)]}'
{
  "log": [
    {
      "commit": "455c6fdbd219161bd09b1165f11699d6d73de11c",
      "tree": "dd65155eb5a9410e83c5b0d760c099f9fbb7719c",
      "parents": [
        "fedc1ed0f11be666de066b0c78443254736a942e"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 20:40:15 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 20:40:15 2014 -0700"
      },
      "message": "Linux 3.14\n"
    },
    {
      "commit": "fedc1ed0f11be666de066b0c78443254736a942e",
      "tree": "d1911f00911c9fc5d22f90fb8492ee87ff9506e4",
      "parents": [
        "01358e562a8b97f50ec04025c009c71508e6d373",
        "38129a13e6e71f666e0468e99fdd932a687b4d7e"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:26:08 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:26:08 2014 -0700"
      },
      "message": "Merge branch \u0027for-linus-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull vfs fixes from Al Viro:\n \"Switch mnt_hash to hlist, turning the races between __lookup_mnt() and\n  hash modifications into false negatives from __lookup_mnt() (instead\n  of hangs)\"\n\nOn the false negatives from __lookup_mnt():\n \"The *only* thing we care about is not getting stuck in __lookup_mnt().\n  If it misses an entry because something in front of it just got moved\n  around, etc, we are fine.  We\u0027ll notice that mount_lock mismatch and\n  that\u0027ll be it\"\n\n* \u0027for-linus-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n  switch mnt_hash to hlist\n  don\u0027t bother with propagate_mnt() unless the target is shared\n  keep shadowed vfsmounts together\n  resizable namespace.c hashes\n"
    },
    {
      "commit": "01358e562a8b97f50ec04025c009c71508e6d373",
      "tree": "010d6ae6339d1da38bc17de894c5d4e59d7b1c60",
      "parents": [
        "915ac4e26ef9c39a0f831e935509243732abedc0"
      ],
      "author": {
        "name": "Randy Dunlap",
        "email": "rdunlap@infradead.org",
        "time": "Fri Mar 28 09:45:33 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:22:28 2014 -0700"
      },
      "message": "MAINTAINERS: resume as Documentation maintainer\n\nI am the new kernel tree Documentation maintainer (except for parts that\nare handled by other people, of course).\n\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nAcked-by: Rob Landley \u003crob@landley.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "915ac4e26ef9c39a0f831e935509243732abedc0",
      "tree": "68ac39fbf077217583f9936eada2c072301becb6",
      "parents": [
        "aa4af831bb4f3168f2f574b2620124699c09c4a3",
        "e4dbedc7eac7da9db363a36f2bd4366962eeefcc"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:20:40 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:20:40 2014 -0700"
      },
      "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input\n\nPull input updates from Dmitry Torokhov:\n \"Some more updates for the input subsystem.\n\n  You will get a fix for race in mousedev that has been causing quite a\n  few oopses lately and a small fixup for force feedback support in\n  evdev\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:\n  Input: mousedev - fix race when creating mixed device\n  Input: don\u0027t modify the id of ioctl-provided ff effect on upload failure\n"
    },
    {
      "commit": "aa4af831bb4f3168f2f574b2620124699c09c4a3",
      "tree": "e92e2484fce4b342e6051109c4ce07cdde40b2ca",
      "parents": [
        "00a1a053ebe5febcfc2ec498bd894f035ad2aa06"
      ],
      "author": {
        "name": "Eric Paris",
        "email": "eparis@redhat.com",
        "time": "Sun Mar 30 19:07:54 2014 -0400"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:02:53 2014 -0700"
      },
      "message": "AUDIT: Allow login in non-init namespaces\n\nIt its possible to configure your PAM stack to refuse login if audit\nmessages (about the login) were unable to be sent.  This is common in\nmany distros and thus normal configuration of many containers.  The PAM\nmodules determine if audit is enabled/disabled in the kernel based on\nthe return value from sending an audit message on the netlink socket.\nIf userspace gets back ECONNREFUSED it believes audit is disabled in the\nkernel.  If it gets any other error else it refuses to let the login\nproceed.\n\nJust about ever since the introduction of namespaces the kernel audit\nsubsystem has returned EPERM if the task sending a message was not in\nthe init user or pid namespace.  So many forms of containers have never\nworked if audit was enabled in the kernel.\n\nBUT if the container was not in net_init then the kernel network code\nwould send ECONNREFUSED (instead of the audit code sending EPERM).  Thus\nby pure accident/dumb luck/bug if an admin configured the PAM stack to\nreject all logins that didn\u0027t talk to audit, but then ran the login\nuntility in the non-init_net namespace, it would work!! Clearly this was\na bug, but it is a bug some people expected.\n\nWith the introduction of network namespace support in 3.14-rc1 the two\nbugs stopped cancelling each other out.  Now, containers in the\nnon-init_net namespace refused to let users log in (just like PAM was\nconfigfured!) Obviously some people were not happy that what used to let\nusers log in, now didn\u0027t!\n\nThis fix is kinda hacky.  We return ECONNREFUSED for all non-init\nrelevant namespaces.  That means that not only will the old broken\nnon-init_net setups continue to work, now the broken non-init_pid or\nnon-init_user setups will \u0027work\u0027.  They don\u0027t really work, since audit\nisn\u0027t logging things.  But it\u0027s what most users want.\n\nIn 3.15 we should have patches to support not only the non-init_net\n(3.14) namespace but also the non-init_pid and non-init_user namespace.\nSo all will be right in the world.  This just opens the doors wide open\non 3.14 and hopefully makes users happy, if not the audit system...\n\nReported-by: Andre Tomt \u003candre@tomt.net\u003e\nReported-by: Adam Richter \u003cadam_richter2004@yahoo.com\u003e\nSigned-off-by: Eric Paris \u003ceparis@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "00a1a053ebe5febcfc2ec498bd894f035ad2aa06",
      "tree": "2a6206f1b3d195168cc1d835b459ea99c2db7811",
      "parents": [
        "981e893ed537cbaa08f8089d480db4165e6d71c0"
      ],
      "author": {
        "name": "Theodore Ts\u0027o",
        "email": "tytso@mit.edu",
        "time": "Sun Mar 30 10:20:01 2014 -0400"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Mar 30 17:02:06 2014 -0700"
      },
      "message": "ext4: atomically set inode-\u003ei_flags in ext4_set_inode_flags()\n\nUse cmpxchg() to atomically set i_flags instead of clearing out the\nS_IMMUTABLE, S_APPEND, etc. flags and then setting them from the\nEXT4_IMMUTABLE_FL, EXT4_APPEND_FL flags, since this opens up a race\nwhere an immutable file has the immutable flag cleared for a brief\nwindow of time.\n\nReported-by: John Sullivan \u003cjsrhbz@kanargh.force9.co.uk\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "38129a13e6e71f666e0468e99fdd932a687b4d7e",
      "tree": "438e817fdf7d224f9fda1186eb24b1bbc37a4b5c",
      "parents": [
        "0b1b901b5a98bb36943d10820efc796f7cd45ff3"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Thu Mar 20 21:10:51 2014 -0400"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 30 19:18:51 2014 -0400"
      },
      "message": "switch mnt_hash to hlist\n\nfixes RCU bug - walking through hlist is safe in face of element moves,\nsince it\u0027s self-terminating.  Cyclic lists are not - if we end up jumping\nto another hash chain, we\u0027ll loop infinitely without ever hitting the\noriginal list head.\n\n[fix for dumb braino folded]\n\nSpotted by: Max Kellermann \u003cmk@cm4all.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "0b1b901b5a98bb36943d10820efc796f7cd45ff3",
      "tree": "72aa314fd3f3c37d152aba79b302b86ce7f6a1f6",
      "parents": [
        "1d6a32acd70ab18499829c0a9a5dbe2bace72a13"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Fri Mar 21 10:14:08 2014 -0400"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 30 19:18:50 2014 -0400"
      },
      "message": "don\u0027t bother with propagate_mnt() unless the target is shared\n\nIf the dest_mnt is not shared, propagate_mnt() does nothing -\nthere\u0027s no mounts to propagate to and thus no copies to create.\nMight as well don\u0027t bother calling it in that case.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "1d6a32acd70ab18499829c0a9a5dbe2bace72a13",
      "tree": "4f48079847ac500a4bb102c872fa10c41084009b",
      "parents": [
        "0818bf27c05b2de56c5b2bd08cfae2a939bd5f52"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Thu Mar 20 20:34:43 2014 -0400"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 30 19:18:50 2014 -0400"
      },
      "message": "keep shadowed vfsmounts together\n\npreparation to switching mnt_hash to hlist\n\nCc: stable@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "0818bf27c05b2de56c5b2bd08cfae2a939bd5f52",
      "tree": "5e66b630daecbd5514f2beb908802ce26cc900ea",
      "parents": [
        "981e893ed537cbaa08f8089d480db4165e6d71c0"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Fri Feb 28 13:46:44 2014 -0500"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 30 19:18:49 2014 -0400"
      },
      "message": "resizable namespace.c hashes\n\n* switch allocation to alloc_large_system_hash()\n* make sizes overridable by boot parameters (mhash_entries\u003d, mphash_entries\u003d)\n* switch mountpoint_hashtable from list_head to hlist_head\n\nCc: stable@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "981e893ed537cbaa08f8089d480db4165e6d71c0",
      "tree": "b0e8f100477be9e25feced30a9c1cbd75e0dc408",
      "parents": [
        "0f2776e6151a123552fd06b666fe755fa780a967",
        "cab5e127eef040399902caa8e1510795583fa03a"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 29 15:01:09 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 29 15:01:09 2014 -0700"
      },
      "message": "Merge branch \u0027timers-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull timer fix from Ingo Molnar:\n \"A late breaking fix from John.  (The bug fixed has a hard lockup\n  potential, but that was not observed, warnings were)\"\n\n* \u0027timers-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  time: Revert to calling clock_was_set_delayed() while in irq context\n"
    },
    {
      "commit": "0f2776e6151a123552fd06b666fe755fa780a967",
      "tree": "57afca1fa00e0c21697e3acdcccd7ed7c544ef77",
      "parents": [
        "49d8137a4039c63c834827f4bfe875e27bb9c521",
        "638c323c4d1f8eaf25224946e21ce8818f1bcee1"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 29 15:00:27 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 29 15:00:27 2014 -0700"
      },
      "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client\n\nPull Ceph fix from Sage Weil:\n \"This drops a bad assert that a few users have been hitting but we\u0027ve\n  only recently been able to track down\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:\n  rbd: drop an unsafe assertion\n"
    },
    {
      "commit": "e4dbedc7eac7da9db363a36f2bd4366962eeefcc",
      "tree": "79330a0e67bdea49030b68173ca016b9ce424a3c",
      "parents": [
        "fc7392aa1b20debc7f398acc39ffc817630f11e6"
      ],
      "author": {
        "name": "Dmitry Torokhov",
        "email": "dmitry.torokhov@gmail.com",
        "time": "Thu Mar 06 12:57:24 2014 -0800"
      },
      "committer": {
        "name": "Dmitry Torokhov",
        "email": "dmitry.torokhov@gmail.com",
        "time": "Sat Mar 29 14:44:23 2014 -0700"
      },
      "message": "Input: mousedev - fix race when creating mixed device\n\nWe should not be using static variable mousedev_mix in methods that can be\ncalled before that singleton gets assigned. While at it let\u0027s add open and\nclose methods to mousedev structure so that we do not need to test if we\nare dealing with multiplexor or normal device and simply call appropriate\nmethod directly.\n\nThis fixes: https://bugzilla.kernel.org/show_bug.cgi?id\u003d71551\n\nReported-by: GiulioDP \u003cdepasquale.giulio@gmail.com\u003e\nTested-by: GiulioDP \u003cdepasquale.giulio@gmail.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n"
    },
    {
      "commit": "fc7392aa1b20debc7f398acc39ffc817630f11e6",
      "tree": "135856bdb6fef8d7809ffe4e195c70c972a5af6a",
      "parents": [
        "8a0435d958fb36d93b8df610124a0e91e5675c82"
      ],
      "author": {
        "name": "Elias Vanderstuyft",
        "email": "elias.vds@gmail.com",
        "time": "Sat Mar 29 12:08:45 2014 -0700"
      },
      "committer": {
        "name": "Dmitry Torokhov",
        "email": "dmitry.torokhov@gmail.com",
        "time": "Sat Mar 29 12:13:09 2014 -0700"
      },
      "message": "Input: don\u0027t modify the id of ioctl-provided ff effect on upload failure\n\nIf a new (id \u003d\u003d -1) ff effect was uploaded from userspace,\nff-core.c::input_ff_upload() will have assigned a positive number to the\nnew effect id.  Currently, evdev.c::evdev_do_ioctl() will save this new id\nto userspace, regardless of whether the upload succeeded or not.\n\nOn upload failure, this can be confusing because the dev-\u003eff-\u003eeffects[]\narray will not contain an element at the index of that new effect id.\n\nThis patch fixes this by leaving the id unchanged after upload fails.\n\nNote: Unfortunately applications should still expect changed effect id for\nquite some time.\n\nThis has been discussed on:\nhttp://www.mail-archive.com/linux-input@vger.kernel.org/msg08513.html\n(\"ff-core effect id handling in case of a failed effect upload\")\n\nSuggested-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nSigned-off-by: Elias Vanderstuyft \u003celias.vds@gmail.com\u003e\nSigned-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n"
    },
    {
