)]}' { "log": [ { "commit": "04ed3e741d0f133e02bed7fa5c98edba128f90e7", "tree": "3dde4ca8306e98536faa69bccf0e47a2549c088f", "parents": [ "57422dc530115e427dff464cc0a32bcd0efb5008" ], "author": { "name": "Michał Mirosław", "email": "mirq-linux@rere.qmqm.pl", "time": "Mon Jan 24 15:32:47 2011 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 24 15:32:47 2011 -0800" }, "message": "net: change netdev-\u003efeatures to u32\n\nQuoting Ben Hutchings: we presumably won\u0027t be defining features that\ncan only be enabled on 64-bit architectures.\n\nOccurences found by `grep -r` on net/, drivers/net, include/\n\n[ Move features and vlan_features next to each other in\n struct netdev, as per Eric Dumazet\u0027s suggestion -DaveM ]\n\nSigned-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fe6c791570efe717946ea7b7dd50aec96b70d551", "tree": "1becb5e8aea7a9c9a7d78f987bd73b0a5d8ee434", "parents": [ "f8bf5681cf15f77692c8ad8cb95d059ff7c622c9", "f19872575ff7819a3723154657a497d9bca66b33" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 08 13:15:38 2010 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 08 13:47:38 2010 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/net/wireless/ath/ath9k/ar9003_eeprom.c\n\tnet/llc/af_llc.c\n" }, { "commit": "c39508d6f118308355468314ff414644115a07f3", "tree": "c69cfd271855fa4b81398f9ddcb3dc286489d71a", "parents": [ "4448008eb12f4b6bb9993584de8ec1d20b708d6f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 24 11:47:22 2010 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 24 11:47:22 2010 -0800" }, "message": "tcp: Make TCP_MAXSEG minimum more correct.\n\nUse TCP_MIN_MSS instead of constant 64.\n\nReported-by: Min Zhang \u003cmzhang@mvista.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c25ecd0a21d5e08160cb5cc984f9e2b8ee347443", "tree": "0e4dcacf1bf603f259b8d27445a10e60fa8d00d7", "parents": [ "190683a9d5457e6d962c232ffbecac3ab158dddd", "9457b24a0955bbdd2e89220a75de69fe09501bba" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 14 11:57:05 2010 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 14 11:57:05 2010 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n" }, { "commit": "7a1abd08d52fdeddb3e9a5a33f2f15cc6a5674d2", "tree": "f566af2a622e5416056c70633576a1a46b8bf6aa", "parents": [ "8d987e5c75107ca7515fa19e857cfa24aab6ec8f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 10 21:35:37 2010 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 10 21:35:37 2010 -0800" }, "message": "tcp: Increase TCP_MAXSEG socket option minimum.\n\nAs noted by Steve Chen, since commit\nf5fff5dc8a7a3f395b0525c02ba92c95d42b7390 (\"tcp: advertise MSS\nrequested by user\") we can end up with a situation where\ntcp_select_initial_window() does a divide by a zero (or\neven negative) mss value.\n\nThe problem is that sometimes we effectively subtract\nTCPOLEN_TSTAMP_ALIGNED and/or TCPOLEN_MD5SIG_ALIGNED from the mss.\n\nFix this by increasing the minimum from 8 to 64.\n\nReported-by: Steve Chen \u003cschen@mvista.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8d987e5c75107ca7515fa19e857cfa24aab6ec8f", "tree": "6392c5f08f0df39d42a079336f6be3960ac404dc", "parents": [ "67286640f638f5ad41a946b9a3dc75327950248f" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Tue Nov 09 23:24:26 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 10 12:12:00 2010 -0800" }, "message": "net: avoid limits overflow\n\nRobin Holt tried to boot a 16TB machine and found some limits were\nreached : sysctl_tcp_mem[2], sysctl_udp_mem[2]\n\nWe can switch infrastructure to use long \"instead\" of \"int\", now\natomic_long_t primitives are available for free.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nReported-by: Robin Holt \u003cholt@sgi.com\u003e\nReviewed-by: Robin Holt \u003cholt@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2af6fd8b18ceed416c9dfa675287c765aabf7d43", "tree": "6ddfb64096961839e0cf5a349ff8ff977ca9d542", "parents": [ "b194a3674fba6d9f9e470084d192c7cb99194a62" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Sat Oct 30 11:08:53 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 09 09:22:32 2010 -0800" }, "message": "net/ipv4/tcp.c: Update WARN uses\n\nCoalesce long formats.\nAlign arguments.\nRemove KERN_\u003clevel\u003e.\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "21a180cda012e1f93e362dd4a9b0bfd3d8c92940", "tree": "0e0d10baa3fdcd8ffbc6881076ff1695808dad9d", "parents": [ "c7d4426a98a5f6654cd0b4b33d9dab2e77192c18", "51e97a12bef19b7e43199fc153cf9bd5f2140362" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Oct 04 11:56:38 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Oct 04 11:56:38 2010 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tnet/ipv4/Kconfig\n\tnet/ipv4/tcp_timer.c\n" }, { "commit": "01db403cf99f739f86903314a489fb420e0e254f", "tree": "bf04fbfb3ed88d6cf7abeea1ab5209be36907882", "parents": [ "0b20406cda621c2495d10baab1e87127ceb43337" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 27 20:24:54 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 27 20:24:54 2010 -0700" }, "message": "tcp: Fix \u003e4GB writes on 64-bit.\n\nFixes kernel bugzilla #16603\n\ntcp_sendmsg() truncates iov_len to an \u0027int\u0027 which a 4GB write to write\nzero bytes, for example.\n\nThere is also the problem higher up of how verify_iovec() works. It\nwants to prevent the total length from looking like an error return\nvalue.\n\nHowever it does this using \u0027int\u0027, but syscalls return \u0027long\u0027 (and\nthus signed 64-bit on 64-bit machines). So it could trigger\nfalse-positives on 64-bit as written. So fix it to use \u0027long\u0027.\n\nReported-by: Olaf Bonorden \u003cbono@onlinehome.de\u003e\nReported-by: Daniel Büse \u003cdbuese@gmx.de\u003e\nReported-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e40051d134f7ee95c8c1f7a3471e84eafc9ab326", "tree": "88eb44e49a75721ae926665a2c42f08badac9d07", "parents": [ "42099d7a3941d4aaf853caac92b3ae76149fc6e7", "2cc6d2bf3d6195fabcf0febc192c01f99519a8f3" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 27 01:03:03 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 27 01:03:03 2010 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/net/qlcnic/qlcnic_init.c\n\tnet/ipv4/ip_output.c\n" }, { "commit": "a4d258036ed9b2a1811c3670c6099203a0f284a0", "tree": "58412022664aef745a99d254d20cf1a66aabda44", "parents": [ "9828e6e6e3f19efcb476c567b9999891d051f52f" ], "author": { "name": "Tom Marshall", "email": "tdm.code@gmail.com", "time": "Mon Sep 20 15:42:05 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 20 15:42:05 2010 -0700" }, "message": "tcp: Fix race in tcp_poll\n\nIf a RST comes in immediately after checking sk-\u003esk_err, tcp_poll will\nreturn POLLIN but not POLLOUT. Fix this by checking sk-\u003esk_err at the end\nof tcp_poll. Additionally, ensure the correct order of operations on SMP\nmachines with memory barriers.\n\nSigned-off-by: Tom Marshall \u003ctdm.code@gmail.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e548833df83c3554229eff0672900bfe958b45fd", "tree": "85efc4a76dc356593d6d394776aeb845dc580fb6", "parents": [ "cbd9da7be869f676afc204e1a664163778c770bd", "053d8f6622701f849fda2ca2c9ae596c13599ba9" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 09 22:27:33 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Sep 09 22:27:33 2010 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tnet/mac80211/main.c\n" }, { "commit": "dca43c75e7e545694a9dd6288553f55c53e2a3a3", "tree": "4df6b0b295ecd571fa95004b486d9af1636d6a30", "parents": [ "409456b10f87b28303643fec37543103f9ada00c" ], "author": { "name": "Jerry Chu", "email": "hkchu@google.com", "time": "Fri Aug 27 19:13:28 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Aug 30 13:23:33 2010 -0700" }, "message": "tcp: Add TCP_USER_TIMEOUT socket option.\n\nThis patch provides a \"user timeout\" support as described in RFC793. The\nsocket option is also needed for the the local half of RFC5482 \"TCP User\nTimeout Option\".\n\nTCP_USER_TIMEOUT is a TCP level socket option that takes an unsigned int,\nwhen \u003e 0, to specify the maximum amount of time in ms that transmitted\ndata may remain unacknowledged before TCP will forcefully close the\ncorresponding connection and return ETIMEDOUT to the application. If\n0 is given, TCP will continue to use the system default.\n\nIncreasing the user timeouts allows a TCP connection to survive extended\nperiods without end-to-end connectivity. Decreasing the user timeouts\nallows applications to \"fail fast\" if so desired. Otherwise it may take\nupto 20 minutes with the current system defaults in a normal WAN\nenvironment.\n\nThe socket option can be made during any state of a TCP connection, but\nis only effective during the synchronized states of a connection\n(ESTABLISHED, FIN-WAIT-1, FIN-WAIT-2, CLOSE-WAIT, CLOSING, or LAST-ACK).\nMoreover, when used with the TCP keepalive (SO_KEEPALIVE) option,\nTCP_USER_TIMEOUT will overtake keepalive to determine when to close a\nconnection due to keepalive failure.\n\nThe option does not change in anyway when TCP retransmits a packet, nor\nwhen a keepalive probe will be sent.\n\nThis option, like many others, will be inherited by an acceptor from its\nlistener.\n\nSigned-off-by: H.K. Jerry Chu \u003chkchu@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d84ba638e4ba3c40023ff997aa5e8d3ed002af36", "tree": "45bff01d15301cb6fc2f27d53fdfcbbf1f2355b1", "parents": [ "c5ed63d66f24fd4f7089b5a6e087b0ce7202aa8e" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Tue Aug 24 16:05:48 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Aug 25 23:02:48 2010 -0700" }, "message": "tcp: select(writefds) don\u0027t hang up when a peer close connection\n\nThis issue come from ruby language community. Below test program\nhang up when only run on Linux.\n\n\t% uname -mrsv\n\tLinux 2.6.26-2-486 #1 Sat Dec 26 08:37:39 UTC 2009 i686\n\t% ruby -rsocket -ve \u0027\n\tBasicSocket.do_not_reverse_lookup \u003d true\n\tserv \u003d TCPServer.open(\"127.0.0.1\", 0)\n\ts1 \u003d TCPSocket.open(\"127.0.0.1\", serv.addr[1])\n\ts2 \u003d serv.accept\n\ts2.close\n\ts1.write(\"a\") rescue p $!\n\ts1.write(\"a\") rescue p $!\n\tThread.new {\n\t s1.write(\"a\")\n\t}.join\u0027\n\truby 1.9.3dev (2010-07-06 trunk 28554) [i686-linux]\n\t#\u003cErrno::EPIPE: Broken pipe\u003e\n\t[Hang Here]\n\nFreeBSD, Solaris, Mac doesn\u0027t. because Ruby\u0027s write() method call\nselect() internally. and tcp_poll has a bug.\n\nSUS defined \u0027ready for writing\u0027 of select() as following.\n\n| A descriptor shall be considered ready for writing when a call to an output\n| function with O_NONBLOCK clear would not block, whether or not the function\n| would transfer data successfully.\n\nThat said, EPIPE situation is clearly one of \u0027ready for writing\u0027.\n\nWe don\u0027t have read-side issue because tcp_poll() already has read side\nshutdown care.\n\n| if (sk-\u003esk_shutdown \u0026 RCV_SHUTDOWN)\n| mask |\u003d POLLIN | POLLRDNORM | POLLRDHUP;\n\nSo, Let\u0027s insert same logic in write side.\n\n- reference url\n http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/31065\n http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/31068\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c5ed63d66f24fd4f7089b5a6e087b0ce7202aa8e", "tree": "140df25fef28c94f9ec1e5c29b4a8d0876da2fe8", "parents": [ "ad1af0fedba14f82b240a03fe20eb9b2fdbd0357" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Wed Aug 25 23:02:17 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Aug 25 23:02:17 2010 -0700" }, "message": "tcp: fix three tcp sysctls tuning\n\nAs discovered by Anton Blanchard, current code to autotune \ntcp_death_row.sysctl_max_tw_buckets, sysctl_tcp_max_orphans and\nsysctl_max_syn_backlog makes little sense.