: xt_time: fix failure to match on Sundays\n [NETFILTER]: nfnetlink_log: fix computation of netlink skb size\n [NETFILTER]: nfnetlink_queue: fix computation of allocated size for netlink skb.\n [NETFILTER]: nfnetlink: fix ifdef in nfnetlink_compat.h\n [NET]: include \u003clinux/types.h\u003e into linux/ethtool.h for __u* typedef\n [NET]: Make /proc/net a symlink on /proc/self/net (v3)\n RxRPC: fix rxrpc_recvmsg()\u0027s returning of msg_name\n net/enc28j60: oops fix\n ...\n" }, { "commit": "3fedb3c5a80595d94f7cbe47a6dba9184d869eb8", "tree": "50254172a14b3b67c7ab289341e4938cb2cd1273", "parents": [ "c48cbb405c4f338ce3263c44d621eff41d9a95fc" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Tue Mar 11 14:31:40 2008 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 12 12:37:34 2008 -0700" }, "message": "SVCRDMA: Fix erroneous BUG_ON in send_write\n\nThe assertion that checks for sge context overflow is\nincorrectly hard-coded to 32. This causes a kernel bug\ncheck when using big-data mounts. Changed the BUG_ON to\nuse the computed value RPCSVC_MAXPAGES.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ], "next": "c48cbb405c4f338ce3263c44d621eff41d9a95fc" }