      "commit": "638c323c4d1f8eaf25224946e21ce8818f1bcee1",
      "tree": "e45a1de3d0c8936f14a5933b72dd423abe870e35",
      "parents": [
        "0414855fdc4a40da05221fc6062cccbc0c30f169"
      ],
      "author": {
        "name": "Alex Elder",
        "email": "elder@linaro.org",
        "time": "Tue Mar 25 15:36:02 2014 +0200"
      },
      "committer": {
        "name": "Sage Weil",
        "email": "sage@inktank.com",
        "time": "Sat Mar 29 10:38:14 2014 -0700"
      },
      "message": "rbd: drop an unsafe assertion\n\nOlivier Bonvalet reported having repeated crashes due to a failed\nassertion he was hitting in rbd_img_obj_callback():\n\n    Assertion failure in rbd_img_obj_callback() at line 2165:\n\trbd_assert(which \u003e\u003d img_request-\u003enext_completion);\n\nWith a lot of help from Olivier with reproducing the problem\nwe were able to determine the object and image requests had\nalready been completed (and often freed) at the point the\nassertion failed.\n\nThere was a great deal of discussion on the ceph-devel mailing list\nabout this.  The problem only arose when there were two (or more)\nobject requests in an image request, and the problem was always\nseen when the second request was being completed.\n\nThe problem is due to a race in the window between setting the\n\"done\" flag on an object request and checking the image request\u0027s\nnext completion value.  When the first object request completes, it\nchecks to see if its successor request is marked \"done\", and if\nso, that request is also completed.  In the process, the image\nrequest\u0027s next_completion value is updated to reflect that both\nthe first and second requests are completed.  By the time the\nsecond request is able to check the next_completion value, it\nhas been set to a value *greater* than its own \"which\" value,\nwhich caused an assertion to fail.\n\nFix this problem by skipping over any completion processing\nunless the completing object request is the next one expected.\nTest only for inequality (not \u003e\u003d), and eliminate the bad\nassertion.\n\nTested-by: Olivier Bonvalet \u003cob@daevel.fr\u003e\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nReviewed-by: Sage Weil \u003csage@inktank.com\u003e\nReviewed-by: Ilya Dryomov \u003cilya.dryomov@inktank.com\u003e\n"
    },
    {
      "commit": "49d8137a4039c63c834827f4bfe875e27bb9c521",
      "tree": "9994eb621829cdac210f215ba4cbe6d9e2bf78be",
      "parents": [
        "bc53267e97e4942e7f22c31cd8cf87840796accb",
        "5f2feca20e0f4878fa16787bd50fcd82fb365be7"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 15:09:37 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 15:09:37 2014 -0700"
      },
      "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) We\u0027ve discovered a common error in several networking drivers, they\n    put VLAN offload features into -\u003evlan_features, which would suggest\n    that they support offloading 2 or more levels of VLAN encapsulation.\n    Not only do these devices not do that, but we don\u0027t have the\n    infrastructure yet to handle that at all.\n\n    Fixes from Vlad Yasevich.\n\n 2) Fix tcpdump crash with bridging and vlans, also from Vlad.\n\n 3) Some MAINTAINERS updates for random32 and bonding.\n\n 4) Fix late reseeds of prandom generator, from Sasha Levin.\n\n 5) Bridge doesn\u0027t handle stacked vlans properly, fix from Toshiaki\n    Makita.\n\n 6) Fix deadlock in openvswitch, from Flavio Leitner.\n\n 7) get_timewait4_sock() doesn\u0027t report delay times correctly, fix from\n    Eric Dumazet.\n\n 8) Duplicate address detection and addrconf verification need to run in\n    contexts where RTNL can be obtained.  Move them to run from a\n    workqueue.  From Hannes Frederic Sowa.\n\n 9) Fix route refcount leaking in ip tunnels, from Pravin B Shelar.\n\n10) Don\u0027t return -EINTR from non-blocking recvmsg() on AF_UNIX sockets,\n    from Eric Dumazet.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (28 commits)\n  vlan: Warn the user if lowerdev has bad vlan features.\n  veth: Turn off vlan rx acceleration in vlan_features\n  ifb: Remove vlan acceleration from vlan_features\n  qlge: Do not propaged vlan tag offloads to vlans\n  bridge: Fix crash with vlan filtering and tcpdump\n  net: Account for all vlan headers in skb_mac_gso_segment\n  MAINTAINERS: bonding: change email address\n  MAINTAINERS: bonding: change email address\n  ipv6: move DAD and addrconf_verify processing to workqueue\n  tcp: fix get_timewait4_sock() delay computation on 64bit\n  openvswitch: fix a possible deadlock and lockdep warning\n  bridge: Fix handling stacked vlan tags\n  bridge: Fix inabillity to retrieve vlan tags when tx offload is disabled\n  vhost: validate vhost_get_vq_desc return value\n  vhost: fix total length when packets are too short\n  random32: avoid attempt to late reseed if in the middle of seeding\n  random32: assign to network folks in MAINTAINERS\n  net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset\n  core, nfqueue, openvswitch: Orphan frags in skb_zerocopy and handle errors\n  vlan: Set hard_header_len according to available acceleration\n  ...\n"
    },
    {
      "commit": "5f2feca20e0f4878fa16787bd50fcd82fb365be7",
      "tree": "1761471976beb97edab677ab211326b62a2809ad",
      "parents": [
        "fc92f745f8d0d3736ce5afb00a905d7cc61f9c46",
        "2adb956b084d6d49f519541a4b5f9947e96f8ef7"
      ],
      "author": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:17:16 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:17:16 2014 -0400"
      },
      "message": "Merge branch \u0027vlan_offloads\u0027\n\nVlad Yasevich says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nAudit all drivers for correct vlan_features.\n\nSome drivers set vlan acceleration features in vlan_features.  This causes\nissues with Q-in-Q/802.1ad configurations.\n\nAudit all the drivers for correct vlan_features.  Fix broken ones.\nAdd a warning to vlan code to help catch future offenders.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "2adb956b084d6d49f519541a4b5f9947e96f8ef7",
      "tree": "1761471976beb97edab677ab211326b62a2809ad",
      "parents": [
        "3f8c707b9a83cd956af65796081b6c8cb8716089"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Thu Mar 27 22:14:49 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:16:51 2014 -0400"
      },
      "message": "vlan: Warn the user if lowerdev has bad vlan features.\n\nSome drivers incorrectly assign vlan acceleration features to\nvlan_features thus causing issues for Q-in-Q vlan configurations.\nWarn the user of such cases.\n\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "3f8c707b9a83cd956af65796081b6c8cb8716089",
      "tree": "041ab3011deddf178994633d479e14cb3f9c037c",
      "parents": [
        "8dd6e147b0c29723ec10d0e836c7f3466d61a19b"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Thu Mar 27 22:14:48 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:16:51 2014 -0400"
      },
      "message": "veth: Turn off vlan rx acceleration in vlan_features\n\nFor completeness, turn off vlan rx acceleration in vlan_features so\nthat it doesn\u0027t show up on q-in-q setups.\n\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "8dd6e147b0c29723ec10d0e836c7f3466d61a19b",
      "tree": "0a0af7135b0d78b55399f544222c4b460d99609b",
      "parents": [
        "f6d1ac4b5f15f57929fe0fa283b3a45dfec717a0"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Thu Mar 27 22:14:47 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:16:50 2014 -0400"
      },
      "message": "ifb: Remove vlan acceleration from vlan_features\n\nDo not include vlan acceleration features in vlan_features as that\nprecludes correct Q-in-Q operation.\n\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "f6d1ac4b5f15f57929fe0fa283b3a45dfec717a0",
      "tree": "153e011f17d30c20afded26face72c79ecaf9c53",
      "parents": [
        "fc92f745f8d0d3736ce5afb00a905d7cc61f9c46"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Thu Mar 27 22:14:46 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:16:50 2014 -0400"
      },
      "message": "qlge: Do not propaged vlan tag offloads to vlans\n\nqlge driver turns off NETIF_F_HW_CTAG_FILTER, but forgets to\nturn off HW_CTAG_TX and HW_CTAG_RX on vlan devices.  With the\ncurrent settings, q-in-q will only generate a single vlan header.\nRemember to mask off CTAG_TX and CTAG_RX features in vlan_features.\n\nCC: Shahed Shaikh \u003cshahed.shaikh@qlogic.com\u003e\nCC: Jitendra Kalsaria \u003cjitendra.kalsaria@qlogic.com\u003e\nCC: Ron Mercer \u003cron.mercer@qlogic.com\u003e\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nAcked-by: Jitendra Kalsaria \u003cjitendra.kalsaria@qlogic.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "fc92f745f8d0d3736ce5afb00a905d7cc61f9c46",
      "tree": "606412dc5f865c5e2b11b6def1b75eca213aca30",
      "parents": [
        "53d6471cef17262d3ad1c7ce8982a234244f68ec"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Thu Mar 27 21:51:18 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:14:02 2014 -0400"
      },
      "message": "bridge: Fix crash with vlan filtering and tcpdump\n\nWhen the vlan filtering is enabled on the bridge, but\nthe filter is not configured on the bridge device itself,\nrunning tcpdump on the bridge device will result in a\nan Oops with NULL pointer dereference.  The reason\nis that br_pass_frame_up() will bypass the vlan\ncheck because promisc flag is set.  It will then try\nto get the table pointer and process the packet based\non the table.  Since the table pointer is NULL, we oops.\nCatch this special condition in br_handle_vlan().\n\nReported-by: Toshiaki Makita \u003cmakita.toshiaki@lab.ntt.co.jp\u003e\nCC: Toshiaki Makita \u003cmakita.toshiaki@lab.ntt.co.jp\u003e\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nAcked-by: Toshiaki Makita \u003cmakita.toshiaki@lab.ntt.co.jp\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "53d6471cef17262d3ad1c7ce8982a234244f68ec",
      "tree": "b7adb468b9fb7b500a96222742a3752545ff47b6",
      "parents": [
        "898602a049504dea256e696ee3152dc0b788a393"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Thu Mar 27 17:26:18 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 17:10:36 2014 -0400"
      },
      "message": "net: Account for all vlan headers in skb_mac_gso_segment\n\nskb_network_protocol() already accounts for multiple vlan\nheaders that may be present in the skb.  However, skb_mac_gso_segment()\ndoesn\u0027t know anything about it and assumes that skb-\u003emac_len\nis set correctly to skip all mac headers.  That may not\nalways be the case.  If we are simply forwarding the packet (via\nbridge or macvtap), all vlan headers may not be accounted for.\n\nA simple solution is to allow skb_network_protocol to return\nthe vlan depth it has calculated.  This way skb_mac_gso_segment\nwill correctly skip all mac headers.\n\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "898602a049504dea256e696ee3152dc0b788a393",
      "tree": "9e069b39c94e29318d17897eb913fd29f79884f5",
      "parents": [
        "79b30750d99a7f0fb4eaaa1c69139a7821bf6771"
      ],
      "author": {
        "name": "Veaceslav Falico",
        "email": "vfalico@redhat.com",
        "time": "Thu Mar 27 18:43:50 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:57:38 2014 -0400"
      },
      "message": "MAINTAINERS: bonding: change email address\n\nSigned-off-by: Veaceslav Falico \u003cvfalico@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "79b30750d99a7f0fb4eaaa1c69139a7821bf6771",
      "tree": "b5365cbc744017b14becf3b01d67ade979c2817c",
      "parents": [
        "c15b1ccadb323ea50023e8f1cca2954129a62b51"
      ],
      "author": {
        "name": "Jay Vosburgh",
        "email": "fubar@us.ibm.com",
        "time": "Thu Mar 27 10:33:44 2014 -0700"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:57:38 2014 -0400"
      },
      "message": "MAINTAINERS: bonding: change email address\n\nUpdate my email address.\n\nSigned-off-by: Jay Vosburgh \u003cfubar@us.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "bc53267e97e4942e7f22c31cd8cf87840796accb",
      "tree": "e343b4754bf55f5bd21c5863846b85e5939efda1",
      "parents": [
        "2946369ee277fa9fcc3372aabddc9c15dfabf744",
        "825600c0f20e595daaa7a6dd8970f84fa2a2ee57"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 13:57:13 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 13:57:13 2014 -0700"
      },
      "message": "Merge branch \u0027akpm\u0027 (patches from Andrew Morton)\n\nMerge two fixes from Andrew Morton:\n \"The x86 fix should come from x86 guys but they appear to be\n  conferencing or otherwise distracted.\n\n  The ocfs2 fix is a bit of a mess - the code runs into an immediate\n  NULL deref and we\u0027re trying to work out how this got through test and\n  review, but we haven\u0027t heard from Goldwyn in the past few days.\n  Sasha\u0027s patch fixes the oops, but the feature as a whole is probably\n  broken.  So this is a stopgap for 3.14 - I\u0027ll aim to get the real\n  fixes into 3.14.x\"\n\n* emailed patches from Andrew Morton akpm@linux-foundation.org\u003e:\n  x86: fix boot on uniprocessor systems\n  ocfs2: check if cluster name exists before deref\n"
    },
    {
      "commit": "825600c0f20e595daaa7a6dd8970f84fa2a2ee57",
      "tree": "ae739547c15bb979d412c6eced02796310b04afb",
      "parents": [
        "d9060742fbf630fe31951dfc10b798deb2813f01"
      ],
      "author": {
        "name": "Artem Fetishev",
        "email": "artem_fetishev@epam.com",
        "time": "Fri Mar 28 13:33:39 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 13:56:58 2014 -0700"
      },