\n\nThe bigger a page is, the less tcp_max_orphans is : 4096 on a 512GB\nmachine in Anton\u0027s case.\n\n(tcp_hashinfo.bhash_size * sizeof(struct inet_bind_hashbucket))\nis much bigger if spinlock debugging is on. Its wrong to select bigger\nlimits in this case (where kernel structures are also bigger)\n\nbhash_size max is 65536, and we get this value even for small machines. \n\nA better ground is to use size of ehash table, this also makes code\nshorter and more obvious.\n\nBased on a patch from Anton, and another from David.\n\nReported-and-tested-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ad1af0fedba14f82b240a03fe20eb9b2fdbd0357", "tree": "4d53aa8bc2d9df782aa792e52670ab55c7a44d5b", "parents": [ "b2bc85631e72485b984bcd202a104591874babba" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Aug 25 02:27:49 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Aug 25 02:27:49 2010 -0700" }, "message": "tcp: Combat per-cpu skew in orphan tests.\n\nAs reported by Anton Blanchard when we use\npercpu_counter_read_positive() to make our orphan socket limit checks,\nthe check can be off by up to num_cpus_online() * batch (which is 32\nby default) which on a 128 cpu machine can be as large as the default\norphan limit itself.\n\nFix this by doing the full expensive sum check if the optimized check\ntriggers.\n\nReported-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\n" }, { "commit": "00dad5e479531e379aba7358600cc768725d4f1f", "tree": "652199eacee32c2cf317f5a54da8265973d29978", "parents": [ "c477d0447db08068a497e7beb892b2b2a7bff64b", "3a3dfb062c2e086c202d34f09ce29634515ad256" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Aug 02 22:22:46 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Aug 02 22:22:46 2010 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/net/e1000e/hw.h\n\tnet/bridge/br_device.c\n\tnet/bridge/br_input.c\n" }, { "commit": "3c0fef0b7d36e5f8d3ea3731a8228102274e3c23", "tree": "71e201a342dcd11946891da0d900e7fd81a23217", "parents": [ "99e6d06521f2a522ff5aaa812552f68220507c67" ], "author": { "name": "Josh Hunt", "email": "johunt@akamai.com", "time": "Fri Jul 30 13:49:35 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Aug 02 17:25:06 2010 -0700" }, "message": "net: Add getsockopt support for TCP thin-streams\n\nInitial TCP thin-stream commit did not add getsockopt support for the new\nsocket options: TCP_THIN_LINEAR_TIMEOUTS and TCP_THIN_DUPACK. This adds support\nfor them.\n\nSigned-off-by: Josh Hunt \u003cjohunt@akamai.com\u003e\nTested-by: Andreas Petlund \u003capetlund@simula.no\u003e\nAcked-by: Andreas Petlund \u003capetlund@simula.no\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a3bdb549e30e7a263f7a589747c40e9c50110315", "tree": "ac14f82ebe454fdbc00c3f1298a293e03e1f1592", "parents": [ "eeaf61d8891f9c9ed12c1a667e72bf83f0857954" ], "author": { "name": "Dmitry Popov", "email": "dp@highloadlab.com", "time": "Thu Jul 29 01:59:36 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jul 30 23:04:07 2010 -0700" }, "message": "tcp: cookie transactions setsockopt memory leak\n\nThere is a bug in do_tcp_setsockopt(net/ipv4/tcp.c),\nTCP_COOKIE_TRANSACTIONS case.\nIn some cases (when tp-\u003ecookie_values \u003d\u003d NULL) new tcp_cookie_values\nstructure can be allocated (at cvp), but not bound to\ntp-\u003ecookie_values. So a memory leak occurs.\n\nSigned-off-by: Dmitry Popov \u003cdp@highloadlab.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "11fe883936980fe242869d671092a466cf1db3e3", "tree": "14ff24e81eb4326e94eb5aa6432a1dd55cef5ece", "parents": [ "70d4bf6d467a330ccc947df9b2608e329d9e7708", "573201f36fd9c7c6d5218cdcd9948cee700b277d" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Jul 20 18:25:24 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Jul 20 18:25:24 2010 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/vhost/net.c\n\tnet/bridge/br_device.c\n\nFix merge conflict in drivers/vhost/net.c with guidance from\nStephen Rothwell.\n\nRevert the effects of net-2.6 commit 573201f36fd9c7c6d5218cdcd9948cee700b277d\nsince net-next-2.6 has fixes that make bridge netpoll work properly thus\nwe don\u0027t need it disabled.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3a047bf87b1b6f69c62ab9fb28072c639cb7e2fa", "tree": "b134b9942d2b378b1afd81b07ffdfb948cd1439e", "parents": [ "d809ec895505e6f35fb1965f0946381ab4eaa474" ], "author": { "name": "Changli Gao", "email": "xiaosuo@gmail.com", "time": "Mon Jul 12 21:00:12 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Jul 14 14:45:15 2010 -0700" }, "message": "rfs: call sock_rps_record_flow() in tcp_splice_read()\n\nrfs: call sock_rps_record_flow() in tcp_splice_read()\n\ncall sock_rps_record_flow() in tcp_splice_read(), so the applications using\nsplice(2) or sendfile(2) can utilize RFS.\n\nSigned-off-by: Changli Gao \u003cxiaosuo@gmail.com\u003e\n----\n net/ipv4/tcp.c | 1 +\n 1 file changed, 1 insertion(+)\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7ba42910073f8432934d61a6c08b1023c408fb62", "tree": "4f74648133eaf6cbea26d59c43d34e7153648f5d", "parents": [ "53d3176b282cc105493babb0fef36c8b873f6201" ], "author": { "name": "Changli Gao", "email": "xiaosuo@gmail.com", "time": "Sat Jul 10 20:41:55 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jul 12 20:21:46 2010 -0700" }, "message": "inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and inet_sendpage()\n\na new boolean flag no_autobind is added to structure proto to avoid the autobind\ncalls when the protocol is TCP. Then sock_rps_record_flow() is called int the\nTCP\u0027s sendmsg() and sendpage() pathes.\n\nSigned-off-by: Changli Gao \u003cxiaosuo@gmail.com\u003e\n----\n include/net/inet_common.h | 4 ++++\n include/net/sock.h | 1 +\n include/net/tcp.h | 8 ++++----\n net/ipv4/af_inet.c | 15 +++++++++------\n net/ipv4/tcp.c | 11 +++++------\n net/ipv4/tcp_ipv4.c | 3 +++\n net/ipv6/af_inet6.c | 8 ++++----\n net/ipv6/tcp_ipv6.c | 3 +++\n 8 files changed, 33 insertions(+), 20 deletions(-)\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4bc2f18ba4f22a90ab593c0a580fc9a19c4777b6", "tree": "bb39f427bd468d0e34a73ebc788db726406fa261", "parents": [ "ba80a2522899ea71a5b201bae60bdfd99126af95" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Fri Jul 09 21:22:10 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jul 12 12:57:54 2010 -0700" }, "message": "net/ipv4: EXPORT_SYMBOL cleanups\n\nCodingStyle cleanups\n\nEXPORT_SYMBOL should immediately follow the symbol declaration.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7a9b2d59507d85569b8a456688ef40cf2ac73e48", "tree": "436533da6f680a0b4b6860e0219a2e594ac16735", "parents": [ "b6b3ecc71a0664d44ed8d087d583aee98fbf492a" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Thu Jun 24 00:52:37 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jun 28 23:24:29 2010 -0700" }, "message": "net: use this_cpu_ptr()\n\nuse this_cpu_ptr(p) instead of per_cpu_ptr(p, smp_processor_id())\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "565b7b2d2e632b5792879c0c9cccdd9eecd31195", "tree": "f5bbe0e7a2719f6076a567cb3a68ff47f0c4bcb9", "parents": [ "7a938f80264f2cbfb0c0841b450eab42a8093281" ], "author": { "name": "Konstantin Khorenko", "email": "khorenko@openvz.org", "time": "Thu Jun 24 21:54:58 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jun 24 21:54:58 2010 -0700" }, "message": "tcp: do not send reset to already closed sockets\n\ni\u0027ve found that tcp_close() can be called for an already closed\nsocket, but still sends reset in this case (tcp_send_active_reset())\nwhich seems to be incorrect. Moreover, a packet with reset is sent\nwith different source port as original port number has been already\ncleared on socket. Besides that incrementing stat counter for\nLINUX_MIB_TCPABORTONCLOSE also does not look correct in this case.\n\nInitially this issue was found on 2.6.18-x RHEL5 kernel, but the same\nseems to be true for the current mainstream kernel (checked on\n2.6.35-rc3). Please, correct me if i missed something.\n\nHow that happens:\n\n1) the server receives a packet for socket in TCP_CLOSE_WAIT state\n that triggers a tcp_reset():\n\nCall Trace:\n \u003cIRQ\u003e [\u003cffffffff8025b9b9\u003e] tcp_reset+0x12f/0x1e8\n [\u003cffffffff80046125\u003e] tcp_rcv_state_process+0x1c0/0xa08\n [\u003cffffffff8003eb22\u003e] tcp_v4_do_rcv+0x310/0x37a\n [\u003cffffffff80028bea\u003e] tcp_v4_rcv+0x74d/0xb43\n [\u003cffffffff8024ef4c\u003e] ip_local_deliver_finish+0x0/0x259\n [\u003cffffffff80037131\u003e] ip_local_deliver+0x200/0x2f4\n [\u003cffffffff8003843c\u003e] ip_rcv+0x64c/0x69f\n [\u003cffffffff80021d89\u003e] netif_receive_skb+0x4c4/0x4fa\n [\u003cffffffff80032eca\u003e] process_backlog+0x90/0xec\n [\u003cffffffff8000cc50\u003e] net_rx_action+0xbb/0x1f1\n [\u003cffffffff80012d3a\u003e] __do_softirq+0xf5/0x1ce\n [\u003cffffffff8001147a\u003e] handle_IRQ_event+0x56/0xb0\n [\u003cffffffff8006334c\u003e] call_softirq+0x1c/0x28\n [\u003cffffffff80070476\u003e] do_softirq+0x2c/0x85\n [\u003cffffffff80070441\u003e] do_IRQ+0x149/0x152\n [\u003cffffffff80062665\u003e] ret_from_intr+0x0/0xa\n \u003cEOI\u003e [\u003cffffffff80008a2e\u003e] __handle_mm_fault+0x6cd/0x1303\n [\u003cffffffff80008903\u003e] __handle_mm_fault+0x5a2/0x1303\n [\u003cffffffff80033a9d\u003e] cache_free_debugcheck+0x21f/0x22e\n [\u003cffffffff8006a263\u003e] do_page_fault+0x49a/0x7dc\n [\u003cffffffff80066487\u003e] thread_return+0x89/0x174\n [\u003cffffffff800c5aee\u003e] audit_syscall_exit+0x341/0x35c\n [\u003cffffffff80062e39\u003e] error_exit+0x0/0x84\n\ntcp_rcv_state_process()\n... // (sk_state \u003d\u003d TCP_CLOSE_WAIT here)\n...\n /* step 2: check RST bit */\n if(th-\u003erst) {\n tcp_reset(sk);\n goto discard;\n }\n...\n---------------------------------\ntcp_rcv_state_process\n tcp_reset\n tcp_done\n tcp_set_state(sk, TCP_CLOSE);\n inet_put_port\n __inet_put_port\n inet_sk(sk)-\u003enum \u003d 0;\n\n sk-\u003esk_shutdown \u003d SHUTDOWN_MASK;\n\n2) After that the process (socket owner) tries to write something to\n that socket and \"inet_autobind\" sets a _new_ (which differs from\n the original!) port number for the socket:\n\n Call Trace:\n [\u003cffffffff80255a12\u003e] inet_bind_hash+0x33/0x5f\n [\u003cffffffff80257180\u003e] inet_csk_get_port+0x216/0x268\n [\u003cffffffff8026bcc9\u003e] inet_autobind+0x22/0x8f\n [\u003cffffffff80049140\u003e] inet_sendmsg+0x27/0x57\n [\u003cffffffff8003a9d9\u003e] do_sock_write+0xae/0xea\n [\u003cffffffff80226ac7\u003e] sock_writev+0xdc/0xf6\n [\u003cffffffff800680c7\u003e] _spin_lock_irqsave+0x9/0xe\n [\u003cffffffff8001fb49\u003e] __pollwait+0x0/0xdd\n [\u003cffffffff8008d533\u003e] default_wake_function+0x0/0xe\n [\u003cffffffff800a4f10\u003e] autoremove_wake_function+0x0/0x2e\n [\u003cffffffff800f0b49\u003e] do_readv_writev+0x163/0x274\n [\u003cffffffff80066538\u003e] thread_return+0x13a/0x174\n [\u003cffffffff800145d8\u003e] tcp_poll+0x0/0x1c9\n [\u003cffffffff800c56d3\u003e] audit_syscall_entry+0x180/0x1b3\n [\u003cffffffff800f0dd0\u003e] sys_writev+0x49/0xe4\n [\u003cffffffff800622dd\u003e] tracesys+0xd5/0xe0\n\n3) sendmsg fails at last with -EPIPE (\u003d\u003e \u0027write\u0027 returns -EPIPE in userspace):\n\nF: tcp_sendmsg1 -EPIPE: sk\u003dffff81000bda00d0, sport\u003d49847, old_state\u003d7, new_state\u003d7, sk_err\u003d0, sk_shutdown\u003d3\n\nCall Trace:\n [\u003cffffffff80027557\u003e] tcp_sendmsg+0xcb/0xe87\n [\u003cffffffff80033300\u003e] release_sock+0x10/0xae\n [\u003cffffffff8016f20f\u003e] vgacon_cursor+0x0/0x1a7\n [\u003cffffffff8026bd32\u003e] inet_autobind+0x8b/0x8f\n [\u003cffffffff8003a9d9\u003e] do_sock_write+0xae/0xea\n [\u003cffffffff80226ac7\u003e] sock_writev+0xdc/0xf6\n [\u003cffffffff800680c7\u003e] _spin_lock_irqsave+0x9/0xe\n [\u003cffffffff8001fb49\u003e] __pollwait+0x0/0xdd\n [\u003cffffffff8008d533\u003e] default_wake_function+0x0/0xe\n [\u003cffffffff800a4f10\u003e] autoremove_wake_function+0x0/0x2e\n [\u003cffffffff800f0b49\u003e] do_readv_writev+0x163/0x274\n [\u003cffffffff80066538\u003e] thread_return+0x13a/0x174\n [\u003cffffffff800145d8\u003e] tcp_poll+0x0/0x1c9\n [\u003cffffffff800c56d3\u003e] audit_syscall_entry+0x180/0x1b3\n [\u003cffffffff800f0dd0\u003e] sys_writev+0x49/0xe4\n [\u003cffffffff800622dd\u003e] tracesys+0xd5/0xe0\n\ntcp_sendmsg()\n...