      "message": "x86: fix boot on uniprocessor systems\n\nOn x86 uniprocessor systems topology_physical_package_id() returns -1\nwhich causes rapl_cpu_prepare() to leave rapl_pmu variable uninitialized\nwhich leads to GPF in rapl_pmu_init().\n\nSee arch/x86/kernel/cpu/perf_event_intel_rapl.c.\n\nIt turns out that physical_package_id and core_id can actually be\nretreived for uniprocessor systems too.  Enabling them also fixes\nrapl_pmu code.\n\nSigned-off-by: Artem Fetishev \u003cartem_fetishev@epam.com\u003e\nCc: Stephane Eranian \u003ceranian@google.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "d9060742fbf630fe31951dfc10b798deb2813f01",
      "tree": "a211f223fc34fa78e8b420fcd9e61e5de68eac73",
      "parents": [
        "1fac1fa965a9fbdb38d8a4b945db3521f801c946"
      ],
      "author": {
        "name": "Sasha Levin",
        "email": "sasha.levin@oracle.com",
        "time": "Fri Mar 28 13:33:38 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 13:56:58 2014 -0700"
      },
      "message": "ocfs2: check if cluster name exists before deref\n\nCommit c74a3bdd9b52 (\"ocfs2: add clustername to cluster connection\") is\ntrying to strlcpy a string which was explicitly passed as NULL in the\nvery same patch, triggering a NULL ptr deref.\n\n  BUG: unable to handle kernel NULL pointer dereference at           (null)\n  IP: strlcpy (lib/string.c:388 lib/string.c:151)\n  CPU: 19 PID: 19426 Comm: trinity-c19 Tainted: G        W     3.14.0-rc7-next-20140325-sasha-00014-g9476368-dirty #274\n  RIP:  strlcpy (lib/string.c:388 lib/string.c:151)\n  Call Trace:\n   ocfs2_cluster_connect (fs/ocfs2/stackglue.c:350)\n   ocfs2_cluster_connect_agnostic (fs/ocfs2/stackglue.c:396)\n   user_dlm_register (fs/ocfs2/dlmfs/userdlm.c:679)\n   dlmfs_mkdir (fs/ocfs2/dlmfs/dlmfs.c:503)\n   vfs_mkdir (fs/namei.c:3467)\n   SyS_mkdirat (fs/namei.c:3488 fs/namei.c:3472)\n   tracesys (arch/x86/kernel/entry_64.S:749)\n\nakpm: this patch probably disables the feature.  A temporary thing to\navoid triviel oopses.\n\nSigned-off-by: Sasha Levin \u003csasha.levin@oracle.com\u003e\nCc: Goldwyn Rodrigues \u003crgoldwyn@suse.com\u003e\nCc: Mark Fasheh \u003cmfasheh@suse.de\u003e\nCc: Joel Becker \u003cjlbec@evilplan.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "c15b1ccadb323ea50023e8f1cca2954129a62b51",
      "tree": "14c5c7e2fcb15546bda819ef76f2d22d5d9ea1d4",
      "parents": [
        "e2a1d3e47bb904082b758dec9d07edf241c45d05"
      ],
      "author": {
        "name": "Hannes Frederic Sowa",
        "email": "hannes@stressinduktion.org",
        "time": "Thu Mar 27 18:28:07 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:54:50 2014 -0400"
      },
      "message": "ipv6: move DAD and addrconf_verify processing to workqueue\n\naddrconf_join_solict and addrconf_join_anycast may cause actions which\nneed rtnl locked, especially on first address creation.\n\nA new DAD state is introduced which defers processing of the initial\nDAD processing into a workqueue.\n\nTo get rtnl lock we need to push the code paths which depend on those\ncalls up to workqueues, specifically addrconf_verify and the DAD\nprocessing.\n\n(v2)\naddrconf_dad_failure needs to be queued up to the workqueue, too. This\npatch introduces a new DAD state and stop the DAD processing in the\nworkqueue (this is because of the possible ipv6_del_addr processing\nwhich removes the solicited multicast address from the device).\n\naddrconf_verify_lock is removed, too. After the transition it is not\nneeded any more.\n\nAs we are not processing in bottom half anymore we need to be a bit more\ncareful about disabling bottom half out when we lock spin_locks which are also\nused in bh.\n\nRelevant backtrace:\n[  541.030090] RTNL: assertion failed at net/core/dev.c (4496)\n[  541.031143] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G           O 3.10.33-1-amd64-vyatta #1\n[  541.031145] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007\n[  541.031146]  ffffffff8148a9f0 000000000000002f ffffffff813c98c1 ffff88007c4451f8\n[  541.031148]  0000000000000000 0000000000000000 ffffffff813d3540 ffff88007fc03d18\n[  541.031150]  0000880000000006 ffff88007c445000 ffffffffa0194160 0000000000000000\n[  541.031152] Call Trace:\n[  541.031153]  \u003cIRQ\u003e  [\u003cffffffff8148a9f0\u003e] ? dump_stack+0xd/0x17\n[  541.031180]  [\u003cffffffff813c98c1\u003e] ? __dev_set_promiscuity+0x101/0x180\n[  541.031183]  [\u003cffffffff813d3540\u003e] ? __hw_addr_create_ex+0x60/0xc0\n[  541.031185]  [\u003cffffffff813cfe1a\u003e] ? __dev_set_rx_mode+0xaa/0xc0\n[  541.031189]  [\u003cffffffff813d3a81\u003e] ? __dev_mc_add+0x61/0x90\n[  541.031198]  [\u003cffffffffa01dcf9c\u003e] ? igmp6_group_added+0xfc/0x1a0 [ipv6]\n[  541.031208]  [\u003cffffffff8111237b\u003e] ? kmem_cache_alloc+0xcb/0xd0\n[  541.031212]  [\u003cffffffffa01ddcd7\u003e] ? ipv6_dev_mc_inc+0x267/0x300 [ipv6]\n[  541.031216]  [\u003cffffffffa01c2fae\u003e] ? addrconf_join_solict+0x2e/0x40 [ipv6]\n[  541.031219]  [\u003cffffffffa01ba2e9\u003e] ? ipv6_dev_ac_inc+0x159/0x1f0 [ipv6]\n[  541.031223]  [\u003cffffffffa01c0772\u003e] ? addrconf_join_anycast+0x92/0xa0 [ipv6]\n[  541.031226]  [\u003cffffffffa01c311e\u003e] ? __ipv6_ifa_notify+0x11e/0x1e0 [ipv6]\n[  541.031229]  [\u003cffffffffa01c3213\u003e] ? ipv6_ifa_notify+0x33/0x50 [ipv6]\n[  541.031233]  [\u003cffffffffa01c36c8\u003e] ? addrconf_dad_completed+0x28/0x100 [ipv6]\n[  541.031241]  [\u003cffffffff81075c1d\u003e] ? task_cputime+0x2d/0x50\n[  541.031244]  [\u003cffffffffa01c38d6\u003e] ? addrconf_dad_timer+0x136/0x150 [ipv6]\n[  541.031247]  [\u003cffffffffa01c37a0\u003e] ? addrconf_dad_completed+0x100/0x100 [ipv6]\n[  541.031255]  [\u003cffffffff8105313a\u003e] ? call_timer_fn.isra.22+0x2a/0x90\n[  541.031258]  [\u003cffffffffa01c37a0\u003e] ? addrconf_dad_completed+0x100/0x100 [ipv6]\n\nHunks and backtrace stolen from a patch by Stephen Hemminger.\n\nReported-by: Stephen Hemminger \u003cstephen@networkplumber.org\u003e\nSigned-off-by: Stephen Hemminger \u003cstephen@networkplumber.org\u003e\nSigned-off-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "e2a1d3e47bb904082b758dec9d07edf241c45d05",
      "tree": "5ed95c32972d85fd1e452ca1572c261ad7e94f0e",
      "parents": [
        "4f647e0a3c37b8d5086214128614a136064110c3"
      ],
      "author": {
        "name": "Eric Dumazet",
        "email": "edumazet@google.com",
        "time": "Thu Mar 27 07:19:19 2014 -0700"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:43:14 2014 -0400"
      },
      "message": "tcp: fix get_timewait4_sock() delay computation on 64bit\n\nIt seems I missed one change in get_timewait4_sock() to compute\nthe remaining time before deletion of IPV4 timewait socket.\n\nThis could result in wrong output in /proc/net/tcp for tm-\u003ewhen field.\n\nFixes: 96f817fedec4 (\"tcp: shrink tcp6_timewait_sock by one cache line\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "4f647e0a3c37b8d5086214128614a136064110c3",
      "tree": "32b03e568ed7e2c381de3b709a50161fbb3e6fe0",
      "parents": [
        "99b192da9c99284ad3374132e56f66995cadc6b4"
      ],
      "author": {
        "name": "Flavio Leitner",
        "email": "fbl@redhat.com",
        "time": "Thu Mar 27 11:05:34 2014 -0300"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:41:53 2014 -0400"
      },
      "message": "openvswitch: fix a possible deadlock and lockdep warning\n\nThere are two problematic situations.\n\nA deadlock can happen when is_percpu is false because it can get\ninterrupted while holding the spinlock. Then it executes\novs_flow_stats_update() in softirq context which tries to get\nthe same lock.\n\nThe second sitation is that when is_percpu is true, the code\ncorrectly disables BH but only for the local CPU, so the\nfollowing can happen when locking the remote CPU without\ndisabling BH:\n\n       CPU#0                            CPU#1\n  ovs_flow_stats_get()\n   stats_read()\n +-\u003espin_lock remote CPU#1        ovs_flow_stats_get()\n |  \u003cinterrupted\u003e                  stats_read()\n |  ...                       +--\u003e  spin_lock remote CPU#0\n |                            |     \u003cinterrupted\u003e\n |  ovs_flow_stats_update()   |     ...\n |   spin_lock local CPU#0 \u003c--+     ovs_flow_stats_update()\n +---------------------------------- spin_lock local CPU#1\n\nThis patch disables BH for both cases fixing the deadlocks.\nAcked-by: Jesse Gross \u003cjesse@nicira.com\u003e\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ INFO: inconsistent lock state ]\n3.14.0-rc8-00007-g632b06a #1 Tainted: G          I\n---------------------------------\ninconsistent {SOFTIRQ-ON-W} -\u003e {IN-SOFTIRQ-W} usage.\nswapper/0/0 [HC0[0]:SC1[5]:HE1:SE0] takes:\n(\u0026(\u0026cpu_stats-\u003elock)-\u003erlock){+.?...}, at: [\u003cffffffffa05dd8a1\u003e] ovs_flow_stats_update+0x51/0xd0 [openvswitch]\n{SOFTIRQ-ON-W} state was registered at:\n[\u003cffffffff810f973f\u003e] __lock_acquire+0x68f/0x1c40\n[\u003cffffffff810fb4e2\u003e] lock_acquire+0xa2/0x1d0\n[\u003cffffffff817d8d9e\u003e] _raw_spin_lock+0x3e/0x80\n[\u003cffffffffa05dd9e4\u003e] ovs_flow_stats_get+0xc4/0x1e0 [openvswitch]\n[\u003cffffffffa05da855\u003e] ovs_flow_cmd_fill_info+0x185/0x360 [openvswitch]\n[\u003cffffffffa05daf05\u003e] ovs_flow_cmd_build_info.constprop.27+0x55/0x90 [openvswitch]\n[\u003cffffffffa05db41d\u003e] ovs_flow_cmd_new_or_set+0x4dd/0x570 [openvswitch]\n[\u003cffffffff816c245d\u003e] genl_family_rcv_msg+0x1cd/0x3f0\n[\u003cffffffff816c270e\u003e] genl_rcv_msg+0x8e/0xd0\n[\u003cffffffff816c0239\u003e] netlink_rcv_skb+0xa9/0xc0\n[\u003cffffffff816c0798\u003e] genl_rcv+0x28/0x40\n[\u003cffffffff816bf830\u003e] netlink_unicast+0x100/0x1e0\n[\u003cffffffff816bfc57\u003e] netlink_sendmsg+0x347/0x770\n[\u003cffffffff81668e9c\u003e] sock_sendmsg+0x9c/0xe0\n[\u003cffffffff816692d9\u003e] ___sys_sendmsg+0x3a9/0x3c0\n[\u003cffffffff8166a911\u003e] __sys_sendmsg+0x51/0x90\n[\u003cffffffff8166a962\u003e] SyS_sendmsg+0x12/0x20\n[\u003cffffffff817e3ce9\u003e] system_call_fastpath+0x16/0x1b\nirq event stamp: 1740726\nhardirqs last  enabled at (1740726): [\u003cffffffff8175d5e0\u003e] ip6_finish_output2+0x4f0/0x840\nhardirqs last disabled at (1740725): [\u003cffffffff8175d59b\u003e] ip6_finish_output2+0x4ab/0x840\nsoftirqs last  enabled at (1740674): [\u003cffffffff8109be12\u003e] _local_bh_enable+0x22/0x50\nsoftirqs last disabled at (1740675): [\u003cffffffff8109db05\u003e] irq_exit+0xc5/0xd0\n\nother info that might help us debug this:\n Possible unsafe locking scenario:\n\n       CPU0\n       ----\n  lock(\u0026(\u0026cpu_stats-\u003elock)-\u003erlock);\n  \u003cInterrupt\u003e\n    lock(\u0026(\u0026cpu_stats-\u003elock)-\u003erlock);\n\n *** DEADLOCK ***\n\n5 locks held by swapper/0/0:\n #0:  (((\u0026ifa-\u003edad_timer))){+.-...}, at: [\u003cffffffff810a7155\u003e] call_timer_fn+0x5/0x320\n #1:  (rcu_read_lock){.+.+..}, at: [\u003cffffffff81788a55\u003e] mld_sendpack+0x5/0x4a0\n #2:  (rcu_read_lock_bh){.+....}, at: [\u003cffffffff8175d149\u003e] ip6_finish_output2+0x59/0x840\n #3:  (rcu_read_lock_bh){.+....}, at: [\u003cffffffff8168ba75\u003e] __dev_queue_xmit+0x5/0x9b0\n #4:  (rcu_read_lock){.+.+..}, at: [\u003cffffffffa05e41b5\u003e] internal_dev_xmit+0x5/0x110 [openvswitch]\n\nstack backtrace:\nCPU: 0 PID: 0 Comm: swapper/0 Tainted: G          I  3.14.0-rc8-00007-g632b06a #1\nHardware name:                  /DX58SO, BIOS SOX5810J.86A.5599.2012.0529.2218 05/29/2012\n 0000000000000000 0fcf20709903df0c ffff88042d603808 ffffffff817cfe3c\n ffffffff81c134c0 ffff88042d603858 ffffffff817cb6da 0000000000000005\n ffffffff00000001 ffff880400000000 0000000000000006 ffffffff81c134c0\nCall Trace:\n \u003cIRQ\u003e  [\u003cffffffff817cfe3c\u003e] dump_stack+0x4d/0x66\n [\u003cffffffff817cb6da\u003e] print_usage_bug+0x1f4/0x205\n [\u003cffffffff810f7f10\u003e] ? check_usage_backwards+0x180/0x180\n [\u003cffffffff810f8963\u003e] mark_lock+0x223/0x2b0\n [\u003cffffffff810f96d3\u003e] __lock_acquire+0x623/0x1c40\n [\u003cffffffff810f5707\u003e] ? __lock_is_held+0x57/0x80\n [\u003cffffffffa05e26c6\u003e] ? masked_flow_lookup+0x236/0x250 [openvswitch]\n [\u003cffffffff810fb4e2\u003e] lock_acquire+0xa2/0x1d0\n [\u003cffffffffa05dd8a1\u003e] ? ovs_flow_stats_update+0x51/0xd0 [openvswitch]\n [\u003cffffffff817d8d9e\u003e] _raw_spin_lock+0x3e/0x80\n [\u003cffffffffa05dd8a1\u003e] ? ovs_flow_stats_update+0x51/0xd0 [openvswitch]\n [\u003cffffffffa05dd8a1\u003e] ovs_flow_stats_update+0x51/0xd0 [openvswitch]\n [\u003cffffffffa05dcc64\u003e] ovs_dp_process_received_packet+0x84/0x120 [openvswitch]\n [\u003cffffffff810f93f7\u003e] ? __lock_acquire+0x347/0x1c40\n [\u003cffffffffa05e3bea\u003e] ovs_vport_receive+0x2a/0x30 [openvswitch]\n [\u003cffffffffa05e4218\u003e] internal_dev_xmit+0x68/0x110 [openvswitch]\n [\u003cffffffffa05e41b5\u003e] ? internal_dev_xmit+0x5/0x110 [openvswitch]\n [\u003cffffffff8168b4a6\u003e] dev_hard_start_xmit+0x2e6/0x8b0\n [\u003cffffffff8168be87\u003e] __dev_queue_xmit+0x417/0x9b0\n [\u003cffffffff8168ba75\u003e] ? __dev_queue_xmit+0x5/0x9b0\n [\u003cffffffff8175d5e0\u003e] ? ip6_finish_output2+0x4f0/0x840\n [\u003cffffffff8168c430\u003e] dev_queue_xmit+0x10/0x20\n [\u003cffffffff8175d641\u003e] ip6_finish_output2+0x551/0x840\n [\u003cffffffff8176128a\u003e] ? ip6_finish_output+0x9a/0x220\n [\u003cffffffff8176128a\u003e] ip6_finish_output+0x9a/0x220\n [\u003cffffffff8176145f\u003e] ip6_output+0x4f/0x1f0\n [\u003cffffffff81788c29\u003e] mld_sendpack+0x1d9/0x4a0\n [\u003cffffffff817895b8\u003e] mld_send_initial_cr.part.32+0x88/0xa0\n [\u003cffffffff817691b0\u003e] ? addrconf_dad_completed+0x220/0x220\n [\u003cffffffff8178e301\u003e] ipv6_mc_dad_complete+0x31/0x50\n [\u003cffffffff817690d7\u003e] addrconf_dad_completed+0x147/0x220\n [\u003cffffffff817691b0\u003e] ? addrconf_dad_completed+0x220/0x220\n [\u003cffffffff8176934f\u003e] addrconf_dad_timer+0x19f/0x1c0\n [\u003cffffffff810a71e9\u003e] call_timer_fn+0x99/0x320\n [\u003cffffffff810a7155\u003e] ? call_timer_fn+0x5/0x320\n [\u003cffffffff817691b0\u003e] ? addrconf_dad_completed+0x220/0x220\n [\u003cffffffff810a76c4\u003e] run_timer_softirq+0x254/0x3b0\n [\u003cffffffff8109d47d\u003e] __do_softirq+0x12d/0x480\n\nSigned-off-by: Flavio Leitner \u003cfbl@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "99b192da9c99284ad3374132e56f66995cadc6b4",
      "tree": "5d76b7b88909e15e5126fb250b3ad03d6ecb0d15",
      "parents": [
        "12464bb8de021a01fa7ec9299c273c247df7f198"
      ],
      "author": {
        "name": "Toshiaki Makita",
        "email": "makita.toshiaki@lab.ntt.co.jp",
        "time": "Thu Mar 27 21:46:56 2014 +0900"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:33:09 2014 -0400"
      },
      "message": "bridge: Fix handling stacked vlan tags\n\nIf a bridge with vlan_filtering enabled receives frames with stacked\nvlan tags, i.e., they have two vlan tags, br_vlan_untag() strips not\nonly the outer tag but also the inner tag.\n\nbr_vlan_untag() is called only from br_handle_vlan(), and in this case,\nit is enough to set skb-\u003evlan_tci to 0 here, because vlan_tci has already\nbeen set before calling br_handle_vlan().\n\nSigned-off-by: Toshiaki Makita \u003cmakita.toshiaki@lab.ntt.co.jp\u003e\nAcked-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "12464bb8de021a01fa7ec9299c273c247df7f198",
      "tree": "c4fb3f4b93f4571483327501011be6668ceb3a76",
      "parents": [
        "a39ee449f96a2cd44ce056d8a0a112211a9b1a1f"
      ],
      "author": {
        "name": "Toshiaki Makita",
        "email": "makita.toshiaki@lab.ntt.co.jp",
        "time": "Thu Mar 27 21:46:55 2014 +0900"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:33:09 2014 -0400"
      },
      "message": "bridge: Fix inabillity to retrieve vlan tags when tx offload is disabled\n\nBridge vlan code (br_vlan_get_tag()) assumes that all frames have vlan_tci\nif they are tagged, but if vlan tx offload is manually disabled on bridge\ndevice and frames are sent from vlan device on the bridge device, the tags\nare embedded in skb-\u003edata and they break this assumption.\nExtract embedded vlan tags and move them to vlan_tci at ingress.\n\nSigned-off-by: Toshiaki Makita \u003cmakita.toshiaki@lab.ntt.co.jp\u003e\nAcked-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "a39ee449f96a2cd44ce056d8a0a112211a9b1a1f",
      "tree": "0a28633b16e59b3ce6211daf4f34466ad0129430",
      "parents": [
        "d8316f3991d207fe32881a9ac20241be8fa2bad0"
      ],
      "author": {
        "name": "Michael S. Tsirkin",
        "email": "mst@redhat.com",
        "time": "Thu Mar 27 12:53:37 2014 +0200"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:10:35 2014 -0400"
      },
      "message": "vhost: validate vhost_get_vq_desc return value\n\nvhost fails to validate negative error code\nfrom vhost_get_vq_desc causing\na crash: we are using -EFAULT which is 0xfffffff2\nas vector size, which exceeds the allocated size.\n\nThe code in question was introduced in commit\n8dd014adfea6f173c1ef6378f7e5e7924866c923\n    vhost-net: mergeable buffers support\n\nCVE-2014-0055\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "d8316f3991d207fe32881a9ac20241be8fa2bad0",
      "tree": "76802d9368e0115d9c53ac7cf0477b53011a4486",
      "parents": [
        "05efa8c943b1d5d90fa8c8147571837573338bb6"
      ],
      "author": {
        "name": "Michael S. Tsirkin",
        "email": "mst@redhat.com",
        "time": "Thu Mar 27 12:00:26 2014 +0200"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:07:31 2014 -0400"
      },
      "message": "vhost: fix total length when packets are too short\n\nWhen mergeable buffers are disabled, and the\nincoming packet is too large for the rx buffer,\nget_rx_bufs returns success.\n\nThis was intentional in order for make recvmsg\ntruncate the packet and then handle_rx would\ndetect err !\u003d sock_len and drop it.\n\nUnfortunately we pass the original sock_len to\nrecvmsg - which means we use parts of iov not fully\nvalidated.\n\nFix this up by detecting this overrun and doing packet drop\nimmediately.\n\nCVE-2014-0077\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "05efa8c943b1d5d90fa8c8147571837573338bb6",
      "tree": "7882312770611805d896178894810f0d9467469b",
      "parents": [
        "335a67d2ad481b03607bf30a38c28178fa1ad61a"
      ],
      "author": {
        "name": "Sasha Levin",
        "email": "sasha.levin@oracle.com",
        "time": "Fri Mar 28 17:38:42 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:04:19 2014 -0400"
      },
      "message": "random32: avoid attempt to late reseed if in the middle of seeding\n\nCommit 4af712e8df (\"random32: add prandom_reseed_late() and call when\nnonblocking pool becomes initialized\") has added a late reseed stage\nthat happens as soon as the nonblocking pool is marked as initialized.\n\nThis fails in the case that the nonblocking pool gets initialized\nduring __prandom_reseed()\u0027s call to get_random_bytes(). In that case\nwe\u0027d double back into __prandom_reseed() in an attempt to do a late\nreseed - deadlocking on \u0027lock\u0027 early on in the boot process.\n\nInstead, just avoid even waiting to do a reseed if a reseed is already\noccuring.\n\nFixes: 4af712e8df99 (\"random32: add prandom_reseed_late() and call when nonblocking pool becomes initialized\")\nSigned-off-by: Sasha Levin \u003csasha.levin@oracle.com\u003e\nAcked-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdborkman@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "2946369ee277fa9fcc3372aabddc9c15dfabf744",
      "tree": "f6aa776b63db2a5352cd05a3ad2437e212f72885",
      "parents": [
        "1fac1fa965a9fbdb38d8a4b945db3521f801c946",
        "8a0435d958fb36d93b8df610124a0e91e5675c82"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 13:03:00 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 13:03:00 2014 -0700"
      },
      "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input\n\nPull input subsystem fixes from Dmitry Torokhov:\n \"Updates to Synaptics touchpad to better cope with devices in Lenovo\n  laptops, and a couple more fixes\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:\n  Input: synaptics - add manual min/max quirk for ThinkPad X240\n  Input: synaptics - add manual min/max quirk\n  Input: cypress_ps2 - don\u0027t report as a button pads\n  Input: da9052_onkey - use correct register bit for key status\n  Input: adp5588-keys - get value from data out when dir is out\n"
    },
    {
      "commit": "335a67d2ad481b03607bf30a38c28178fa1ad61a",
      "tree": "42fc172eb17aa9adfacdb676ec1991579f4ed99f",
      "parents": [
        "97a5221f56bad2e1c7e8ab55da4ac4748ef59c64"
      ],
      "author": {
        "name": "Sasha Levin",
        "email": "sasha.levin@oracle.com",
        "time": "Thu Mar 27 02:01:34 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Mar 28 16:03:00 2014 -0400"
      },
      "message": "random32: assign to network folks in MAINTAINERS\n\nlib/random32.c was split out of the network code and is de-facto\nstill maintained by the almighty net/ gods.\n\nMake it a bit more official so that people who aren\u0027t aware of\nthat know where to send their patches.\n\nSigned-off-by: Sasha Levin \u003csasha.levin@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "1fac1fa965a9fbdb38d8a4b945db3521f801c946",
      "tree": "26a1152acc6faaca6427471a3934255812bbcf38",
      "parents": [
        "350bb4bebf5d80937eb41b39509779888bc8aa75",
        "8ee661b505613ef2747b350ca2871a31b3781bee"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 10:58:10 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 10:58:10 2014 -0700"
      },
      "message": "Merge branch \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes from Dave Airlie:\n \"I didn\u0027t want these to wait for stable cycle.\n\n  The nouveau and radeon ones are the same problem, where the runtime pm\n  stuff broke non-runtime pm managed secondary GPUs.\n\n  The udl fix is for an oops on unplug, and the i915 fix is for a\n  regression on Sandybridge even though it may break haswell (regression\n  wins)\"\n\nDaniel Vetter comments:\n \"My apologies for the i915 regression fumble, that thing somehow fell\n  through the cracks here for almost half a year :( Imo that\u0027s more than\n  enough flailing to just go ahead with the revert, and the re-broken\n  hsw should get peoples attention ...\"\n\n* \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux:\n  drm/i915: Undo gtt scratch pte unmapping again\n  drm/radeon: fix runtime suspend breaking secondary GPUs\n  drm/nouveau: fail runtime pm properly.\n  drm/udl: take reference to device struct for dma-bufs\n"
    },
    {
      "commit": "350bb4bebf5d80937eb41b39509779888bc8aa75",
      "tree": "e4d53867511a508ec8363725fa0c976b1d554520",
      "parents": [
        "81250437a55044edf16d2cc484da4fb837489317",
        "5f12c5eca6e6b7aeb4b2028d579f614b4fe7a81f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 10:55:44 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 10:55:44 2014 -0700"
      },
      "message": "Merge branch \u0027i2c/for-current\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux\n\nPull i2c build fix from Wolfram Sang:\n \"The build fix from my last request unveiled another build problem\n  which is fixed with this patch\"\n\n* \u0027i2c/for-current\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:\n  i2c: cpm: Fix build by adding of_address.h and of_irq.h\n"
    },
    {
      "commit": "81250437a55044edf16d2cc484da4fb837489317",
      "tree": "07f12a763928ed767fd39be4c6d35bfc28d30aea",
      "parents": [
        "75c5a52da3fc2a06abb6c6192bdf5d680e56d37d",
        "5926f87fdaad4be3ed10cec563bf357915e55a86"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 10:52:05 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Mar 28 10:52:05 2014 -0700"
      },
      "message": "Merge tag \u0027stable/for-linus-3.14-rc8-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip\n\nPull Xen bugfixes from David Vrabel:\n \"Fix two bugs that cause x86 PV guest crashes.\n\n  1. Ballooning a 32-bit guest would eventually crash it.\n\n  2. Revert a broken fix for a regression with NUMA_BALACING.  The bad\n     fix caused PV guests to crash after migration.  This is not ideal\n     but unpicking the madness that is _PAGE_NUMA \u003d\u003d _PAGE_PROTNONE will\n     take a while longer\"\n\n* tag \u0027stable/for-linus-3.14-rc8-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:\n  Revert \"xen: properly account for _PAGE_NUMA during xen pte translations\"\n  xen/balloon: flush persistent kmaps in correct position\n"
    },
    {
      "commit": "8a0435d958fb36d93b8df610124a0e91e5675c82",
      "tree": "d7624c8e4af79a747de51a650cb35f95e61d4050",
      "parents": [
        "421e08c41fda1f0c2ff6af81a67b491389b653a5"
      ],
      "author": {
        "name": "Hans de Goede",
        "email": "hdegoede@redhat.com",
        "time": "Fri Mar 28 01:01:38 2014 -0700"
      },
      "committer": {
        "name": "Dmitry Torokhov",
        "email": "dmitry.torokhov@gmail.com",
        "time": "Fri Mar 28 09:02:01 2014 -0700"
      },
      "message": "Input: synaptics - add manual min/max quirk for ThinkPad X240\n\nThis extends Benjamin Tissoires manual min/max quirk table with support for\nthe ThinkPad X240.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nSigned-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n"
    },
    {
      "commit": "421e08c41fda1f0c2ff6af81a67b491389b653a5",
      "tree": "ed7512d06b79fb0f47fa1c4dfba86db2b6f9cdef",
      "parents": [
        "6797b39e6f6f34c74177736e146406e894b9482b"
      ],
      "author": {
        "name": "Benjamin Tissoires",
        "email": "benjamin.tissoires@redhat.com",
        "time": "Fri Mar 28 00:43:00 2014 -0700"
      },
      "committer": {
        "name": "Dmitry Torokhov",
        "email": "dmitry.torokhov@gmail.com",
        "time": "Fri Mar 28 09:02:00 2014 -0700"
      },
      "message": "Input: synaptics - add manual min/max quirk\n\nThe new Lenovo Haswell series (-40\u0027s) contains a new Synaptics touchpad.\nHowever, these new Synaptics devices report bad axis ranges.\nUnder Windows, it is not a problem because the Windows driver uses RMI4\nover SMBus to talk to the device. Under Linux, we are using the PS/2\nfallback interface and it occurs the reported ranges are wrong.\n\nOf course, it would be too easy to have only one range for the whole\nseries, each touchpad seems to be calibrated in a different way.\n\nWe can not use SMBus to get the actual range because I suspect the firmware\nwill switch into the SMBus mode and stop talking through PS/2 (this is the\ncase for hybrid HID over I2C / PS/2 Synaptics touchpads).\n\nSo as a temporary solution (until RMI4 land into upstream), start a new\nlist of quirks with the min/max manually set.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nCC: stable@vger.kernel.org\nSigned-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n"
    },
    {
      "commit": "cab5e127eef040399902caa8e1510795583fa03a",
      "tree": "9e9a5713d4d58ffafea24354537e8207bd013c7e",
      "parents": [
        "75c5a52da3fc2a06abb6c6192bdf5d680e56d37d"
      ],
      "author": {
        "name": "John Stultz",
        "email": "john.stultz@linaro.org",
        "time": "Thu Mar 27 16:30:49 2014 -0700"
      },
      "committer": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Fri Mar 28 08:07:07 2014 +0100"
      },