\n /* Wait for a connection to finish. */\n if ((1 \u003c\u003c sk-\u003esk_state) \u0026 ~(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)) {\n int old_state \u003d sk-\u003esk_state;\n if ((err \u003d sk_stream_wait_connect(sk, \u0026timeo)) !\u003d 0) {\nif (f_d \u0026\u0026 (err \u003d\u003d -EPIPE)) {\n printk(\"F: tcp_sendmsg1 -EPIPE: sk\u003d%p, sport\u003d%u, old_state\u003d%d, new_state\u003d%d, \"\n \"sk_err\u003d%d, sk_shutdown\u003d%d\\n\",\n sk, ntohs(inet_sk(sk)-\u003esport), old_state, sk-\u003esk_state,\n sk-\u003esk_err, sk-\u003esk_shutdown);\n dump_stack();\n}\n goto out_err;\n }\n }\n...\n\n4) Then the process (socket owner) understands that it\u0027s time to close\n that socket and does that (and thus triggers sending reset packet):\n\nCall Trace:\n...\n [\u003cffffffff80032077\u003e] dev_queue_xmit+0x343/0x3d6\n [\u003cffffffff80034698\u003e] ip_output+0x351/0x384\n [\u003cffffffff80251ae9\u003e] dst_output+0x0/0xe\n [\u003cffffffff80036ec6\u003e] ip_queue_xmit+0x567/0x5d2\n [\u003cffffffff80095700\u003e] vprintk+0x21/0x33\n [\u003cffffffff800070f0\u003e] check_poison_obj+0x2e/0x206\n [\u003cffffffff80013587\u003e] poison_obj+0x36/0x45\n [\u003cffffffff8025dea6\u003e] tcp_send_active_reset+0x15/0x14d\n [\u003cffffffff80023481\u003e] dbg_redzone1+0x1c/0x25\n [\u003cffffffff8025dea6\u003e] tcp_send_active_reset+0x15/0x14d\n [\u003cffffffff8000ca94\u003e] cache_alloc_debugcheck_after+0x189/0x1c8\n [\u003cffffffff80023405\u003e] tcp_transmit_skb+0x764/0x786\n [\u003cffffffff8025df8a\u003e] tcp_send_active_reset+0xf9/0x14d\n [\u003cffffffff80258ff1\u003e] tcp_close+0x39a/0x960\n [\u003cffffffff8026be12\u003e] inet_release+0x69/0x80\n [\u003cffffffff80059b31\u003e] sock_release+0x4f/0xcf\n [\u003cffffffff80059d4c\u003e] sock_close+0x2c/0x30\n [\u003cffffffff800133c9\u003e] __fput+0xac/0x197\n [\u003cffffffff800252bc\u003e] filp_close+0x59/0x61\n [\u003cffffffff8001eff6\u003e] sys_close+0x85/0xc7\n [\u003cffffffff800622dd\u003e] tracesys+0xd5/0xe0\n\nSo, in brief:\n\n* a received packet for socket in TCP_CLOSE_WAIT state triggers\n tcp_reset() which clears inet_sk(sk)-\u003enum and put socket into\n TCP_CLOSE state\n\n* an attempt to write to that socket forces inet_autobind() to get a\n new port (but the write itself fails with -EPIPE)\n\n* tcp_close() called for socket in TCP_CLOSE state sends an active\n reset via socket with newly allocated port\n\nThis adds an additional check in tcp_close() for already closed\nsockets. We do not want to send anything to closed sockets.\n\nSigned-off-by: Konstantin Khorenko \u003ckhorenko@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a3433f35a55f7604742cae620c6dc6edfc70db6a", "tree": "4ce6d169e402c03a48319a9db4055cc31df00d42", "parents": [ "f350a0a87374418635689471606454abc7beaa3a" ], "author": { "name": "Changli Gao", "email": "xiaosuo@gmail.com", "time": "Sat Jun 12 14:01:43 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Jun 15 11:56:19 2010 -0700" }, "message": "tcp: unify tcp flag macros\n\nunify tcp flag macros: TCPHDR_FIN, TCPHDR_SYN, TCPHDR_RST, TCPHDR_PSH,\nTCPHDR_ACK, TCPHDR_URG, TCPHDR_ECE and TCPHDR_CWR. TCBCB_FLAG_* are replaced\nwith the corresponding TCPHDR_*.\n\nSigned-off-by: Changli Gao \u003cxiaosuo@gmail.com\u003e\n----\n include/net/tcp.h | 24 ++++++-------\n net/ipv4/tcp.c | 8 ++--\n net/ipv4/tcp_input.c | 2 -\n net/ipv4/tcp_output.c | 59 ++++++++++++++++-----------------\n net/netfilter/nf_conntrack_proto_tcp.c | 32 ++++++-----------\n net/netfilter/xt_TCPMSS.c | 4 --\n 6 files changed, 58 insertions(+), 71 deletions(-)\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d7fd1b5747fff3bde92777bcaf705d98ae6f8b6f", "tree": "8cdac5dba6b350fd759eadd324e706639180a162", "parents": [ "15e83ed78864d0625e87a85f09b297c0919a4797" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Mon May 17 20:40:51 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon May 31 00:24:02 2010 -0700" }, "message": "tcp: tcp_md5_hash_skb_data() frag_list handling\n\ntcp_md5_hash_skb_data() should handle skb-\u003efrag_list, and eventually\nrecurse.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ccbd6a5a4f76e821ed36f69fdaf59817c3a7f18e", "tree": "9fd681809e588913cc312f63ae0e701f18599afc", "parents": [ "935e2a26b85003c0bd52b6c92712c2f77a5f9d33" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Fri May 14 10:58:26 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon May 17 17:44:35 2010 -0700" }, "message": "net: Remove unnecessary semicolons after switch statements\n\nAlso added an explicit break; to avoid\na fallthrough in net/ipv4/tcp_input.c\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6811d58fc148c393f80a9f5a9db49d7e75cdc546", "tree": "c25d5b0e49ec848943d35f819e748d157ccb492e", "parents": [ "c4949f074332a64baeb2ead6ab9319ca37642f96", "c02db8c6290bb992442fec1407643c94cc414375" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun May 16 22:26:58 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun May 16 22:26:58 2010 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tinclude/linux/if_link.h\n" }, { "commit": "35790c0421121364883a167bab8a2e37e1f67f78", "tree": "a3e032a9ba6bb77337176bef407d7408d6a10a4d", "parents": [ "d77f873fdd21912803836da78f627d2efd267082" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Sun May 16 00:34:04 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun May 16 00:34:04 2010 -0700" }, "message": "tcp: fix MD5 (RFC2385) support\n\nTCP MD5 support uses percpu data for temporary storage. It currently\ndisables preemption so that same storage cannot be reclaimed by another\nthread on same cpu.\n\nWe also have to make sure a softirq handler wont try to use also same\ncontext. Various bug reports demonstrated corruptions.\n\nFix is to disable preemption and BH.\n\nReported-by: Bhaskar Dutta \u003cbhaskie@gmail.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6c37e5de456987f5bc80879afde05aa120784095", "tree": "1459ee69ef3c7172a82d07a7b8aea00836eedda5", "parents": [ "dcdca2c49bb6328bbc7cd8d73434c308b5dd0df2" ], "author": { "name": "Flavio Leitner", "email": "fleitner@redhat.com", "time": "Mon Apr 26 18:33:27 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 27 12:53:25 2010 -0700" }, "message": "TCP: avoid to send keepalive probes if receiving data\n\nRFC 1122 says the following:\n...\n Keep-alive packets MUST only be sent when no data or\n acknowledgement packets have been received for the\n connection within an interval.\n...\n\nThe acknowledgement packet is reseting the keepalive\ntimer but the data packet isn\u0027t. This patch fixes it by\nchecking the timestamp of the last received data packet\ntoo when the keepalive timer expires.\n\nSigned-off-by: Flavio Leitner \u003cfleitner@redhat.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nAcked-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0eae88f31ca2b88911ce843452054139e028771f", "tree": "90f50197d8e1e50e3621e9c525bd4ca2634fe368", "parents": [ "cb903bf4ee2d6e53210e2174d363e10698112042" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Tue Apr 20 19:06:52 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 20 19:06:52 2010 -0700" }, "message": "net: Fix various endianness glitches\n\nSparse can help us find endianness bugs, but we need to make some\ncleanups to be able to more easily spot real bugs.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "aa395145165cb06a0d0885221bbe0ce4a564391d", "tree": "118b0403621f10db8dc3dbf12079f9af5b19e05d", "parents": [ "ab9304717f7624c41927f442e6b6d418b2d8b3e4" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Tue Apr 20 13:03:51 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Apr 20 16:37:13 2010 -0700" }, "message": "net: sk_sleep() helper\n\nDefine a new function to return the waitqueue of a \"struct sock\".\n\nstatic inline wait_queue_head_t *sk_sleep(struct sock *sk)\n{\n\treturn sk-\u003esk_sleep;\n}\n\nChange all read occurrences of sk_sleep by a call to this function.\n\nNeeded for a future RCU conversion. sk_sleep wont be a field directly\navailable.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cb4361c1dc29cd870f664c004b1817106fbce0fa", "tree": "c31533210d738c9074b21b8f31fe0399b39545f0", "parents": [ "309361e09ca9e9670dc8664e5d14125bf82078af", "fb9e2d887243499b8d28efcf80821c4f6a092395" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 06 08:34:06 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 06 08:34:06 2010 -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: (37 commits)\n smc91c92_cs: fix the problem of \"Unable to find hardware address\"\n r8169: clean up my printk uglyness\n net: Hook up cxgb4 to Kconfig and Makefile\n cxgb4: Add main driver file and driver Makefile\n cxgb4: Add remaining driver headers and L2T management\n cxgb4: Add packet queues and packet DMA code\n cxgb4: Add HW and FW support code\n cxgb4: Add register, message, and FW definitions\n netlabel: Fix several rcu_dereference() calls used without RCU read locks\n bonding: fix potential deadlock in bond_uninit()\n net: check the length of the socket address passed to connect(2)\n stmmac: add documentation for the driver.\n stmmac: fix kconfig for crc32 build error\n be2net: fix bug in vlan rx path for big endian architecture\n be2net: fix flashing on big endian architectures\n be2net: fix a bug in flashing the redboot section\n bonding: bond_xmit_roundrobin() fix\n drivers/net: Add missing unlock\n net: gianfar - align BD ring size console messages\n net: gianfar - initialize per-queue statistics\n ...\n" }, { "commit": "baff42ab1494528907bf4d5870359e31711746ae", "tree": "82cfffb254ea1f5b95701d328375228f7ab343e6", "parents": [ "c0cd884af045338476b8e69a61fceb3f34ff22f1" ], "author": { "name": "Steven J. Magnani", "email": "steve@digidescorp.com", "time": "Tue Mar 30 13:56:01 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Mar 30 13:56:01 2010 -0700" }, "message": "net: Fix oops from tcp_collapse() when using splice()\n\ntcp_read_sock() can have a eat skbs without immediately advancing copied_seq.\nThis can cause a panic in tcp_collapse() if it is called as a result\nof the recv_actor dropping the socket lock.\n\nA userspace program that splices data from a socket to either another\nsocket or to a file can trigger this bug.\n\nSigned-off-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5a0e3ad6af8660be21ca98a971cd00f331318c05", "tree": "5bfb7be11a03176a87296a43ac6647975c00a1d1", "parents": [ "ed391f4ebf8f701d3566423ce8f17e614cde9806" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Wed Mar 24 17:04:11 2010 +0900" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Tue Mar 30 22:02:32 2010 +0900" }, "message": "include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h\n\npercpu.h is included by sched.h and module.h and thus ends up being\nincluded when building most .c files. percpu.h includes slab.h which\nin turn includes gfp.h making everything defined by the two files\nuniversally available and complicating inclusion dependencies.\n\npercpu.h -\u003e slab.h dependency is about to be removed. Prepare for\nthis change by updating users of gfp and slab facilities include those\nheaders directly instead of assuming availability. As this conversion\nneeds to touch large number of source files, the following script is\nused as the basis of conversion.\n\n http://userweb.kernel.org/~tj/misc/slabh-sweep.py\n\nThe script does the followings.\n\n* Scan files for gfp and slab usages and update includes such that\n only the necessary includes are there. ie. if only gfp is used,\n gfp.h, if slab is used, slab.h.\n\n* When the script inserts a new include, it looks at the include\n blocks and try to put the new include such that its order conforms\n to its surrounding. It\u0027s put in the include block which contains\n core kernel includes, in the same order that the rest are ordered -\n alphabetical, Christmas tree, rev-Xmas-tree or at the end if there\n doesn\u0027t seem to be any matching order.\n\n* If the script can\u0027t find a place to put a new include (mostly\n because the file doesn\u0027t have fitting include block), it prints out\n an error message indicating which .h file needs to be added to the\n file.\n\nThe conversion was done in the following steps.\n\n1. The initial automatic conversion of all .c files updated slightly\n over 4000 files, deleting around 700 includes and adding ~480 gfp.h\n and ~3000 slab.h inclusions. The script emitted errors for ~400\n files.\n\n2. Each error was manually checked. Some didn\u0027t need the inclusion,\n some needed manual addition while adding it to implementation .h or\n embedding .c file was more appropriate for others. This step added\n inclusions to around 150 files.\n\n3. The script was run again and the output was compared to the edits\n from #2 to make sure no file was left behind.\n\n4. Several build tests were done and a couple of problems were fixed.\n e.g. lib/decompress_*.c used malloc/free() wrappers around slab\n APIs requiring slab.h to be added manually.\n\n5. The script was run on all .h files but without automatically\n editing them as sprinkling gfp.h and slab.h inclusions around .h\n files could easily lead to inclusion dependency hell. Most gfp.h\n inclusion directives were ignored as stuff from gfp.h was usually\n wildly available and often used in preprocessor macros. Each\n slab.h inclusion directive was examined and added manually as\n necessary.\n\n6. percpu.h was updated not to include slab.h.\n\n7. Build test were done on the following configurations and failures\n were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my\n distributed build env didn\u0027t work with gcov compiles) and a few\n more options had to be turned off depending on archs to make things\n build (like ipr on powerpc/64 which failed due to missing writeq).\n\n * x86 and x86_64 UP and SMP allmodconfig and a custom test config.\n * powerpc and powerpc64 SMP allmodconfig\n * sparc and sparc64 SMP allmodconfig\n * ia64 SMP allmodconfig\n * s390 SMP allmodconfig\n * alpha SMP allmodconfig\n * um on x86_64 SMP allmodconfig\n\n8. percpu.h modifications were reverted so that it could be applied as\n a separate patch and serve as bisection point.\n\nGiven the fact that I had only a couple of failures from tests on step\n6, I\u0027m fairly confident about the coverage of this conversion patch.\nIf there is a breakage, it\u0027s likely to be something in one of the arch\nheaders which should be easily discoverable easily on most builds of\nthe specific arch.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nGuess-its-ok-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\n" }, { "commit": "73852e8151b7d7a529fbe019ab6d2d0c02d8f3f2", "tree": "724151d30c232e38ea947705986cfe60a3359241", "parents": [ "f5d410f2ea7ba340f11815a56e05b9fa9421c421" ], "author": { "name": "Steven J. Magnani", "email": "steve@digidescorp.com", "time": "Tue Mar 16 05:22:44 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 20 14:29:02 2010 -0700" }, "message": "NET_DMA: free skbs periodically\n\nUnder NET_DMA, data transfer can grind to a halt when userland issues a\nlarge read on a socket with a high RCVLOWAT (i.e., 512 KB for both).\nThis appears to be because the NET_DMA design queues up lots of memcpy\noperations, but doesn\u0027t issue or wait for them (and thus free the\nassociated skbs) until it is time for tcp_recvmesg() to return.\nThe socket hangs when its TCP window goes to zero before enough data is\navailable to satisfy the read.\n\nPeriodically issue asynchronous memcpy operations, and free skbs for ones\nthat have completed, to prevent sockets from going into zero-window mode.\n\nSigned-off-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b634f87522dff87712df8bda2a6c9061954d552a", "tree": "fad85992b6d2566ba635434339da7567966ce1a6", "parents": [ "22001a13d09d82772e831dcdac0553994a4bac5d" ], "author": { "name": "Alexandra Kossovsky", "email": "Alexandra.Kossovsky@oktetlabs.ru", "time": "Thu Mar 18 20:29:24 2010 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 18 20:29:24 2010 -0700" }, "message": "tcp: Fix OOB POLLIN avoidance.\n\nFrom: Alexandra.Kossovsky@oktetlabs.ru\n\nFixes kernel bugzilla #15541\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7e38017557bc0b87434d184f8804cadb102bb903", "tree": "30c4a0793bd28f6731ade58786c06f9cf33115df", "parents": [ "36e31b0af58728071e8023cf8e20c5166b700717" ], "author": { "name": "Andreas Petlund", "email": "apetlund@simula.no", "time": "Thu Feb 18 04:48:19 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 18 15:43:09 2010 -0800" }, "message": "net: TCP thin dupack\n\nThis patch enables fast retransmissions after one dupACK for\nTCP if the stream is identified as thin. This will reduce\nlatencies for thin streams that are not able to trigger fast\nretransmissions due to high packet interarrival time. This\nmechanism is only active if enabled by iocontrol or syscontrol\nand the stream is identified as thin.\n\nSigned-off-by: Andreas Petlund \u003capetlund@simula.no\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "36e31b0af58728071e8023cf8e20c5166b700717", "tree": "8b4d251bf78965ac7501bea9011786b8255a3312", "parents": [ "5aa4b32fc86408705337e941ed716880c63d1590" ], "author": { "name": "Andreas Petlund", "email": "apetlund@simula.no", "time": "Thu Feb 18 02:47:01 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 18 15:43:08 2010 -0800" }, "message": "net: TCP thin linear timeouts\n\nThis patch will make TCP use only linear timeouts if the\nstream is thin. This will help to avoid the very high latencies\nthat thin stream suffer because of exponential backoff. This\nmechanism is only active if enabled by iocontrol or syscontrol\nand the stream is identified as thin. A maximum of 6 linear\ntimeouts is tried before exponential backoff is resumed.\n\nSigned-off-by: Andreas Petlund \u003capetlund@simula.no\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7d720c3e4f0c4fc152a6bf17e24244a3c85412d2", "tree": "36e037187ce79acb211702bea22e99c625787757", "parents": [ "2bb4646fce8d09916b351d1a62f98db7cec6fc41" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Tue Feb 16 15:20:26 2010 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 16 23:05:38 2010 -0800" }, "message": "percpu: add __percpu sparse annotations to net\n\nAdd __percpu sparse annotations to net.\n\nThese annotations are to make sparse consider percpu variables to be\nin a different address space and warn if accessed without going\nthrough percpu accessors. This patch doesn\u0027t affect normal builds.\n\nThe macro and type tricks around snmp stats make things a bit\ninteresting. DEFINE/DECLARE_SNMP_STAT() macros mark the target field\nas __percpu and SNMP_UPD_PO_STATS() macro is updated accordingly. All\nsnmp_mib_*() users which used to cast the argument to (void **) are\nupdated to cast it to (void __percpu **).\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nAcked-by: David S. Miller \u003cdavem@davemloft.net\u003e\nCc: Patrick McHardy \u003ckaber@trash.net\u003e\nCc: Arnaldo Carvalho de Melo \u003cacme@ghostprotocols.net\u003e\nCc: Vlad Yasevich \u003cvladislav.yasevich@hp.com\u003e\nCc: netdev@vger.kernel.org\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "def87cf42069a6d4fd42a2ede8f19c620a292568", "tree": "1e8e97b452dd95d389feb4f04c0f25b6a9264412", "parents": [ "afeca340c078e17ca233b3c68c3c3a70c56bfe1d" ], "author": { "name": "Krishna Kumar", "email": "krkumar2@in.ibm.com", "time": "Thu Dec 10 07:16:59 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 23 14:13:29 2009 -0800" }, "message": "tcp: Slightly optimize tcp_sendmsg\n\nSlightly optimize tcp_sendmsg since NETIF_F_SG is used many\ntimes iteratively in the loop. The only other modification is\nto change:\n\t\t\t} else if (i \u003d\u003d MAX_SKB_FRAGS ||\n\t\t\t\t (!i \u0026\u0026\n\t\t\t\t !(sk-\u003esk_route_caps \u0026 NETIF_F_SG))) {\n\tto:\n\t\t\t} else if (i \u003d\u003d MAX_SKB_FRAGS || !sg) {\n\nThe reason why this change is correct: this code (other than\nthe MAX_SKB_FRAGS case) executes only due to the else part\nof: \"if (skb_tailroom(skb) \u003e 0) {\" - i.e. there was no space\nin the skb to put the data inline. Hence SG is false is a\nsufficient condition, and there is no way a fragment can be\nadded to the skb.\n\nChangelog:\n\t- Added the above explanation for the change\n\nSigned-off-by: Krishna Kumar \u003ckrkumar2@in.ibm.com\u003e\nAcked-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "afeca340c078e17ca233b3c68c3c3a70c56bfe1d", "tree": "f7b308856bdd21f41ea61cf6b74cd28e50b0448f", "parents": [ "12d50c46dc0f7fd2e625c4befaa5fa5740a7a594" ], "author": { "name": "Krishna Kumar", "email": "krkumar2@in.ibm.com", "time": "Thu Dec 10 07:16:52 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 23 14:13:28 2009 -0800" }, "message": "tcp: Remove unrequired operations in tcp_push()\n\nRemove unrequired operations in tcp_push()\n\nChangelog:\n\tRemoved a temporary skb variable from tcp_push()\n\nSigned-off-by: Krishna Kumar \u003ckrkumar2@in.ibm.com\u003e\nAcked-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3dc789320e1b310cb505dcd94512c279abcd5e1c", "tree": "b6173953ccff6c9bfd83096c97027c2d338287d1", "parents": [ "e61444d9209821469ed0d5a9128dcf2e494533ea" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Dec 08 20:07:54 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Dec 08 20:07:54 2009 -0800" }, "message": "tcp: Remove runtime check that can never be true.\n\nGCC even warns about it, as reported by Andrew Morton:\n\nnet/ipv4/tcp.c: In function \u0027do_tcp_getsockopt\u0027:\nnet/ipv4/tcp.c:2544: warning: comparison is always false due to limited range of data type\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e56fb50f2b7958b931c8a2fc0966061b3f3c8f3a", "tree": "392f548e99f5d098286cea54fb9e18ac8c31e672", "parents": [ "435cf559f02ea3a3159eb316f97dc88bdebe9432" ], "author": { "name": "William Allen Simpson", "email": "william.allen.simpson@gmail.com", "time": "Wed Dec 02 18:19:30 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 02 22:07:25 2009 -0800" }, "message": "TCPCT part 1e: implement socket option TCP_COOKIE_TRANSACTIONS\n\nProvide per socket control of the TCP cookie option and SYN/SYNACK data.