      "message": "time: Revert to calling clock_was_set_delayed() while in irq context\n\nIn commit 47a1b796306356f35 (\"tick/timekeeping: Call\nupdate_wall_time outside the jiffies lock\"), we moved to calling\nclock_was_set() due to the fact that we were no longer holding\nthe timekeeping or jiffies lock.\n\nHowever, there is still the problem that clock_was_set()\ntriggers an IPI, which cannot be done from the timer\u0027s hard irq\ncontext, and will generate WARN_ON warnings.\n\nApparently in my earlier testing, I\u0027m guessing I didn\u0027t bump the\ndmesg log level, so I somehow missed the WARN_ONs.\n\nThus we need to revert back to calling clock_was_set_delayed().\n\nSigned-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: http://lkml.kernel.org/r/1395963049-11923-1-git-send-email-john.stultz@linaro.org\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n"
    },
    {
      "commit": "8ee661b505613ef2747b350ca2871a31b3781bee",
      "tree": "b000442236a2a6316911ae095a9ca3ae785793e9",
      "parents": [
        "c53c6beddc0ebb6feb5d64488c2c0cd23dd7f43d"
      ],
      "author": {
        "name": "Daniel Vetter",
        "email": "daniel.vetter@ffwll.ch",
        "time": "Wed Mar 26 20:10:09 2014 +0100"
      },
      "committer": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Fri Mar 28 12:33:50 2014 +1000"
      },
      "message": "drm/i915: Undo gtt scratch pte unmapping again\n\nIt apparently blows up on some machines. This functionally reverts\n\ncommit 828c79087cec61eaf4c76bb32c222fbe35ac3930\nAuthor: Ben Widawsky \u003cbenjamin.widawsky@intel.com\u003e\nDate:   Wed Oct 16 09:21:30 2013 -0700\n\n    drm/i915: Disable GGTT PTEs on GEN6+ suspend\n\nBugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d64841\nReported-and-Tested-by: Brad  Jackson \u003cbjackson0971@gmail.com\u003e\nCc: stable@vger.kernel.org\nCc: Takashi Iwai \u003ctiwai@suse.de\u003e\nCc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\nCc: Todd Previte \u003ctprevite@gmail.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n"
    },
    {
      "commit": "c53c6beddc0ebb6feb5d64488c2c0cd23dd7f43d",
      "tree": "cf467d9de5d6cbc67d3966588595d232dc8c0252",
      "parents": [
        "adbbdbac04f093c0abf946b1e93e4e5291808491"
      ],
      "author": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Thu Mar 27 02:31:08 2014 +0000"
      },
      "committer": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Fri Mar 28 12:27:25 2014 +1000"
      },
      "message": "drm/radeon: fix runtime suspend breaking secondary GPUs\n\nSame fix as for nouveau, when we fail with EINVAL, subsequent\ngets fail hard, causing the device not to open.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n"
    },
    {
      "commit": "97a5221f56bad2e1c7e8ab55da4ac4748ef59c64",
      "tree": "0762600e19c196843f868d158a44aa63b0fdc0f4",
      "parents": [
        "36d5fe6a000790f56039afe26834265db0a3ad4c"
      ],
      "author": {
        "name": "Wei Yang",
        "email": "weiyang@linux.vnet.ibm.com",
        "time": "Thu Mar 27 09:28:31 2014 +0800"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 27 15:35:33 2014 -0400"
      },
      "message": "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset\n\nThe second parameter of __mlx4_init_one() is used to identify whether the\npci_dev is a PF or VF. Currently, when it is invoked in mlx4_pci_slot_reset()\nthis information is missed.\n\nThis patch match the pci_dev with mlx4_pci_table and passes the\npci_device_id.driver_data to __mlx4_init_one() in mlx4_pci_slot_reset().\n\nSigned-off-by: Wei Yang \u003cweiyang@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "36d5fe6a000790f56039afe26834265db0a3ad4c",
      "tree": "391de83f18ede9ca651ed347112ce4400e4645d5",
      "parents": [
        "fc0d48b8fb449ca007b2057328abf736cb516168"
      ],
      "author": {
        "name": "Zoltan Kiss",
        "email": "zoltan.kiss@citrix.com",
        "time": "Wed Mar 26 22:37:45 2014 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 27 15:29:38 2014 -0400"
      },
      "message": "core, nfqueue, openvswitch: Orphan frags in skb_zerocopy and handle errors\n\nskb_zerocopy can copy elements of the frags array between skbs, but it doesn\u0027t\norphan them. Also, it doesn\u0027t handle errors, so this patch takes care of that\nas well, and modify the callers accordingly. skb_tx_error() is also added to\nthe callers so they will signal the failed delivery towards the creator of the\nskb.\n\nSigned-off-by: Zoltan Kiss \u003czoltan.kiss@citrix.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "fc0d48b8fb449ca007b2057328abf736cb516168",
      "tree": "9c0a39d0db9ee11261c6261035ff355157720668",
      "parents": [
        "14a0d635d18d0fb552dcc979d6d25106e6541f2e"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Wed Mar 26 11:47:56 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 27 15:00:37 2014 -0400"
      },
      "message": "vlan: Set hard_header_len according to available acceleration\n\nCurrently, if the card supports CTAG acceleration we do not\naccount for the vlan header even if we are configuring an\n8021AD vlan.  This may not be best since we\u0027ll do software\ntagging for 8021AD which will cause data copy on skb head expansion\nConfigure the length based on available hw offload capabilities and\nvlan protocol.\n\nCC: Patrick McHardy \u003ckaber@trash.net\u003e\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "14a0d635d18d0fb552dcc979d6d25106e6541f2e",
      "tree": "2fa1bce86903a0f841ab662eb85de4445bbfe398",
      "parents": [
        "681daee2443291419c57cccb0671f5f94a839005"
      ],
      "author": {
        "name": "Oliver Neukum",
        "email": "oneukum@suse.de",
        "time": "Wed Mar 26 14:32:51 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 27 14:59:10 2014 -0400"
      },
      "message": "usbnet: include wait queue head in device structure\n\nThis fixes a race which happens by freeing an object on the stack.\nQuoting Julius:\n\u003e The issue is\n\u003e that it calls usbnet_terminate_urbs() before that, which temporarily\n\u003e installs a waitqueue in dev-\u003ewait in order to be able to wait on the\n\u003e tasklet to run and finish up some queues. The waiting itself looks\n\u003e okay, but the access to \u0027dev-\u003ewait\u0027 is totally unprotected and can\n\u003e race arbitrarily. I think in this case usbnet_bh() managed to succeed\n\u003e it\u0027s dev-\u003ewait check just before usbnet_terminate_urbs() sets it back\n\u003e to NULL. The latter then finishes and the waitqueue_t structure on its\n\u003e stack gets overwritten by other functions halfway through the\n\u003e wake_up() call in usbnet_bh().\n\nThe fix is to just not allocate the data structure on the stack.\nAs dev-\u003ewait is abused as a flag it also takes a runtime PM change\nto fix this bug.\n\nSigned-off-by: Oliver Neukum \u003coneukum@suse.de\u003e\nReported-by: Grant Grundler \u003cgrundler@google.com\u003e\nTested-by: Grant Grundler \u003cgrundler@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "681daee2443291419c57cccb0671f5f94a839005",
      "tree": "e6c47044894798c2b838e535bbca0d412a4ce3c6",
      "parents": [
        "de1443916791d75fdd26becb116898277bb0273f"
      ],
      "author": {
        "name": "Jason Wang",
        "email": "jasowang@redhat.com",
        "time": "Wed Mar 26 13:03:00 2014 +0800"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 27 13:13:21 2014 -0400"
      },
      "message": "virtio-net: correct error handling of virtqueue_kick()\n\nCurrent error handling of virtqueue_kick() was wrong in two places:\n- The skb were freed immediately when virtqueue_kick() fail during\n  xmit. This may lead double free since the skb was not detached from\n  the virtqueue.\n- try_fill_recv() returns false when virtqueue_kick() fail. This will\n  lead unnecessary rescheduling of refill work.\n\nActually, it\u0027s safe to just ignore the kick failure in those two\nplaces. So this patch fixes this by partially revert commit\n67975901183799af8e93ec60e322f9e2a1940b9b.\n\nFixes 67975901183799af8e93ec60e322f9e2a1940b9b\n(virtio_net: verify if virtqueue_kick() succeeded).\n\nCc: Heinz Graalfs \u003cgraalfs@linux.vnet.ibm.com\u003e\nCc: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nCc: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nAcked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "75c5a52da3fc2a06abb6c6192bdf5d680e56d37d",
      "tree": "ad31cd7d9c1d5893d6b2ce392b170c068b625688",
      "parents": [
        "f217c44ebd41ce7369d2df07622b2839479183b0"
      ],
      "author": {
        "name": "Jan Kara",
        "email": "jack@suse.cz",
        "time": "Wed Mar 26 06:20:14 2014 +0100"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 27 09:52:54 2014 -0700"
      },
      "message": "vfs: Allocate anon_inode_inode in anon_inode_init()\n\nCurrently we allocated anon_inode_inode in anon_inodefs_mount. This is\nsomewhat fragile as if that function ever gets called again, it will\noverwrite anon_inode_inode pointer. So move the initialization of\nanon_inode_inode to anon_inode_init().\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n[ Further simplified on suggestion from Dave Jones ]\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "adbbdbac04f093c0abf946b1e93e4e5291808491",
      "tree": "8c54c687948f03ff4a261550c9e5bd454fe6aab0",
      "parents": [
        "347cf10aed1657a2b385a95f92763a67062c5ad3"
      ],
      "author": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Wed Mar 26 14:09:37 2014 +1000"
      },
      "committer": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Thu Mar 27 02:20:37 2014 +0000"
      },
      "message": "drm/nouveau: fail runtime pm properly.\n\nIf we were on a non-optimus device, we\u0027d return -EINVAL, this would\nlead to the over engineered runtime pm system to go into an error\nstate, subsequent get_sync\u0027s would fail, so we\u0027d never be able\nto open the device again.\n\n(like really get_sync shouldn\u0027t fail if the device isn\u0027t powered\ndown).\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n"
    },
    {
      "commit": "347cf10aed1657a2b385a95f92763a67062c5ad3",
      "tree": "0e63e153ca002dc47a37e1505b432ac54038e61e",
      "parents": [
        "b098d6726bbfb94c06d6e1097466187afddae61f"
      ],
      "author": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Tue Mar 25 14:38:44 2014 +1000"
      },
      "committer": {
        "name": "Dave Airlie",
        "email": "airlied@redhat.com",
        "time": "Thu Mar 27 02:19:50 2014 +0000"
      },
      "message": "drm/udl: take reference to device struct for dma-bufs\n\nthis stops the device from being deleted before all the dma-bufs\non it are freed, this fixes an oops when you unplug a udl device while\nit has imported a buffer from another device.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n"
    },
    {
      "commit": "de1443916791d75fdd26becb116898277bb0273f",
      "tree": "26fa8bd2f7ea36df46746a2b57e48f3247733d13",
      "parents": [
        "dc0fe58f356b109c662c6c85562460b0e8be9f5f"
      ],
      "author": {
        "name": "Eric Dumazet",
        "email": "edumazet@google.com",
        "time": "Tue Mar 25 18:42:27 2014 -0700"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 17:05:40 2014 -0400"
      },
      "message": "net: unix: non blocking recvmsg() should not return -EINTR\n\nSome applications didn\u0027t expect recvmsg() on a non blocking socket\ncould return -EINTR. This possibility was added as a side effect\nof commit b3ca9b02b00704 (\"net: fix multithreaded signal handling in\nunix recv routines\").\n\nTo hit this bug, you need to be a bit unlucky, as the u-\u003ereadlock\nmutex is usually held for very small periods.\n\nFixes: b3ca9b02b00704 (\"net: fix multithreaded signal handling in unix recv routines\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Rainer Weikusat \u003crweikusat@mobileactivedefense.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "dc0fe58f356b109c662c6c85562460b0e8be9f5f",
      "tree": "b2acdc5f918c94602776b1ad3cd05d075b625096",
      "parents": [
        "51dfe7b944998eaeb2b34d314f3a6b16a5fd621b",
        "b5f3b75d9d3b5526d973fc0bfee5680bdc6acf2a"
      ],
      "author": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 16:53:04 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 16:53:04 2014 -0400"
      },
      "message": "Merge branch \u0027mvneta\u0027\n\nThomas Petazzoni says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: mvneta: fix usage as a module\n\nThe following set of two patches fix the usage of the mvneta driver\nwhen built as a module, and used in RGMII configurations. It is\nsomewhat similar to a previous fix that was made by Arnaud Patard, but\nwhich was limited to SGMII configurations.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "b5f3b75d9d3b5526d973fc0bfee5680bdc6acf2a",
      "tree": "b2acdc5f918c94602776b1ad3cd05d075b625096",
      "parents": [
        "e3a8786c10e75903f1269474e21fe8cb49c3a670"
      ],
      "author": {
        "name": "Thomas Petazzoni",
        "email": "thomas.petazzoni@free-electrons.com",
        "time": "Wed Mar 26 00:26:55 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 16:52:42 2014 -0400"
      },
      "message": "net: mvneta: use devm_ioremap_resource() instead of of_iomap()\n\nThe mvneta driver currently uses of_iomap(), which has two drawbacks:\nit doesn\u0027t request the resource, and it isn\u0027t devm-style so some error\nhandling is needed.\n\nThis commit switches to use devm_ioremap_resource() instead, which\nautomatically requests the resource (so the I/O registers region shows\nup properly in /proc/iomem), and also is devm-style, which allows to\nget rid of some error handling to unmap the I/O registers region.\n\nSigned-off-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "e3a8786c10e75903f1269474e21fe8cb49c3a670",
      "tree": "b4e84134707f4e7a64a0e0685543d8b5103901de",
      "parents": [
        "a79121d3b57e7ad61f0b5d23eae05214054f3ccd"
      ],
      "author": {
        "name": "Thomas Petazzoni",
        "email": "thomas.petazzoni@free-electrons.com",