\n\nThis is a straightforward re-implementation of an earlier (year-old)\npatch that no longer applies cleanly, with permission of the original\nauthor (Adam Langley):\n\n http://thread.gmane.org/gmane.linux.network/102586\n\nThe principle difference is using a TCP option to carry the cookie nonce,\ninstead of a user configured offset in the data.\n\nAllocations have been rearranged to avoid requiring GFP_ATOMIC.\n\nRequires:\n net: TCP_MSS_DEFAULT, TCP_MSS_DESIRED\n TCPCT part 1c: sysctl_tcp_cookie_size, socket option TCP_COOKIE_TRANSACTIONS\n TCPCT part 1d: define TCP cookie option, extend existing struct\u0027s\n\nSigned-off-by: William.Allen.Simpson@gmail.com\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "da5c78c82629a167794436e4306b4cf1faddea90", "tree": "c933fdc0583b592c01885890ebf770840555d28b", "parents": [ "e6b4d11367519bc71729c09d05a126b133c755be" ], "author": { "name": "William Allen Simpson", "email": "william.allen.simpson@gmail.com", "time": "Wed Dec 02 18:12:09 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Dec 02 22:07:23 2009 -0800" }, "message": "TCPCT part 1b: generate Responder Cookie secret\n\nDefine (missing) hash message size for SHA1.\n\nDefine hashing size constants specific to TCP cookies.\n\nAdd new function: tcp_cookie_generator().\n\nMaintain global secret values for tcp_cookie_generator().\n\nThis is a significantly revised implementation of earlier (15-year-old)\nPhoturis [RFC-2522] code for the KA9Q cooperative multitasking platform.\n\nLinux RCU technique appears to be well-suited to this application, though\nneither of the circular queue items are freed.\n\nThese functions will also be used in subsequent patches that implement\nadditional features.\n\nSigned-off-by: William.Allen.Simpson@gmail.com\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ff9c38bba37937adb909cceb2a6521f2e92e17c6", "tree": "93bd6152d9fa28348be99ef1c788040cc7b7a94d", "parents": [ "65c0cfafce9575319fb6f70080fbe226e5617e3b", "b2722b1c3a893ec6021508da15b32282ec79f4da" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Dec 01 22:13:38 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Dec 01 22:13:38 2009 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tnet/mac80211/ht.c\n" }, { "commit": "1fdf475aa141a669af8db6ccc7015f0b725087de", "tree": "7298a884935feb59352566ca6119856a0e322cb5", "parents": [ "0cae200eec6330cd2c20b24279597be1da50dc93" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Mon Nov 30 12:53:30 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 30 12:53:30 2009 -0800" }, "message": "tcp: tcp_disconnect() should clear window_clamp\n\nNFS can reuse its TCP socket after calling tcp_disconnect().\n\nWe noticed window scaling was not negotiated in SYN packet of next\nconnection request.\n\nFix is to clear tp-\u003ewindow_clamp in tcp_disconnect().\n\nReported-by: Krzysztof Oledzki \u003cole@ans.pl\u003e\nTested-by: Krzysztof Oledzki \u003cole@ans.pl\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a2bfbc072e279ff81e6b336acff612b9bc2e5281", "tree": "89cfe21911e1539e1b7f89d8a10d42abecba2a18", "parents": [ "5c427ff9e4cc61625d48172ea082ae99e21eea6a", "82b3cc1a2f5e46300a9dec4a8cc8106dc20a4c23" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 17 00:05:02 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 17 00:05:02 2009 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/net/can/Kconfig\n" }, { "commit": "d792c1006fe92448217b71513d3955868358271d", "tree": "918d3fea89ab7f053a0a3f15c2b2feff24f81ef5", "parents": [ "d01032e4fd33110f9f3a085a36cb819c1dfc5827" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Fri Nov 13 13:56:33 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 13 13:56:33 2009 -0800" }, "message": "tcp: provide more information on the tcp receive_queue bugs\n\nThe addition of rcv_nxt allows to discern whether the skb\nwas out of place or tp-\u003ecopied. Also catch fancy combination\nof flags if necessary (sadly we might miss the actual causer\nflags as it might have already returned).\n\nBtw, we perhaps would want to forward copied_seq in\nsomewhere or otherwise we might have some nice loop with\nWARN stuff within but where to do that safely I don\u0027t\nknow at this stage until more is known (but it is not\nmade significantly worse by this patch).\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cfadf853f6cd9689f79a63ca960c6f9d6665314f", "tree": "35418e342d9783f0974ea33ef03875aa21d2362a", "parents": [ "05423b241311c9380b7280179295bac7794281b6", "f568a926a353d6816b3704d7367f34371df44ce7" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 27 01:03:26 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 27 01:03:26 2009 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/net/sh_eth.c\n" }, { "commit": "c62f4c453ab4b0240ab857bfd089da2c01ad91e7", "tree": "2c99ed102d752d762cc051ee1e89539e0453bdc0", "parents": [ "74eee2e8d08048c847d1998c686e12a477ff939a" ], "author": { "name": "Arjan van de Ven", "email": "arjan@linux.intel.com", "time": "Thu Oct 22 21:37:56 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 22 21:37:56 2009 -0700" }, "message": "net: use WARN() for the WARN_ON in commit b6b39e8f3fbbb\n\nCommit b6b39e8f3fbbb (tcp: Try to catch MSG_PEEK bug) added a printk()\nto the WARN_ON() that\u0027s in tcp.c. This patch changes this combination\nto WARN(); the advantage of WARN() is that the printk message shows up\ninside the message, so that kerneloops.org will collect the message.\n\nIn addition, this gets rid of an extra if() statement.\n\nSigned-off-by: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b6b39e8f3fbbb31001b836afec87bcaf4811a7bf", "tree": "33cf8d3732bde60abc7651b0ecf78e297d7b0afb", "parents": [ "55b8050353c4a212c94d7156e2bd5885225b869b" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Mon Oct 19 19:41:06 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 20 00:51:57 2009 -0700" }, "message": "tcp: Try to catch MSG_PEEK bug\n\nThis patch tries to print out more information when we hit the\nMSG_PEEK bug in tcp_recvmsg. It\u0027s been around since at least\n2005 and it\u0027s about time that we finally fix it.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b103cf34382f26ff48a87931b83f13b177b47c1a", "tree": "6922379bca3fc9935d394bc474a91b859ac06813", "parents": [ "0c3d79bce48034018e840468ac5a642894a521a3" ], "author": { "name": "Julian Anastasov", "email": "ja@ssi.bg", "time": "Mon Oct 19 10:10:40 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Oct 19 19:19:06 2009 -0700" }, "message": "tcp: fix TCP_DEFER_ACCEPT retrans calculation\n\nFix TCP_DEFER_ACCEPT conversion between seconds and\nretransmission to match the TCP SYN-ACK retransmission periods\nbecause the time is converted to such retransmissions. The old\nalgorithm selects one more retransmission in some cases. Allow\nup to 255 retransmissions.\n\nSigned-off-by: Julian Anastasov \u003cja@ssi.bg\u003e\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c720c7e8383aff1cb219bddf474ed89d850336e3", "tree": "4f12337e6690fccced376db9f501eaf98614a65e", "parents": [ "988ade6b8e27e79311812f83a87b5cea11fabcd7" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Thu Oct 15 06:30:45 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Oct 18 18:52:53 2009 -0700" }, "message": "inet: rename some inet_sock fields\n\nIn order to have better cache layouts of struct sock (separate zones\nfor rx/tx paths), we need this preliminary patch.\n\nGoal is to transfert fields used at lookup time in the first\nread-mostly cache line (inside struct sock_common) and move sk_refcnt\nto a separate cache line (only written by rx path)\n\nThis patch adds inet_ prefix to daddr, rcv_saddr, dport, num, saddr,\nsport and id fields. This allows a future patch to define these\nfields as macros, like sk_refcnt, without name clashes.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f373b53b5fe67aa4a6f28f921a529cc90f88e79b", "tree": "b2cae9152aed2e30b7a39c114678b9355ab5c14f", "parents": [ "c3faca053d0a9c877597935b434150b422dbc6fb" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Fri Oct 09 00:16:19 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 13 03:44:02 2009 -0700" }, "message": "tcp: replace ehash_size by ehash_mask\n\nStoring the mask (size - 1) instead of the size allows fast path to be\na bit faster.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "42324c62704365d6a3e89138dea55909d2f26afe", "tree": "6a56f042aebb4e0dbc8da7e904d79989e42b3666", "parents": [ "914a9ab386a288d0f22252fc268ecbc048cdcbd5" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Thu Oct 01 15:26:00 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Oct 02 09:46:05 2009 -0700" }, "message": "net: splice() from tcp to pipe should take into account O_NONBLOCK\n\ntcp_splice_read() doesnt take into account socket\u0027s O_NONBLOCK flag\n\nBefore this patch :\n\nsplice(socket,0,pipe,0,128*1024,SPLICE_F_MOVE);\ncauses a random endless block (if pipe is full) and\nsplice(socket,0,pipe,0,128*1024,SPLICE_F_MOVE | SPLICE_F_NONBLOCK);\nwill return 0 immediately if the TCP buffer is empty.\n\nUser application has no way to instruct splice() that socket should be in blocking mode\nbut pipe in nonblock more.\n\nMany projects cannot use splice(tcp -\u003e pipe) because of this flaw.\n\nhttp://git.samba.org/?p\u003dsamba.git;a\u003dhistory;f\u003dsource3/lib/recvfile.c;h\u003dea0159642137390a0f7e57a123684e6e63e47581;hb\u003dHEAD\nhttp://lkml.indiana.edu/hypermail/linux/kernel/0807.2/0687.html\n\nLinus introduced SPLICE_F_NONBLOCK in commit 29e350944fdc2dfca102500790d8ad6d6ff4f69d\n(splice: add SPLICE_F_NONBLOCK flag )\n\n It doesn\u0027t make the splice itself necessarily nonblocking (because the\n actual file descriptors that are spliced from/to may block unless they\n have the O_NONBLOCK flag set), but it makes the splice pipe operations\n nonblocking.\n\nLinus intention was clear : let SPLICE_F_NONBLOCK control the splice pipe mode only\n\nThis patch instruct tcp_splice_read() to use the underlying file O_NONBLOCK\nflag, as other socket operations do.\n\nUsers will then call :\n\nsplice(socket,0,pipe,0,128*1024,SPLICE_F_MOVE | SPLICE_F_NONBLOCK );\n\nto block on data coming from socket (if file is in blocking mode),\nand not block on pipe output (to avoid deadlock)\n\nFirst version of this patch was submitted by Octavian Purdila\n\nReported-by: Volker Lendecke \u003cvl@samba.org\u003e\nReported-by: Jason Gunthorpe \u003cjgunthorpe@obsidianresearch.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: Octavian Purdila \u003copurdila@ixiacom.com\u003e\nAcked-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4fdb78d3093a347456e108b77d56d493d29071b2", "tree": "2ad6dc10dc3603d40caaa26fbdb6684f30987a4e", "parents": [ "28ad3957b913855e8d41a27f7b90bed944809625" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Thu Oct 01 15:02:20 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 01 15:02:20 2009 -0700" }, "message": "net/ipv4/tcp.c: fix min() type mismatch warning\n\nnet/ipv4/tcp.c: In function \u0027do_tcp_setsockopt\u0027:\nnet/ipv4/tcp.c:2050: warning: comparison of distinct pointer types lacks a cast\n\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b7058842c940ad2c08dd829b21e5c92ebe3b8758", "tree": "5fe78d599fc345ca0bcd4b083b79095a54b2921b", "parents": [ "eb1cf0f8f7a9e5a6d573d5bd72c015686a042db0" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Sep 30 16:12:20 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Sep 30 16:12:20 2009 -0700" }, "message": "net: Make setsockopt() optlen be unsigned.\n\nThis provides safety against negative optlen at the type\nlevel instead of depending upon (sometimes non-trivial)\nchecks against this sprinkled all over the the place, in\neach and every implementation.