        "time": "Wed Mar 26 00:25:42 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 16:52:41 2014 -0400"
      },
      "message": "net: mvneta: fix usage as a module on RGMII configurations\n\nCommit 5445eaf309ff (\u0027mvneta: Try to fix mvneta when compiled as\nmodule\u0027) fixed the mvneta driver to make it work properly when loaded\nas a module in SGMII configuration, which was tested successful by the\nauthor on the Armada XP OpenBlocks AX3, which uses SGMII.\n\nHowever, it turns out that the Armada XP GP, which uses RGMII, is\naffected by a similar problem: its SERDES configuration is lost when\nmvneta is loaded as a module, because this configuration is set by the\nbootloader, and then lost because the clock is gated by the clock\nframework until the mvneta driver is loaded again and the clock is\nre-enabled.\n\nHowever, it turns out that for the RGMII case, setting the SERDES\nconfiguration is not sufficient: the PCS enable bit in the\nMVNETA_GMAC_CTRL_2 register must also be set, like in the SGMII\nconfiguration.\n\nTherefore, this commit reworks the SGMII/RGMII initialization: the\nonly difference between the two now is a different SERDES\nconfiguration, all the rest is identical.\n\nIn detail, to achieve this, the commit:\n\n * Renames MVNETA_SGMII_SERDES_CFG to MVNETA_SERDES_CFG because it is\n   not specific to SGMII, but also used on RGMII configurations.\n\n * Adds a MVNETA_RGMII_SERDES_PROTO definition, that must be used as\n   the MVNETA_SERDES_CFG value in RGMII configurations.\n\n * Removes the mvneta_gmac_rgmii_set() and mvneta_port_sgmii_config()\n   functions, and instead directly do the SGMII/RGMII configuration in\n   mvneta_port_up(), from where those functions where called. It is\n   worth mentioning that mvneta_gmac_rgmii_set() had an \u0027enable\u0027\n   parameter that was always passed as \u00271\u0027, so it was pretty useless.\n\n * Reworks the mvneta_port_up() function to set the MVNETA_SERDES_CFG\n   register to the appropriate value depending on the RGMII vs. SGMII\n   configuration. It also unconditionally set the PCS_ENABLE bit (was\n   already done for SGMII, but is now also needed for RGMII), and sets\n   the PORT_RGMII bit (which was already done for both SGMII and\n   RGMII).\n\nThis commit was successfully tested with mvneta compiled as a module,\non both the OpenBlocks AX3 (SGMII configuration) and the Armada XP GP\n(RGMII configuration).\n\nReported-by: Steve McIntyre \u003csteve@einval.com\u003e\nCc: stable@vger.kernel.org # 3.11.x: 5445eaf309ff mvneta: Try to fix mvneta when compiled as module\nSigned-off-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "a79121d3b57e7ad61f0b5d23eae05214054f3ccd",
      "tree": "3b2b20be22a22905ab0254f963733bba28a61f9f",
      "parents": [
        "51dfe7b944998eaeb2b34d314f3a6b16a5fd621b"
      ],
      "author": {
        "name": "Thomas Petazzoni",
        "email": "thomas.petazzoni@free-electrons.com",
        "time": "Wed Mar 26 00:25:41 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 16:52:41 2014 -0400"
      },
      "message": "net: mvneta: rename MVNETA_GMAC2_PSC_ENABLE to MVNETA_GMAC2_PCS_ENABLE\n\nBit 3 of the MVNETA_GMAC_CTRL_2 is actually used to enable the PCS,\nnot the PSC: there was a typo in the name of the define, which this\ncommit fixes.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "6797b39e6f6f34c74177736e146406e894b9482b",
      "tree": "f834ae0e2b8e8c72233bbb534f3e0f8cae26d707",
      "parents": [
        "70b0052425ffd549bb27fb08649a4d30daaf40e4"
      ],
      "author": {
        "name": "Hans de Goede",
        "email": "hdegoede@redhat.com",
        "time": "Wed Mar 26 13:30:52 2014 -0700"
      },
      "committer": {
        "name": "Dmitry Torokhov",
        "email": "dmitry.torokhov@gmail.com",
        "time": "Wed Mar 26 13:33:58 2014 -0700"
      },
      "message": "Input: cypress_ps2 - don\u0027t report as a button pads\n\nThe cypress PS/2 trackpad models supported by the cypress_ps2 driver\nemulate BTN_RIGHT events in firmware based on the finger position, as part\nof this no motion events are sent when the finger is in the button area.\n\nThe INPUT_PROP_BUTTONPAD property is there to indicate to userspace that\nBTN_RIGHT events should be emulated in userspace, which is not necessary\nin this case.\n\nWhen INPUT_PROP_BUTTONPAD is advertised userspace will wait for a motion\nevent before propagating the button event higher up the stack, as it needs\ncurrent abs x + y data for its BTN_RIGHT emulation. Since in the\ncypress_ps2 pads don\u0027t report motion events in the button area, this means\nthat clicks in the button area end up being ignored, so\nINPUT_PROP_BUTTONPAD actually causes problems for these touchpads, and\nremoving it fixes:\n\nhttps://bugs.freedesktop.org/show_bug.cgi?id\u003d76341\n\nReported-by: Adam Williamson \u003cawilliam@redhat.com\u003e\nTested-by: Adam Williamson \u003cawilliam@redhat.com\u003e\nReviewed-by: Peter Hutterer \u003cpeter.hutterer@who-t.net\u003e\nSigned-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nSigned-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n"
    },
    {
      "commit": "51dfe7b944998eaeb2b34d314f3a6b16a5fd621b",
      "tree": "395fd66c62e4689c3f27ac86f223f9363d039f97",
      "parents": [
        "fbd02dd405d0724a0f25897ed4a6813297c9b96f"
      ],
      "author": {
        "name": "Vlad Yasevich",
        "email": "vyasevic@redhat.com",
        "time": "Mon Mar 24 17:52:12 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 15:55:10 2014 -0400"
      },
      "message": "tg3: Do not include vlan acceleration features in vlan_features\n\nIncluding hardware acceleration features in vlan_features breaks\nstacked vlans (Q-in-Q) by marking the bottom vlan interface as\ncapable of acceleration.  This causes one of the tags to be lost\nand the packets are sent with a sing vlan header.\n\nCC: Nithin Nayak Sujir \u003cnsujir@broadcom.com\u003e\nCC: Michael Chan \u003cmchan@broadcom.com\u003e\nSigned-off-by: Vlad Yasevich \u003cvyasevic@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "fbd02dd405d0724a0f25897ed4a6813297c9b96f",
      "tree": "4ebae2c7d156ffa99c1c6ec2d14a8061d0a7b742",
      "parents": [
        "632b06aa2842b12c6d6a510ec080fb6ebdb38ea5"
      ],
      "author": {
        "name": "Pravin B Shelar",
        "email": "pshelar@nicira.com",
        "time": "Sun Mar 23 22:06:36 2014 -0700"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Wed Mar 26 15:18:40 2014 -0400"
      },
      "message": "ip_tunnel: Fix dst ref-count.\n\nCommit 10ddceb22ba (ip_tunnel:multicast process cause panic due\nto skb-\u003e_skb_refdst NULL pointer) removed dst-drop call from\nip-tunnel-recv.\n\nFollowing commit reintroduce dst-drop and fix the original bug by\nchecking loopback packet before releasing dst.\nOriginal bug: https://bugzilla.kernel.org/show_bug.cgi?id\u003d70681\n\nCC: Xin Long \u003clucien.xin@gmail.com\u003e\nSigned-off-by: Pravin B Shelar \u003cpshelar@nicira.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "f217c44ebd41ce7369d2df07622b2839479183b0",
      "tree": "c8ee34e6d2b20e068c5241fbe663983ab0a1a88d",
      "parents": [
        "fce7fc79c8f7188dfc5eafa1b937bcc3c5a4c2f5",
        "2c4a33aba5f9ea3a28f2e40351f078d95f00786b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Wed Mar 26 09:09:18 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Wed Mar 26 09:09:18 2014 -0700"
      },
      "message": "Merge tag \u0027trace-fixes-v3.14-rc7-v2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fix from Steven Rostedt:\n \"While on my flight to Linux Collaboration Summit, I was working on my\n  slides for the event trigger tutorial.  I booted a 3.14-rc7 kernel to\n  perform what I wanted to teach and cut and paste it into my slides.\n  When I tried the traceon event trigger with a condition attached to it\n  (turns tracing on only if a field of the trigger event matches a\n  condition set by the user), nothing happened.  Tracing would not turn\n  on.  I stopped working on my presentation in order to find what was\n  wrong.\n\n  It ended up being the way trace event triggers work when they have\n  conditions.  Instead of copying the fields, the condition code just\n  looks at the fields that were copied into the ring buffer.  This works\n  great, unless tracing is off.  That\u0027s because when the event is\n  reserved on the ring buffer, the ring buffer returns a NULL pointer,\n  this tells the tracing code that the ring buffer is disabled.  This\n  ends up being a problem for the traceon trigger if it is using this\n  information to check its condition.\n\n  Luckily the code that checks if tracing is on returns the ring buffer\n  to use (because the ring buffer is determined by the event file also\n  passed to that field).  I was able to easily solve this bug by\n  checking in that helper function if the returned ring buffer entry is\n  NULL, and if so, also check the file flag if it has a trace event\n  trigger condition, and if so, to pass back a temp ring buffer to use.\n  This will allow the trace event trigger condition to still test the\n  event fields, but nothing will be recorded\"\n\n* tag \u0027trace-fixes-v3.14-rc7-v2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n  tracing: Fix traceon trigger condition to actually turn tracing on\n"
    },
    {
      "commit": "2c4a33aba5f9ea3a28f2e40351f078d95f00786b",
      "tree": "657da8f44eaa6331bd8bfa4789f10b152bc8104c",
      "parents": [
        "87291347c49dc40aa339f587b209618201c2e527"
      ],
      "author": {
        "name": "Steven Rostedt (Red Hat)",
        "email": "rostedt@goodmis.org",
        "time": "Tue Mar 25 23:39:41 2014 -0400"
      },
      "committer": {
        "name": "Steven Rostedt",
        "email": "rostedt@goodmis.org",
        "time": "Tue Mar 25 23:39:41 2014 -0400"
      },
      "message": "tracing: Fix traceon trigger condition to actually turn tracing on\n\nWhile working on my tutorial for 2014 Linux Collaboration Summit\nI found that the traceon trigger did not work when conditions were\nused. The other triggers worked fine though. Looking into it, it\nis because of the way the triggers use the ring buffer to store\nthe fields it will use for the condition. But if tracing is off, nothing\nis stored in the buffer, and the tracepoint exits before calling the\ntrigger to test the condition. This is fine for all the triggers that\nonly work when tracing is on, but for traceon trigger that is to\nwork when tracing is off, nothing happens.\n\nThe fix is simple, just use a temp ring buffer to record the event\nif tracing is off and the event has a trace event conditional trigger\nenabled. The rest of the tracepoint code will work just fine, but\nthe tracepoint wont be recorded in the other buffers.\n\nCc: Tom Zanussi \u003ctom.zanussi@linux.intel.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n"
    },
    {
      "commit": "fce7fc79c8f7188dfc5eafa1b937bcc3c5a4c2f5",
      "tree": "1c5e437460385f833f8fd2608e7526f610282d9e",
      "parents": [
        "d6f2589ad561aa5fa39f347eca6942668b7560a1"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 17:43:34 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 17:43:34 2014 -0700"
      },
      "message": "fs: remove now stale label in anon_inode_init()\n\nThe previous commit removed the register_filesystem() call and the\nassociated error handling, but left the label for the error path that no\nlonger exists.  Remove that too.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "d6f2589ad561aa5fa39f347eca6942668b7560a1",
      "tree": "48708ad51abdeed29c5787350f0f2a378297754d",
      "parents": [
        "632b06aa2842b12c6d6a510ec080fb6ebdb38ea5"
      ],
      "author": {
        "name": "Jan Kara",
        "email": "jack@suse.cz",
        "time": "Tue Mar 25 21:37:09 2014 +0100"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 17:42:16 2014 -0700"
      },
      "message": "fs: Avoid userspace mounting anon_inodefs filesystem\n\nanon_inodefs filesystem is a kernel internal filesystem userspace\nshouldn\u0027t mess with. Remove registration of it so userspace cannot\neven try to mount it (which would fail anyway because the filesystem is\nMS_NOUSER).\n\nThis fixes an oops triggered by trinity when it tried mounting\nanon_inodefs which overwrote anon_inode_inode pointer while other CPU\nhas been in anon_inode_getfile() between ihold() and d_instantiate().\nThus effectively creating dentry pointing to an inode without holding a\nreference to it.\n\nReported-by: Sasha Levin \u003csasha.levin@oracle.com\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "632b06aa2842b12c6d6a510ec080fb6ebdb38ea5",
      "tree": "42dee61b79734e9465001ad3bbcb7597ff3eb20e",
      "parents": [
        "3e79d97828841305e3369ad1e07cfed5bf5989ef",
        "1406b916f4a29d5f9660264a28ce609c8c77e7ae"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 15:24:11 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 15:24:11 2014 -0700"
      },
      "message": "Merge branch \u0027nfsd-next\u0027 of git://linux-nfs.org/~bfields/linux\n\nPull nfsd fix frm Bruce Fields:\n \"J R Okajima sent this early and I was just slow to pass it along,\n  apologies.  Fortunately it\u0027s a simple fix\"\n\n* \u0027nfsd-next\u0027 of git://linux-nfs.org/~bfields/linux:\n  nfsd: fix lost nfserrno() call in nfsd_setattr()\n"
    },
    {
      "commit": "3e79d97828841305e3369ad1e07cfed5bf5989ef",
      "tree": "d14264947e9bf1a72922060dfefd08693e550ebc",
      "parents": [
        "b098d6726bbfb94c06d6e1097466187afddae61f",
        "b37199e626b31e1175fb06764c5d1d687723aac2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 15:05:57 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Mar 25 15:05:57 2014 -0700"
      },