\n\nBased upon work done by Arjan van de Ven and feedback\nfrom Linus Torvalds.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4481374ce88ba8f460c8b89f2572027bd27057d0", "tree": "6896601b6a1da0e3e932ffa75fcff966c834c02c", "parents": [ "4738e1b9cf8f9e28d7de080a5e6ce5d0095ea18f" ], "author": { "name": "Jan Beulich", "email": "JBeulich@novell.com", "time": "Mon Sep 21 17:03:05 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Sep 22 07:17:38 2009 -0700" }, "message": "mm: replace various uses of num_physpages by totalram_pages\n\nSizing of memory allocations shouldn\u0027t depend on the number of physical\npages found in a system, as that generally includes (perhaps a huge amount\nof) non-RAM pages. The amount of what actually is usable as storage\nshould instead be used as a basis here.\n\nSome of the calculations (i.e. those not intending to use high memory)\nshould likely even use (totalram_pages - totalhigh_pages).\n\nSigned-off-by: Jan Beulich \u003cjbeulich@novell.com\u003e\nAcked-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nAcked-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Dave Airlie \u003cairlied@linux.ie\u003e\nCc: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nCc: Jeremy Fitzhardinge \u003cjeremy@goop.org\u003e\nCc: Pekka Enberg \u003cpenberg@cs.helsinki.fi\u003e\nCc: Hugh Dickins \u003chugh.dickins@tiscali.co.uk\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0b6a05c1dbebe8c616e2e5b0f52b7a01fd792911", "tree": "18b5a2d16ac3fa8dd6d4528469725c216aed428c", "parents": [ "036d6a673fa0a2e2c5b72a3b1d1b86114c1711c0" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Tue Sep 15 01:30:10 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Sep 15 01:30:10 2009 -0700" }, "message": "tcp: fix ssthresh u16 leftover\n\nIt was once upon time so that snd_sthresh was a 16-bit quantity.\n...That has not been true for long period of time. I run across\nsome ancient compares which still seem to trust such legacy.\nPut all that magic into a single place, I hopefully found all\nof them.\n\nCompile tested, though linking of allyesconfig is ridiculous\nnowadays it seems.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "aa1330766c49199bdab4d4a9096d98b072df9044", "tree": "98787478dbef0faa7caee09c4996abcda723a608", "parents": [ "05c6a8d7a7d778f26d8eb821556988993b766092" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Wed Sep 02 23:45:45 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Sep 02 23:45:45 2009 -0700" }, "message": "tcp: replace hard coded GFP_KERNEL with sk_allocation\n\nThis fixed a lockdep warning which appeared when doing stress\nmemory tests over NFS:\n\n\tinconsistent {RECLAIM_FS-ON-W} -\u003e {IN-RECLAIM_FS-W} usage.\n\n\tpage reclaim \u003d\u003e nfs_writepage \u003d\u003e tcp_sendmsg \u003d\u003e lock sk_lock\n\n\tmount_root \u003d\u003e nfs_root_data \u003d\u003e tcp_close \u003d\u003e lock sk_lock \u003d\u003e\n\t\t\ttcp_send_fin \u003d\u003e alloc_skb_fclone \u003d\u003e page reclaim\n\nDavid raised a concern that if the allocation fails in tcp_send_fin(), and it\u0027s\nGFP_ATOMIC, we are going to yield() (which sleeps) and loop endlessly waiting\nfor the allocation to succeed.\n\nBut fact is, the original GFP_KERNEL also sleeps. GFP_ATOMIC+yield() looks\nweird, but it is no worse the implicit sleep inside GFP_KERNEL. Both could\nloop endlessly under memory pressure.\n\nCC: Arnaldo Carvalho de Melo \u003cacme@ghostprotocols.net\u003e\nCC: David S. Miller \u003cdavem@davemloft.net\u003e\nCC: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "df19a6267705456f463871ae2aabc44299909d2a", "tree": "59654abf9eb5dc58c9f8773dde5738b5bcb18312", "parents": [ "5e9b2dbfcbc9b180a0064d8a473a4652ee99c9ad" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Fri Aug 28 23:48:54 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Aug 28 23:48:54 2009 -0700" }, "message": "tcp: keepalive cleanups\n\nIntroduce keepalive_probes(tp) helper, and use it, like \nkeepalive_time_when(tp) and keepalive_intvl_when(tp)\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a57de0b4336e48db2811a2030bb68dba8dd09d88", "tree": "a01c189d5fd55c69c9e2e842241e84b46728bc60", "parents": [ "1b614fb9a00e97b1eab54d4e442d405229c059dd" ], "author": { "name": "Jiri Olsa", "email": "jolsa@redhat.com", "time": "Wed Jul 08 12:09:13 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 09 17:06:57 2009 -0700" }, "message": "net: adding memory barrier to the poll and receive callbacks\n\nAdding memory barrier after the poll_wait function, paired with\nreceive callbacks. Adding fuctions sock_poll_wait and sk_has_sleeper\nto wrap the memory barrier.\n\nWithout the memory barrier, following race can happen.\nThe race fires, when following code paths meet, and the tp-\u003ercv_nxt\nand __add_wait_queue updates stay in CPU caches.\n\nCPU1 CPU2\n\nsys_select receive packet\n ... ...\n __add_wait_queue update tp-\u003ercv_nxt\n ... ...\n tp-\u003ercv_nxt check sock_def_readable\n ... {\n schedule ...\n if (sk-\u003esk_sleep \u0026\u0026 waitqueue_active(sk-\u003esk_sleep))\n wake_up_interruptible(sk-\u003esk_sleep)\n ...\n }\n\nIf there was no cache the code would work ok, since the wait_queue and\nrcv_nxt are opposit to each other.\n\nMeaning that once tp-\u003ercv_nxt is updated by CPU2, the CPU1 either already\npassed the tp-\u003ercv_nxt check and sleeps, or will get the new value for\ntp-\u003ercv_nxt and will return with new data mask.\nIn both cases the process (CPU1) is being added to the wait queue, so the\nwaitqueue_active (CPU2) call cannot miss and will wake up CPU1.\n\nThe bad case is when the __add_wait_queue changes done by CPU1 stay in its\ncache, and so does the tp-\u003ercv_nxt update on CPU2 side. The CPU1 will then\nendup calling schedule and sleep forever if there are no more data on the\nsocket.\n\nCalls to poll_wait in following modules were ommited:\n\tnet/bluetooth/af_bluetooth.c\n\tnet/irda/af_irda.c\n\tnet/irda/irnet/irnet_ppp.c\n\tnet/mac80211/rc80211_pid_debugfs.c\n\tnet/phonet/socket.c\n\tnet/rds/af_rds.c\n\tnet/rfkill/core.c\n\tnet/sunrpc/cache.c\n\tnet/sunrpc/rpc_pipe.c\n\tnet/tipc/socket.c\n\nSigned-off-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6828b92bd21acd65113dfe0541f19f5df0d9668f", "tree": "be9b1009500891d7598a2c19a63fad9c0a28f17c", "parents": [ "8e5b9dda99cc86bdbd822935fcc37c5808e271b3" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Sun Jun 28 18:06:41 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jun 29 19:41:43 2009 -0700" }, "message": "tcp: Do not tack on TSO data to non-TSO packet\n\nIf a socket starts out on a non-TSO route, and then switches to\na TSO route, then we will tack on data to the tail of the tx queue\neven if it started out life as non-TSO. This is suboptimal because\nall of it will then be copied and checksummed unnecessarily.\n\nThis patch fixes this by ensuring that skb-\u003eip_summed is set to\nCHECKSUM_PARTIAL before appending extra data beyond the MSS.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "915219441d566f1da0caa0e262be49b666159e17", "tree": "91d40271d6601bc7ae10c4ab0cff2a47bb38d4e3", "parents": [ "de1033428baf1940bbbbf9e66b073ee0a577f5e0" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 28 21:35:47 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 28 21:35:47 2009 -0700" }, "message": "tcp: Use SKB queue and list helpers instead of doing it by-hand.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a2a804cddfe65f18f903985e8a8d04c7c9eec354", "tree": "93c81d1a672b52ff8ec38fefa227ab894b37ec62", "parents": [ "9aaa156cf9b7e9d9ed899f254283b91c4e3c36c8" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Tue May 26 18:50:34 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 27 03:26:05 2009 -0700" }, "message": "tcp: Do not check flush when comparing options for GRO\n\nThere is no need to repeatedly check flush when comparing TCP\noptions for GRO as it will be false 99% of the time where it\nmatters.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a5b1cf288d4200506ab62fbb86cc81ace948a306", "tree": "c95339866f4d67220d111811a92e5c6b79de6ab4", "parents": [ "7489594cb249aeb178287c9a43a9e4f366044259" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Tue May 26 18:50:28 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 27 03:26:01 2009 -0700" }, "message": "gro: Avoid unnecessary comparison after skb_gro_header\n\nFor the overwhelming majority of cases, skb_gro_header\u0027s return\nvalue cannot be NULL. Yet we must check it because of its current\nform. This patch splits it up into multiple functions in order\nto avoid this.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "30a3ae30c775e2723f86ef70746ad3cb4404a4c9", "tree": "9a3365e4a2b611930f31674a0162126625adac10", "parents": [ "4a9a2968a17eae42ef5dffca8b37534c864e30cc" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Tue May 26 18:50:26 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 27 03:26:00 2009 -0700" }, "message": "tcp: Optimise len/mss comparison\n\nInstead of checking len \u003e mss || len \u003d\u003d 0, we can accomplish\nboth by checking (len - 1) \u003e mss using the unsigned wraparound.\nAt nearly a million times a second, this might just help.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4a9a2968a17eae42ef5dffca8b37534c864e30cc", "tree": "8fa6f421a04491546a25f18a30e2133d50d83c64", "parents": [ "745898eaf0eb7a04a56dec1188d9148259510863" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Tue May 26 18:50:25 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 27 03:25:59 2009 -0700" }, "message": "tcp: Remove unnecessary window comparisons for GRO\n\nThe window has already been checked as part of the flag word\nso there is no need to check it explicitly.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "745898eaf0eb7a04a56dec1188d9148259510863", "tree": "c4ad29d7bc6e8ceb5a96f68cee363f871b2bf17f", "parents": [ "78d3fd0b7de844a6dad56e9620fc9d2271b32ab9" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Tue May 26 18:50:24 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 27 03:25:57 2009 -0700" }, "message": "tcp: Optimise GRO port comparisons\n\nInstead of doing two 16-bit operations for the source/destination\nports, we can do one 32-bit operation to take care both.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "775273131810caa41dfc7f9e552ea5d8508caf40", "tree": "38ee8e4815d614c483be352524134c6f7fc163c3", "parents": [ "705efc3b03cbee449e4d83b230423894152f7982" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Sun May 10 20:32:34 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon May 18 15:05:40 2009 -0700" }, "message": "tcp: fix MSG_PEEK race check\n\nCommit 518a09ef11 (tcp: Fix recvmsg MSG_PEEK influence of\nblocking behavior) lets the loop run longer than the race check\ndid previously expect, so we need to be more careful with this\ncheck and consider the work we have been doing.\n\nI tried my best to deal with urg hole madness too which happens\nhere:\n\tif (!sock_flag(sk, SOCK_URGINLINE)) {\n\t\t++*seq;\n\t\t...\nby using additional offset by one but I certainly have very\nlittle interest in testing that part.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nTested-by: Frans Pop \u003celendil@planet.nl\u003e\nTested-by: Ian Zimmermann \u003citz@buug.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a0a69a0106dab8d20596f97f6674bed3b394d1ee", "tree": "2477a32cab5f8afdb1c158f0b151048ee7afc594", "parents": [ "9dd014eb9804f19d6230c3cbc10fa25f5416bda7" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Fri Apr 17 02:34:38 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Apr 17 02:34:38 2009 -0700" }, "message": "gro: Fix use after free in tcp_gro_receive\n\nAfter calling skb_gro_receive skb-\u003elen can no longer be relied\non since if the skb was merged using frags, then its pages will\nhave been removed and the length reduced.\n\nThis caused tcp_gro_receive to prematurely end merging which\nresulted in suboptimal performance with ixgbe.\n\nThe fix is to store skb-\u003elen on the stack.\n\nReported-by: Mark Wagner \u003cmwagner@redhat.com\u003e\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "377f0a08e4cb56658d878d22c3aed4716e283c6b", "tree": "9b18ef210d92a04dc95bd694f680e52bbdf51223", "parents": [ "eeb5f5c9b3c8bee5ab4794323c1d23d100a9db59" ], "author": { "name": "Rami Rosen", "email": "ramirose@gmail.com", "time": "Tue Mar 31 14:43:17 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Mar 31 14:43:17 2009 -0700" }, "message": "ipv4: remove unused parameter from tcp_recv_urg().\n\nSigned-off-by: Rami Rosen \u003cramirose@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "beedad923ad6237f03265fdf86eb8a1b50d14ae9", "tree": "14de83c72e68bcae559616f45da16dd990065cd0", "parents": [ "9bdd8d40c8c59435664af6049dabe24b7779b203" ], "author": { "name": "Rami Rosen", "email": "ramirose@gmail.com", "time": "Wed Mar 18 18:50:09 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 18 18:50:09 2009 -0700" }, "message": "tcp: remove parameter from tcp_recv_urg().\n\nThis patch removes an unused parameter (addr_len) from tcp_recv_urg()\nmethod in net/ipv4/tcp.c.\n\nSigned-off-by: Rami Rosen \u003cramirose@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "afece1c6587010cc81d1a43045c855774e8234a3", "tree": "00caa55dc0c2a86c44883154986885f0421d1251", "parents": [ "2a3a041c4e2c1685e668b280c121a5a40a029a03" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Sat Mar 14 14:23:07 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Mar 15 20:09:55 2009 -0700" }, "message": "tcp: make sure xmit goal size never becomes zero\n\nIt\u0027s not too likely to happen, would basically require crafted\npackets (must hit the max guard in tcp_bound_to_half_wnd()).\nIt seems that nothing that bad would happen as there\u0027s tcp_mems\nand congestion window that prevent runaway at some point from\nhurting all too much (I\u0027m not that sure what all those zero\nsized segments we would generate do though in write queue).\nPreventing it regardless is certainly the best way to go.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nCc: Evgeniy Polyakov \u003czbr@ioremap.net\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2a3a041c4e2c1685e668b280c121a5a40a029a03", "tree": "dab327c0cbebec79968c3c2207d064a9152857f0", "parents": [ "0c54b85f2828128274f319a1eb3ce7f604fe2a53" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Sat Mar 14 22:45:16 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Mar 15 20:09:55 2009 -0700" }, "message": "tcp: cache result of earlier divides when mss-aligning things\n\nThe results is very unlikely change every so often so we\nhardly need to divide again after doing that once for a\nconnection. Yet, if divide still becomes necessary we\ndetect that and do the right thing and again settle for\nnon-divide state. Takes the u16 space which was previously\ntaken by the plain xmit_size_goal.\n\nThis should take care part of the tso vs non-tso difference\nwe found earlier.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0c54b85f2828128274f319a1eb3ce7f604fe2a53", "tree": "8ff84b9aa9cd4775f244c2421e11adfbd206a7bf", "parents": [ "72211e90501f954f586481c25521c3724cda3cc7" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Sat Mar 14 14:23:05 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Mar 15 20:09:54 2009 -0700" }, "message": "tcp: simplify tcp_current_mss\n\nThere\u0027s very little need for most of the callsites to get\ntp-\u003exmit_goal_size updated. That will cost us divide as is,\nso slice the function in two. Also, the only users of the\ntp-\u003exmit_goal_size are directly behind tcp_current_mss(),\nso there\u0027s no need to store that variable into tcp_sock\nat all! The drop of xmit_goal_size currently leaves 16-bit\nhole and some reorganization would again be necessary to\nchange that (but I\u0027m aiming to fill that hole with u16\nxmit_goal_size_segs to cache the results of the remaining\ndivide to get that tso on regression).\n\nBring xmit_goal_size parts into tcp.c\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nCc: Evgeniy Polyakov \u003czbr@ioremap.net\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0d6a775e27d975e5f9ea8e2911216d84face50ca", "tree": "a19fc1938325c517cab08ef7706115e784c72bb5", "parents": [ "cabeccbd172cc305f4383f5a4808ae254745275f" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Sat Feb 28 04:44:41 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 02 03:00:16 2009 -0800" }, "message": "tcp: in sendmsg/pages open code the real goto target\n\ncopied was assigned zero right before the goto, so if (copied)\ncannot ever be true.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "aa6320d336971171df1d13c1c284facf10804881", "tree": "3e659464dd24b9634897d631931f334ab1c22179", "parents": [ "a5ad24be728d4352b71a81fba471aa41eb71f83a" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Sun Feb 08 18:00:40 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Feb 08 20:22:19 2009 -0800" }, "message": "gro: Optimise TCP packet reception\n\ngro: Optimise TCP packet reception\n\nAs this function can be called more than half a million times for\n10GbE, it\u0027s important to optimise it as much as we can.\n\nThis patch uses bit ops to logical ops, as well as open coding\nmemcmp to exploit alignment properties.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "05bee4737774881e027bfd9a8b5c40a7d68f6325", "tree": "5fd32eb8fe345d2d30e85910beac241a4439e8cd", "parents": [ "80595d59ba9917227856e663da249c2276a8628d", "905db44087855e3c1709f538ecdc22fd149cadd8" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jan 30 14:31:07 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jan 30 14:31:07 2009 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tdrivers/net/e1000/e1000_main.c\n" }, { "commit": "86911732d3996a9da07914b280621450111bb6da", "tree": "e787240d5ba869ddf4d0adfc3f9c69e0372e96ef", "parents": [ "5d0d9be8ef456afc6c3fb5f8aad06ef19b704b05" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Thu Jan 29 14:19:50 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jan 29 16:33:03 2009 -0800" }, "message": "gro: Avoid copying headers of unmerged packets\n\nUnfortunately simplicity isn\u0027t always the best. The fraginfo\ninterface turned out to be suboptimal. The problem was quite\nobvious. For every packet, we have to copy the headers from\nthe frags structure into skb-\u003ehead, even though for 99% of the\npackets this part is immediately thrown away after the merge.\n\nLRO didn\u0027t have this problem because it directly read the headers\nfrom the frags structure.\n\nThis patch attempts to address this by creating an interface\nthat allows GRO to access the headers in the first frag without\nhaving to copy it. Because all drivers that use frags place the\nheaders in the first frag this optimisation should be enough.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9fa5fdf291c9b58b1cb8b4bb2a0ee57efa21d635", "tree": "cfa28bcf674ef224a5ab66b3b8bc4bc1091ccb11", "parents": [ "a7a41acf99d9150b424839b0d7b4f5ad9d211e2d" ], "author": { "name": "Dimitris Michailidis", "email": "dm@chelsio.com", "time": "Mon Jan 26 22:15:31 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 26 22:15:31 2009 -0800" }, "message": "tcp: Fix length tcp_splice_data_recv passes to skb_splice_bits.\n\ntcp_splice_data_recv has two lengths to consider: the len parameter it\ngets from tcp_read_sock, which specifies the amount of data in the skb,\nand rd_desc-\u003ecount, which is the amount of data the splice caller still\nwants. Currently it passes just the latter to skb_splice_bits, which then\nsplices min(rd_desc-\u003ecount, skb-\u003elen - offset) bytes.\n\nMost of the time this is fine, except when the skb contains urgent data.\nIn that case len goes only up to the urgent byte and is less than\nskb-\u003elen - offset. By ignoring len tcp_splice_data_recv may a) splice\ndata tcp_read_sock told it not to, b) return to tcp_read_sock a value \u003e len.\n\nNow, tcp_read_sock doesn\u0027t handle used \u003e len and leaves the socket in a\nbad state (both sk_receive_queue and copied_seq are bad at that point)\nresulting in duplicated data and corruption.\n\nFix by passing min(rd_desc-\u003ecount, len) to skb_splice_bits.\n\nSigned-off-by: Dimitris Michailidis \u003cdm@chelsio.com\u003e\nAcked-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4e704ee3c2cd38748ca59d835435d6a7e7f6f613", "tree": "24d4f83bb55748fa5ae79f302b641663fa0488ba", "parents": [ "f557206800801410c30e53ce7a27219b2c4cf0ba" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Wed Jan 14 20:41:12 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Jan 14 20:41:12 2009 -0800" }, "message": "gso: Ensure that the packet is long enough\n\nWhen we get a GSO packet from an untrusted source, we need to\nensure that it is sufficiently long so that we don\u0027t end up\ncrashing.\n\nBased on discovery and patch by Ian Campbell.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nTested-by: Ian Campbell \u003cian.campbell@citrix.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "33966dd0e2f68f26943cd9ee93ec6abbc6547a8e", "tree": "75502831bb2868e016ed4f55d3ea3238b50f609b", "parents": [ "9a0811ca4e9cf7be320ae8a5a43a259feb70074f" ], "author": { "name": "Willy Tarreau", "email": "w@1wt.eu", "time": "Tue Jan 13 16:04:36 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Jan 13 16:04:36 2009 -0800" }, "message": "tcp: splice as many packets as possible at once\n\nAs spotted by Willy Tarreau, current splice() from tcp socket to pipe is not\noptimal. It processes at most one segment per call.\nThis results in low performance and very high overhead due to syscall rate\nwhen splicing from interfaces which do not support LRO.\n\nWilly provided a patch inside tcp_splice_read(), but a better fix\nis to let tcp_read_sock() process as many segments as possible, so\nthat tcp_rcv_space_adjust() and tcp_cleanup_rbuf() are called less\noften.\n\nWith this change, splice() behaves like tcp_recvmsg(), being able\nto consume many skbs in one system call. With typical 1460 bytes\nof payload per frame, that means splice(SPLICE_F_NONBLOCK) can return\n16*1460 \u003d 23360 bytes.\n\nSigned-off-by: Willy Tarreau \u003cw@1wt.eu\u003e\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d9e8a3a5b8298a3c814ed37ac5756e6f67b6be41", "tree": "ffa1cf8b367b3f32155f6336d7b86b781a368019", "parents": [ "2150edc6c5cf00f7adb54538b9ea2a3e9cedca3f", "b9bdcbba010c2e49c8f837ea7a49fe006b636f41" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 09 11:52:14 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 09 11:52:14 2009 -0800" }, "message": "Merge branch \u0027next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx\n\n* \u0027next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (22 commits)\n ioat: fix self test for multi-channel case\n dmaengine: bump initcall level to arch_initcall\n dmaengine: advertise all channels on a device to dma_filter_fn\n dmaengine: use idr for registering dma device numbers\n dmaengine: add a release for dma class devices and dependent infrastructure\n ioat: do not perform removal actions at shutdown\n iop-adma: enable module removal\n iop-adma: kill debug BUG_ON\n iop-adma: let devm do its job, don\u0027t duplicate free\n dmaengine: kill enum dma_state_client\n dmaengine: remove \u0027bigref\u0027 infrastructure\n dmaengine: kill struct dma_client and supporting infrastructure\n dmaengine: replace dma_async_client_register with dmaengine_get\n atmel-mci: convert to dma_request_channel and down-level dma_slave\n dmatest: convert to dma_request_channel\n dmaengine: introduce dma_request_channel and private channels\n net_dma: convert to dma_find_channel\n dmaengine: provide a common \u0027issue_pending_all\u0027 implementation\n dmaengine: centralize channel allocation, introduce dma_find_channel\n dmaengine: up-level reference counting to the module level\n ...