      "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull vfs fixes from Al Viro:\n \"These four commits are obvious fixes (a couple of fdget_pos()-related\n  ones from Eric Biggers, prepend_name() fix, missing checks for false\n  negatives from __lookup_mnt() in fs/namei.c)\"\n\nFor now I\u0027m pulling just the four obvious fixes, there\u0027s another four\npending in Al\u0027s \u0027for-linus\u0027 branch wrt the mnt_hash list that were more\ninvolved.\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n  rcuwalk: recheck mount_lock after mountpoint crossing attempts\n  make prepend_name() work correctly when called with negative *buflen\n  vfs: Don\u0027t let __fdget_pos() get FMODE_PATH files\n  vfs: atomic f_pos access in llseek()\n"
    },
    {
      "commit": "5926f87fdaad4be3ed10cec563bf357915e55a86",
      "tree": "f44ee105eb3a1bb59ea098788c6949004438cf41",
      "parents": [
        "09ed3d5ba06137913960f9c9385f71fc384193ab"
      ],
      "author": {
        "name": "David Vrabel",
        "email": "david.vrabel@citrix.com",
        "time": "Tue Mar 25 10:38:37 2014 +0000"
      },
      "committer": {
        "name": "David Vrabel",
        "email": "david.vrabel@citrix.com",
        "time": "Tue Mar 25 11:11:42 2014 +0000"
      },
      "message": "Revert \"xen: properly account for _PAGE_NUMA during xen pte translations\"\n\nThis reverts commit a9c8e4beeeb64c22b84c803747487857fe424b68.\n\nPTEs in Xen PV guests must contain machine addresses if _PAGE_PRESENT\nis set and pseudo-physical addresses is _PAGE_PRESENT is clear.\n\nThis is because during a domain save/restore (migration) the page\ntable entries are \"canonicalised\" and uncanonicalised\". i.e., MFNs are\nconverted to PFNs during domain save so that on a restore the page\ntable entries may be rewritten with the new MFNs on the destination.\nThis canonicalisation is only done for PTEs that are present.\n\nThis change resulted in writing PTEs with MFNs if _PAGE_PROTNONE (or\n_PAGE_NUMA) was set but _PAGE_PRESENT was clear.  These PTEs would be\nmigrated as-is which would result in unexpected behaviour in the\ndestination domain.  Either a) the MFN would be translated to the\nwrong PFN/page; b) setting the _PAGE_PRESENT bit would clear the PTE\nbecause the MFN is no longer owned by the domain; or c) the present\nbit would not get set.\n\nSymptoms include \"Bad page\" reports when munmapping after migrating a\ndomain.\n\nSigned-off-by: David Vrabel \u003cdavid.vrabel@citrix.com\u003e\nAcked-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e        [3.12+]\n"
    },
    {
      "commit": "09ed3d5ba06137913960f9c9385f71fc384193ab",
      "tree": "582150ce743efb10051e0bc13747e3e1a24005b9",
      "parents": [
        "b098d6726bbfb94c06d6e1097466187afddae61f"
      ],
      "author": {
        "name": "Wei Liu",
        "email": "wei.liu2@citrix.com",
        "time": "Sat Mar 15 16:11:47 2014 +0000"
      },
      "committer": {
        "name": "David Vrabel",
        "email": "david.vrabel@citrix.com",
        "time": "Tue Mar 25 10:38:30 2014 +0000"
      },
      "message": "xen/balloon: flush persistent kmaps in correct position\n\nXen balloon driver will update ballooned out pages\u0027 P2M entries to point\nto scratch page for PV guests. In 24f69373e2 (\"xen/balloon: don\u0027t alloc\npage while non-preemptible\", kmap_flush_unused was moved after updating\nP2M table. In that case for 32 bit PV guest we might end up with\n\n  P2M    X -----\u003e S  (S is mfn of balloon scratch page)\n  M2P    Y -----\u003e X  (Y is mfn in persistent kmap entry)\n\nkmap_flush_unused() iterates through all the PTEs in the kmap address\nspace, using pte_to_page() to obtain the page. If the p2m and the m2p\nare inconsistent the incorrect page is returned.  This will clear\npage-\u003eaddress on the wrong page which may cause subsequent oopses if\nthat page is currently kmap\u0027ed.\n\nMove the flush back between get_page and __set_phys_to_machine to fix\nthis.\n\nSigned-off-by: Wei Liu \u003cwei.liu2@citrix.com\u003e\nSigned-off-by: David Vrabel \u003cdavid.vrabel@citrix.com\u003e\nCc: stable@vger.kernel.org # 3.12+\n"
    },
    {
      "commit": "b098d6726bbfb94c06d6e1097466187afddae61f",
      "tree": "dadbc3901075f02dded557ec33a48a05d52218c0",
      "parents": [
        "822316461b15e0207e50ff661f9cf830af116e9f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 19:31:17 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 19:31:17 2014 -0700"
      },
      "message": "Linux 3.14-rc8\n"
    },
    {
      "commit": "822316461b15e0207e50ff661f9cf830af116e9f",
      "tree": "96e42d5eca385746ea6791807c93933952dc1c3e",
      "parents": [
        "56f1f4b24e8787d7ba794dbe2e949d504c054892",
        "a34fe10750ebe524a39f97bd78ab4d232a554edb"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 17:36:58 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 17:36:58 2014 -0700"
      },
      "message": "Merge branch \u0027parisc-3.14\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux\n\nPull parisc updates from Helge Deller:\n - revert parts of the latest patch regarding font selection with STICON\n   console\n - wire up the utimes() syscall for parisc\n - remove the unused parisc tmpalias code and unnecessary arch*relax\n   defines\n\n* \u0027parisc-3.14\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:\n  parisc: locks: remove redundant arch_*_relax operations\n  parisc: wire up sys_utimes\n  parisc: Remove unused CONFIG_PARISC_TMPALIAS code\n  partly revert commit 8a10bc9: parisc/sti_console: prefer Linux fonts over built-in ROM fonts\n"
    },
    {
      "commit": "56f1f4b24e8787d7ba794dbe2e949d504c054892",
      "tree": "b884069e95f50fee18cb0834a304d059d84bf2d0",
      "parents": [
        "8a1094462c7ec47f9d3a31b0a281ce35cb138f50",
        "cb3042d609e30e6144024801c89be3925106752b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 17:30:44 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 17:30:44 2014 -0700"
      },
      "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc\n\nPull sparc fixes from David Miller:\n\n 1) Do serial locking in a way that makes things clear that these are\n    IRQ spinlocks.\n\n 2) Conversion to generic idle loop broke first generation Niagara\n    machines, need to have %pil interrupts enabled during cpu yield\n    hypervisor call.\n\n 3) Do not use magic constants for iterations over tsb tables, from Doug\n    Wilson.\n\n 4) Fix erroneous truncation of 64-bit system call return values to\n    32-bit.  From Dave Kleikamp.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:\n  sparc64: Make sure %pil interrupts are enabled during hypervisor yield.\n  sparc64:tsb.c:use array size macro rather than number\n  sparc64: don\u0027t treat 64-bit syscall return codes as 32-bit\n  sparc: serial: Clean up the locking for -rt\n"
    },
    {
      "commit": "8a1094462c7ec47f9d3a31b0a281ce35cb138f50",
      "tree": "c05accc7db20026a417a6a54a1ca33766e2505af",
      "parents": [
        "774868c7094d35b4518be3d0e654de000a5d11fc",
        "a5d0e7c037119484a7006b883618bfa87996cb41"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 17:07:24 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Mar 24 17:07:24 2014 -0700"
      },
      "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) OpenVswitch\u0027s lookup_datapath() returns error pointers, so don\u0027t\n    check against NULL.  From Jiri Pirko.\n\n 2) pfkey_compile_policy() code path tries to do a GFP_KERNEL allocation\n    under RCU locks, fix by using GFP_ATOMIC when necessary.  From\n    Nikolay Aleksandrov.\n\n 3) phy_suspend() indirectly passes uninitialized data into the ethtool\n    get wake-on-land implementations.  Fix from Sebastian Hesselbarth.\n\n 4) CPSW driver unregisters CPTS twice, fix from Benedikt Spranger.\n\n 5) If SKB allocation of reply packet fails, vxlan\u0027s arp_reduce() defers\n    a NULL pointer.  Fix from David Stevens.\n\n 6) IPV6 neigh handling in vxlan doesn\u0027t validate the destination\n    address properly, and it builds a packet with the src and dst\n    reversed.  Fix also from David Stevens.\n\n 7) Fix spinlock recursion during subscription failures in TIPC stack,\n    from Erik Hugne.\n\n 8) Revert buggy conversion of davinci_emac to devm_request_irq, from\n    Chrstian Riesch.\n\n 9) Wrong flags passed into forwarding database netlink notifications,\n    from Nicolas Dichtel.\n\n10) The netpoll neighbour soliciation handler checks wrong ethertype,\n    needs to be ETH_P_IPV6 rather than ETH_P_ARP.  Fix from Li RongQing.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (34 commits)\n  tipc: fix spinlock recursion bug for failed subscriptions\n  vxlan: fix nonfunctional neigh_reduce()\n  net: davinci_emac: Fix rollback of emac_dev_open()\n  net: davinci_emac: Replace devm_request_irq with request_irq\n  netpoll: fix the skb check in pkt_is_ns\n  net: micrel : ks8851-ml: add vdd-supply support\n  ip6mr: fix mfc notification flags\n  ipmr: fix mfc notification flags\n  rtnetlink: fix fdb notification flags\n  tcp: syncookies: do not use getnstimeofday()\n  netlink: fix setsockopt in mmap examples in documentation\n  openvswitch: Correctly report flow used times for first 5 minutes after boot.\n  via-rhine: Disable device in error path\n  ATHEROS-ATL1E: Convert iounmap to pci_iounmap\n  vxlan: fix potential NULL dereference in arp_reduce()\n  cnic: Update version to 2.5.20 and copyright year.\n  cnic,bnx2i,bnx2fc: Fix inconsistent use of page size\n  cnic: Use proper ulp_ops for per device operations.\n  net: cdc_ncm: fix control message ordering\n  ipv6: ip6_append_data_mtu do not handle the mtu of the second fragment properly\n  ...\n"
    },
    {
      "commit": "a5d0e7c037119484a7006b883618bfa87996cb41",
      "tree": "0240143aaed7d20be2d7198393e4f5c1e4b4f681",
      "parents": [
        "4b29dba9c085a4fb79058fb1c45a2f6257ca3dfa"
      ],
      "author": {
        "name": "Erik Hugne",
        "email": "erik.hugne@ericsson.com",
        "time": "Mon Mar 24 16:56:38 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:36:56 2014 -0400"
      },
      "message": "tipc: fix spinlock recursion bug for failed subscriptions\n\nIf a topology event subscription fails for any reason, such as out\nof memory, max number reached or because we received an invalid\nrequest the correct behavior is to terminate the subscribers\nconnection to the topology server. This is currently broken and\nproduces the following oops:\n\n[27.953662] tipc: Subscription rejected, illegal request\n[27.955329] BUG: spinlock recursion on CPU#1, kworker/u4:0/6\n[27.957066]  lock: 0xffff88003c67f408, .magic: dead4ead, .owner: kworker/u4:0/6, .owner_cpu: 1\n[27.958054] CPU: 1 PID: 6 Comm: kworker/u4:0 Not tainted 3.14.0-rc6+ #5\n[27.960230] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011\n[27.960874] Workqueue: tipc_rcv tipc_recv_work [tipc]\n[27.961430]  ffff88003c67f408 ffff88003de27c18 ffffffff815c0207 ffff88003de1c050\n[27.962292]  ffff88003de27c38 ffffffff815beec5 ffff88003c67f408 ffffffff817f0a8a\n[27.963152]  ffff88003de27c58 ffffffff815beeeb ffff88003c67f408 ffffffffa0013520\n[27.964023] Call Trace:\n[27.964292]  [\u003cffffffff815c0207\u003e] dump_stack+0x45/0x56\n[27.964874]  [\u003cffffffff815beec5\u003e] spin_dump+0x8c/0x91\n[27.965420]  [\u003cffffffff815beeeb\u003e] spin_bug+0x21/0x26\n[27.965995]  [\u003cffffffff81083df6\u003e] do_raw_spin_lock+0x116/0x140\n[27.966631]  [\u003cffffffff815c6215\u003e] _raw_spin_lock_bh+0x15/0x20\n[27.967256]  [\u003cffffffffa0008540\u003e] subscr_conn_shutdown_event+0x20/0xa0 [tipc]\n[27.968051]  [\u003cffffffffa000fde4\u003e] tipc_close_conn+0xa4/0xb0 [tipc]\n[27.968722]  [\u003cffffffffa00101ba\u003e] tipc_conn_terminate+0x1a/0x30 [tipc]\n[27.969436]  [\u003cffffffffa00089a2\u003e] subscr_conn_msg_event+0x1f2/0x2f0 [tipc]\n[27.970209]  [\u003cffffffffa0010000\u003e] tipc_receive_from_sock+0x90/0xf0 [tipc]\n[27.970972]  [\u003cffffffffa000fa79\u003e] tipc_recv_work+0x29/0x50 [tipc]\n[27.971633]  [\u003cffffffff8105dbf5\u003e] process_one_work+0x165/0x3e0\n[27.972267]  [\u003cffffffff8105e869\u003e] worker_thread+0x119/0x3a0\n[27.972896]  [\u003cffffffff8105e750\u003e] ? manage_workers.isra.25+0x2a0/0x2a0\n[27.973622]  [\u003cffffffff810648af\u003e] kthread+0xdf/0x100\n[27.974168]  [\u003cffffffff810647d0\u003e] ? kthread_create_on_node+0x1a0/0x1a0\n[27.974893]  [\u003cffffffff815ce13c\u003e] ret_from_fork+0x7c/0xb0\n[27.975466]  [\u003cffffffff810647d0\u003e] ? kthread_create_on_node+0x1a0/0x1a0\n\nThe recursion occurs when subscr_terminate tries to grab the\nsubscriber lock, which is already taken by subscr_conn_msg_event.\nWe fix this by checking if the request to establish a new\nsubscription was successful, and if not we initiate termination of\nthe subscriber after we have released the subscriber lock.\n\nSigned-off-by: Erik Hugne \u003cerik.hugne@ericsson.com\u003e\nReviewed-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "4b29dba9c085a4fb79058fb1c45a2f6257ca3dfa",
      "tree": "3ee4d0ba0ee4ec64d8c1103dc1b2dc1e5231d3e5",
      "parents": [
        "866b7cdf5f8e440c4921765b7f4e1fe102a7e089"
      ],
      "author": {
        "name": "David Stevens",
        "email": "dlstevens@us.ibm.com",
        "time": "Mon Mar 24 10:39:58 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:35:10 2014 -0400"
      },