\n" }, { "commit": "684f2176015b313ab59cecf574117969cf638f28", "tree": "e87c2285d35f7f308c9046940d4efadb9b6ad11a", "parents": [ "787e9208360117835101f513f7db593dc2525cf8" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Thu Jan 08 10:41:23 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jan 08 10:41:23 2009 -0800" }, "message": "tcp6: Add GRO support\n\nThis patch adds GRO support for TCP over IPv6. The code is exactly\nthe same as the IPv4 version except for the pseudo-header checksum\ncomputation.\n\nNote that I\u0027ve removed the unused tcphdr argument from tcp_v6_check\nrather than invent a bogus value for GRO.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f67b45999205164958de4ec0658d51fa4bee066d", "tree": "2fc3c2e833fdfea2f029dca05d5401ba9ae63229", "parents": [ "2ba05622b8b143b0c95968ba59bddfbd6d2f2559" ], "author": { "name": "Dan Williams", "email": "dan.j.williams@intel.com", "time": "Tue Jan 06 11:38:15 2009 -0700" }, "committer": { "name": "Dan Williams", "email": "dan.j.williams@intel.com", "time": "Tue Jan 06 11:38:15 2009 -0700" }, "message": "net_dma: convert to dma_find_channel\n\nUse the general-purpose channel allocation provided by dmaengine.\n\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\n\n" }, { "commit": "6f49a57aa5a0c6d4e4e27c85f7af6c83325a12d1", "tree": "afba24357d1f4ff69ccb2b39a19542546590a50b", "parents": [ "07f2211e4fbce6990722d78c4f04225da9c0e9cf" ], "author": { "name": "Dan Williams", "email": "dan.j.williams@intel.com", "time": "Tue Jan 06 11:38:14 2009 -0700" }, "committer": { "name": "Dan Williams", "email": "dan.j.williams@intel.com", "time": "Tue Jan 06 11:38:14 2009 -0700" }, "message": "dmaengine: up-level reference counting to the module level\n\nSimply, if a client wants any dmaengine channel then prevent all dmaengine\nmodules from being removed. Once the clients are done re-enable module\nremoval.\n\nWhy?, beyond reducing complication:\n1/ Tracking reference counts per-transaction in an efficient manner, as\n is currently done, requires a complicated scheme to avoid cache-line\n bouncing effects.\n2/ Per-transaction ref-counting gives the false impression that a\n dma-driver can be gracefully removed ahead of its user (net, md, or\n dma-slave)\n3/ None of the in-tree dma-drivers talk to hot pluggable hardware, but\n if such an engine were built one day we still would not need to notify\n clients of remove events. The driver can simply return NULL to a\n -\u003eprep() request, something that is much easier for a client to handle.\n\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nAcked-by: Maciej Sosnowski \u003cmaciej.sosnowski@intel.com\u003e\nSigned-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\n\n\n" }, { "commit": "7945cc6464a4db0caf6dfacdfe05806051c4cb7b", "tree": "8bcd5f9cb276229f1b25844249a4243b6f95f656", "parents": [ "4f7d54f59bc470f0aaa932f747a95232d7ebf8b1" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 05 00:59:00 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 05 00:59:00 2009 -0800" }, "message": "tcp: Kill extraneous SPLICE_F_NONBLOCK checks.\n\nIn splice TCP receive, the SPLICE_F_NONBLOCK flag is used\nto compute the \"timeo\" value. So checking it again inside\nof the main receive loop to trigger -EAGAIN processing is\nentirely unnecessary.\n\nNoticed by Jarek P. and Lennert Buytenhek.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4f7d54f59bc470f0aaa932f747a95232d7ebf8b1", "tree": "9d391a6c4d9fdb1c8534127702b5543b0b5ab089", "parents": [ "129fa44785a399248ae2466b6cb5c655e96668f7" ], "author": { "name": "Lennert Buytenhek", "email": "buytenh@marvell.com", "time": "Mon Jan 05 00:00:12 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 05 00:00:12 2009 -0800" }, "message": "tcp: don\u0027t mask EOF and socket errors on nonblocking splice receive\n\nCurrently, setting SPLICE_F_NONBLOCK on splice from a TCP socket\nresults in masking of EOF (RDHUP) and error conditions on the socket\nby an -EAGAIN return. Move the NONBLOCK check in tcp_splice_read()\nto be after the EOF and error checks to fix this.\n\nSigned-off-by: Lennert Buytenhek \u003cbuytenh@marvell.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b530256d2e0f1a75fab31f9821129fff1bb49faa", "tree": "e3e615d405921fc157169180d53c91d09b9a7186", "parents": [ "cfc3a44c3c32abe48898398d9a92e8524c976803" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Sun Jan 04 16:13:19 2009 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Jan 04 16:13:19 2009 -0800" }, "message": "gro: Use gso_size to store MSS\n\nIn order to allow GRO packets without frag_list at all, we need to\nstore the MSS in the packet itself. The obvious place is gso_size.\nThe only thing to watch out for is if the packet ends up not being\nGRO then we need to clear gso_size before pushing the packet into\nthe stack.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eb4dea5853046727bfbb579f0c9a8cae7369f7c6", "tree": "79e18d6284494ab63a890885b0eecce9431a9597", "parents": [ "0f23174aa8c1aa7a2a6050a72a60d290ef9ee578" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Mon Dec 29 23:04:08 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Dec 29 23:04:08 2008 -0800" }, "message": "net: Fix percpu counters deadlock\n\nWhen we converted the protocol atomic counters such as the orphan\ncount and the total socket count deadlocks were introduced due to\nthe mismatch in BH status of the spots that used the percpu counter\noperations.\n\nBased on the diagnosis and patch by Peter Zijlstra, this patch\nfixes these issues by disabling BH where we may be in process\ncontext.\n\nReported-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\nTested-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bf296b125b21b8d558ceb6ec30bb4eba2730cd6b", "tree": "5a06714c4001ccf1b217b888121f30d57fceea6d", "parents": [ "71d93b39e52e92aea35f1058d957cf12250d0b75" ], "author": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Mon Dec 15 23:43:36 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Dec 15 23:43:36 2008 -0800" }, "message": "tcp: Add GRO support\n\nThis patch adds the TCP-specific portion of GRO. The criterion for\nmerging is extremely strict (the TCP header must match exactly apart\nfrom the checksum) so as to allow refragmentation. Otherwise this\nis pretty much identical to LRO, except that we support the merging\nof ECN packets.\n\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "dd24c00191d5e4a1ae896aafe33c6b8095ab4bd1", "tree": "e955c09e0b288e50c706b6ee409229d5a930c80c", "parents": [ "1748376b6626acf59c24e9592ac67b3fe2a0e026" ], "author": { "name": "Eric Dumazet", "email": "dada1@cosmosbay.com", "time": "Tue Nov 25 21:17:14 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 25 21:17:14 2008 -0800" }, "message": "net: Use a percpu_counter for orphan_count\n\nInstead of using one atomic_t per protocol, use a percpu_counter\nfor \"orphan_count\", to reduce cache line contention on\nheavy duty network servers. \n\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1748376b6626acf59c24e9592ac67b3fe2a0e026", "tree": "65a28205daf1ca92e31389440764fc407365014e", "parents": [ "c1b56878fb68e9c14070939ea4537ad4db79ffae" ], "author": { "name": "Eric Dumazet", "email": "dada1@cosmosbay.com", "time": "Tue Nov 25 21:16:35 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 25 21:16:35 2008 -0800" }, "message": "net: Use a percpu_counter for sockets_allocated\n\nInstead of using one atomic_t per protocol, use a percpu_counter\nfor \"sockets_allocated\", to reduce cache line contention on\nheavy duty network servers. \n\nNote : We revert commit (248969ae31e1b3276fc4399d67ce29a5d81e6fd9\nnet: af_unix can make unix_nr_socks visbile in /proc),\nsince it is not anymore used after sock_prot_inuse_add() addition\n\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3ab5aee7fe840b5b1b35a8d1ac11c3de5281e611", "tree": "468296b7be813643248d4ca67497d6ddb6934fc6", "parents": [ "88ab1932eac721c6e7336708558fa5ed02c85c80" ], "author": { "name": "Eric Dumazet", "email": "dada1@cosmosbay.com", "time": "Sun Nov 16 19:40:17 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 16 19:40:17 2008 -0800" }, "message": "net: Convert TCP \u0026 DCCP hash tables to use RCU / hlist_nulls\n\nRCU was added to UDP lookups, using a fast infrastructure :\n- sockets kmem_cache use SLAB_DESTROY_BY_RCU and dont pay the\n price of call_rcu() at freeing time.\n- hlist_nulls permits to use few memory barriers.\n\nThis patch uses same infrastructure for TCP/DCCP established\nand timewait sockets.\n\nThanks to SLAB_DESTROY_BY_RCU, no slowdown for applications\nusing short lived TCP connections. A followup patch, converting\nrwlocks to spinlocks will even speedup this case.\n\n__inet_lookup_established() is pretty fast now we dont have to\ndirty a contended cache line (read_lock/read_unlock)\n\nOnly established and timewait hashtable are converted to RCU\n(bind table and listen table are still using traditional locking)\n\nSigned-off-by: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9eeda9abd1faf489f3df9a1f557975f4c8650363", "tree": "3e0a58e25b776cfbee193195460324dccb1886c7", "parents": [ "61c9eaf90081cbe6dc4f389e0056bff76eca19ec", "4bab0ea1d42dd1927af9df6fbf0003fc00617c50" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 06 22:43:03 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 06 22:43:03 2008 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\n\tdrivers/net/wireless/ath5k/base.c\n\tnet/8021q/vlan_core.c\n" }, { "commit": "518a09ef11f8454f4676125d47c3e775b300c6a5", "tree": "a22f9f9de58e91d3287bcd21277e9378de49e18f", "parents": [ "efb9a8c28ca0edd9e2572117105ebad9bbc0c368" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 05 03:36:01 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 05 03:36:01 2008 -0800" }, "message": "tcp: Fix recvmsg MSG_PEEK influence of blocking behavior.\n\nVito Caputo noticed that tcp_recvmsg() returns immediately from\npartial reads when MSG_PEEK is used. In particular, this means that\nSO_RCVLOWAT is not respected.\n\nSimply remove the test. And this matches the behavior of several\nother systems, including BSD.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5a5f3a8db9d70c90e9d55b46e02b2d8deb1c2c2e", "tree": "051b35d2766bf5285a695b82d527cc3c2d77f435", "parents": [ "d9319100c1ad7d0ed4045ded767684ad25670436" ], "author": { "name": "Jianjun Kong", "email": "jianjun@zeuux.org", "time": "Mon Nov 03 00:24:34 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 03 00:24:34 2008 -0800" }, "message": "net: clean up net/ipv4/ipip.c raw.c tcp.c tcp_minisocks.c tcp_yeah.c xfrm4_policy.c\n\nSigned-off-by: Jianjun Kong \u003cjianjun@zeuux.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" } ], "next": "33f5f57eeb0c6386fdd85f9c690dc8d700ba7928" }