      "message": "vxlan: fix nonfunctional neigh_reduce()\n\nThe VXLAN neigh_reduce() code is completely non-functional since\ncheck-in. Specific errors:\n\n1) The original code drops all packets with a multicast destination address,\n\teven though neighbor solicitations are sent to the solicited-node\n\taddress, a multicast address. The code after this check was never run.\n2) The neighbor table lookup used the IPv6 header destination, which is the\n\tsolicited node address, rather than the target address from the\n\tneighbor solicitation. So neighbor lookups would always fail if it\n\tgot this far. Also for L3MISSes.\n3) The code calls ndisc_send_na(), which does a send on the tunnel device.\n\tThe context for neigh_reduce() is the transmit path, vxlan_xmit(),\n\twhere the host or a bridge-attached neighbor is trying to transmit\n\ta neighbor solicitation. To respond to it, the tunnel endpoint needs\n\tto do a *receive* of the appropriate neighbor advertisement. Doing a\n\tsend, would only try to send the advertisement, encapsulated, to the\n\tremote destinations in the fdb -- hosts that definitely did not do the\n\tcorresponding solicitation.\n4) The code uses the tunnel endpoint IPv6 forwarding flag to determine the\n\tisrouter flag in the advertisement. This has nothing to do with whether\n\tor not the target is a router, and generally won\u0027t be set since the\n\ttunnel endpoint is bridging, not routing, traffic.\n\n\tThe patch below creates a proxy neighbor advertisement to respond to\nneighbor solicitions as intended, providing proper IPv6 support for neighbor\nreduction.\n\nSigned-off-by: David L Stevens \u003cdlstevens@us.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "866b7cdf5f8e440c4921765b7f4e1fe102a7e089",
      "tree": "06ca0a0ed579d0e8aaab3a7b0f2921f3c9c20855",
      "parents": [
        "c27f0872a3448c46e561e226b5b97f77187b06d2",
        "cd11cf505318ff24e42f35145f9cdf8596fa1958"
      ],
      "author": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:32:34 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:32:34 2014 -0400"
      },
      "message": "Merge branch \u0027davinci_emac\u0027\n\nChristian Riesch says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: davinci_emac: Fix interrupt requests and error handling\n\nsince commit 6892b41d9701283085b655c6086fb57a5d63fa47 (Linux 3.11) the\ndavinci_emac driver is broken. After doing ifconfig down, ifconfig up,\nrequesting the interrupts for the driver fails. The interface remains dead\nuntil the board is rebooted.\n\nThe first patch in this patchset reverts commit\n6892b41d9701283085b655c6086fb57a5d63fa47 partially and makes the driver\nuseable again.\n\nDuring the work on the first patch, a number of bugs in the error handling\nof the driver\u0027s ndo_open code were found. The second patch fixes these bugs.\n\nI believe the first patch meets the rules for stable kernels, I therefore added\nthe stable tag to this patch. The second patch is just cleanup, the code\nthat is fixed by this patch is only executed in case of an error.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "cd11cf505318ff24e42f35145f9cdf8596fa1958",
      "tree": "06ca0a0ed579d0e8aaab3a7b0f2921f3c9c20855",
      "parents": [
        "33b7107f59a61236d94ecd6b45e20283cd5abcc8"
      ],
      "author": {
        "name": "Christian Riesch",
        "email": "christian.riesch@omicron.at",
        "time": "Mon Mar 24 13:46:27 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:32:03 2014 -0400"
      },
      "message": "net: davinci_emac: Fix rollback of emac_dev_open()\n\nIf an error occurs during the initialization in emac_dev_open() (the\ndriver\u0027s ndo_open function), interrupts, DMA descriptors etc. must be freed.\nThe current rollback code is buggy in several ways.\n\n  1) Freeing the interrupts. The current code will not free all interrupts\n     that were requested by the driver. Furthermore,  the code tries to do a\n     platform_get_resource(priv-\u003epdev, IORESOURCE_IRQ, -1) in its last\n     iteration.\n\n     This patch fixes these bugs.\n\n  2) Wrong order of err: and rollback: labels. If the setup of the PHY in\n     the code fails, the interrupts that have been requested before are\n     not freed:\n\n        request irq\n                if requesting irqs fails, goto rollback\n        setup phy\n                if phy setup fails, goto err\n        return 0\n\n     rollback:\n        free irqs\n     err:\n\n     This patch brings the code into the correct order.\n\n  3) The code calls napi_enable() and emac_int_enable(), but does not\n     undo both in case of an error.\n\n     This patch adds calls of emac_int_disable() and napi_disable() to the\n     rollback code.\n\n  4) RX DMA descriptors are not freed in case of an error: Right before\n     requesting the irqs, the function creates DMA descriptors for the\n     RX channel. These RX descriptors are never freed when we jump to either\n     rollback or err.\n\n     This patch adds code for freeing the DMA descriptors in the case of\n     an initialization error. This required a modification of\n     cpdma_ctrl_stop() in davinci_cpdma.c: We must be able to call this\n     function to free the DMA descriptors while the DMA channels are\n     in IDLE state (before cpdma_ctlr_start() was called).\n\nTested on a custom board with the Texas Instruments AM1808.\n\nSigned-off-by: Christian Riesch \u003cchristian.riesch@omicron.at\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "33b7107f59a61236d94ecd6b45e20283cd5abcc8",
      "tree": "26aec0e59cfad25d27c772f8191bf75c039c67c0",
      "parents": [
        "c27f0872a3448c46e561e226b5b97f77187b06d2"
      ],
      "author": {
        "name": "Christian Riesch",
        "email": "christian.riesch@omicron.at",
        "time": "Mon Mar 24 13:46:26 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:32:03 2014 -0400"
      },
      "message": "net: davinci_emac: Replace devm_request_irq with request_irq\n\nIn commit 6892b41d9701283085b655c6086fb57a5d63fa47\n\nAuthor: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\nDate:   Tue Jun 25 21:24:51 2013 +0530\nnet: davinci: emac: Convert to devm_* api\n\nthe call of request_irq is replaced by devm_request_irq and the call\nof free_irq is removed. But since interrupts are requested in\nemac_dev_open, doing ifconfig up/down on the board requests the\ninterrupts again each time, causing devm_request_irq to fail. The\ninterface is dead until the device is rebooted.\n\nThis patch reverts said commit partially: It changes the driver back\nto use request_irq instead of devm_request_irq, puts free_irq back in\nplace, but keeps the remaining changes of the original patch.\n\nReported-by: Jon Ringle \u003cjon@ringle.org\u003e\nSigned-off-by: Christian Riesch \u003cchristian.riesch@omicron.at\u003e\nCc: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "c27f0872a3448c46e561e226b5b97f77187b06d2",
      "tree": "984e53e8277ec8b4bd50ee7b1bd8a803cfa02b85",
      "parents": [
        "ebf4ad955d3e26d4d2a33709624fc7b5b9d3b969"
      ],
      "author": {
        "name": "Li RongQing",
        "email": "roy.qing.li@gmail.com",
        "time": "Fri Mar 21 20:53:57 2014 +0800"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 15:08:40 2014 -0400"
      },
      "message": "netpoll: fix the skb check in pkt_is_ns\n\nNeighbor Solicitation is ipv6 protocol, so we should check\nskb-\u003eprotocol with ETH_P_IPV6\n\nSigned-off-by: Li RongQing \u003croy.qing.li@gmail.com\u003e\nCc: WANG Cong \u003camwang@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "cb3042d609e30e6144024801c89be3925106752b",
      "tree": "269c25b05aa778a030310f1656c501fd3d41c823",
      "parents": [
        "151b628f104566a450125a6a0f6775a35bde58d6"
      ],
      "author": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 14:45:12 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 14:45:12 2014 -0400"
      },
      "message": "sparc64: Make sure %pil interrupts are enabled during hypervisor yield.\n\nIn arch_cpu_idle() we must enable %pil based interrupts before\npotentially invoking the hypervisor cpu yield call.\n\nAs per the Hypervisor API documentation for cpu_yield:\n\n\tInterrupts which are blocked by some mechanism other that\n\tpstate.ie (for example %pil) are not guaranteed to cause\n\ta return from this service.\n\nIt seems that only first generation Niagara chips are hit by this\nbug.  My best guess is that later chips implement this in hardware\nand wake up anyways from %pil events, whereas in first generation\nchips the yield is implemented completely in hypervisor code and\nrequires %pil to be enabled in order to wake properly from this\ncall.\n\nFixes: 87fa05aeb3a5 (\"sparc: Use generic idle loop\")\nReported-by: Fabio M. Di Nitto \u003cfabbione@fabbione.net\u003e\nReported-by: Jan Engelhardt \u003cjengelh@inai.de\u003e\nTested-by: Jan Engelhardt \u003cjengelh@inai.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "5f12c5eca6e6b7aeb4b2028d579f614b4fe7a81f",
      "tree": "c4c7fc149fc1cb021580560b946b82542b1088ba",
      "parents": [
        "dcb99fd9b08cfe1afe426af4d8d3cbc429190f15"
      ],
      "author": {
        "name": "Scott Wood",
        "email": "scottwood@freescale.com",
        "time": "Tue Mar 18 16:10:24 2014 -0500"
      },
      "committer": {
        "name": "Wolfram Sang",
        "email": "wsa@the-dreams.de",
        "time": "Mon Mar 24 14:54:21 2014 +0100"
      },
      "message": "i2c: cpm: Fix build by adding of_address.h and of_irq.h\n\nFixes a build break due to the undeclared use of irq_of_parse_and_map()\nand of_iomap().  This build break was apparently introduced while the\ndriver was unbuildable due to the bug fixed by\n62c19c9d29e65086e5ae76df371ed2e6b23f00cd (\"i2c: Remove usage of\norphaned symbol OF_I2C\").  When 62c19c was added in v3.14-rc7,\nthe driver was enabled again, breaking the powerpc mpc85xx_defconfig\nand mpc85xx_smp_defconfig.\n\n62c19c is marked for stable, so this should go there as well.\n\nReported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nSigned-off-by: Scott Wood \u003cscottwood@freescale.com\u003e\nSigned-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\nCc: stable@kernel.org\n"
    },
    {
      "commit": "ebf4ad955d3e26d4d2a33709624fc7b5b9d3b969",
      "tree": "da93c4744f9787a30027dda791b6101aebaa89e2",
      "parents": [
        "b74d3feccc3f5e6e4cf98cf88daa2e0cdbc4ca62"
      ],
      "author": {
        "name": "Nishanth Menon",
        "email": "nm@ti.com",
        "time": "Fri Mar 21 01:52:48 2014 -0500"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Mar 24 00:36:47 2014 -0400"
      },
      "message": "net: micrel : ks8851-ml: add vdd-supply support\n\nFew platforms use external regulator to keep the ethernet MAC supplied.\nSo, request and enable the regulator for driver functionality.\n\nFixes: 66fda75f47dc (regulator: core: Replace direct ops-\u003edisable usage)\nReported-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSuggested-by: Markus Pargmann \u003cmpa@pengutronix.de\u003e\nSigned-off-by: Nishanth Menon \u003cnm@ti.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "a34fe10750ebe524a39f97bd78ab4d232a554edb",
      "tree": "fbcb1dcbb05ea872117db209fce605e461f268ed",
      "parents": [
        "e9af8b7aba1fd5f2175ec552ec36d5efe57f6c1d"
      ],
      "author": {
        "name": "Will Deacon",
        "email": "will.deacon@arm.com",
        "time": "Fri Feb 21 17:34:57 2014 +0000"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 23 17:01:23 2014 +0100"
      },
      "message": "parisc: locks: remove redundant arch_*_relax operations\n\nNow that the arch_{spin,read,write}_relax macros default to cpu_relax(),\nremove the redundant definitions for parisc.\n\nCc: Helge Deller \u003cdeller@gmx.de\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "e9af8b7aba1fd5f2175ec552ec36d5efe57f6c1d",
      "tree": "28ae17c5299fa4ea4cb3d5c583054922297bf422",
      "parents": [
        "4b02a72a266b58268979cb6f9c5d4dd002148634"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 23 15:24:40 2014 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 23 16:57:13 2014 +0100"
      },
      "message": "parisc: wire up sys_utimes\n\nWe seem to be nearly the only platform which does not provide the\nsys_utimes syscall.  Adding it now makes our life much easier with\nuserspace applications (like dietlibc and e2fsprogs) since we then\nbehave like all other platforms too and don\u0027t need extra patches which\nare hard to get upstream anyway because we are not a mainstream\narchitecture.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\nCc: stable@vger.kernel.org # v3.13\n"
    },
    {
      "commit": "4b02a72a266b58268979cb6f9c5d4dd002148634",
      "tree": "dc77d83fb08376e633ed6c11c0ed6baac15cb3ed",
      "parents": [
        "a2fb4d782c61f77480e586578eeb4dfd27d134ea"
      ],
      "author": {
        "name": "John David Anglin",
        "email": "dave.anglin@bell.net",
        "time": "Sat Mar 01 17:41:22 2014 -0500"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 23 16:46:30 2014 +0100"
      },
      "message": "parisc: Remove unused CONFIG_PARISC_TMPALIAS code\n\nThe attached change removes the unused and experimental\nCONFIG_PARISC_TMPALIAS code. It doesn\u0027t work and I don\u0027t believe it will\never be used.\n\nSigned-off-by: John David Anglin \u003cdave.anglin@bell.net\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "a2fb4d782c61f77480e586578eeb4dfd27d134ea",
      "tree": "d58824ad68b9e632991fb73a8e5ddf081b2f3c41",
      "parents": [
        "774868c7094d35b4518be3d0e654de000a5d11fc"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 23 16:39:52 2014 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 23 16:44:42 2014 +0100"
      },
      "message": "partly revert commit 8a10bc9: parisc/sti_console: prefer Linux fonts over built-in ROM fonts\n\nSTI console is used on parisc and m68k HP machines. This patch partly reverts\nmy previous commit and as such restores the fonts for the m68k machines.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\nCc: stable@vger.kernel.org # v3.13\n"
    },
    {
      "commit": "b37199e626b31e1175fb06764c5d1d687723aac2",
      "tree": "0ac51a84f15fa251800fb40191dcf4a86c4c04c2",
      "parents": [
        "e825196d48d2b89a6ec3a8eff280098d2a78207e"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Thu Mar 20 15:18:22 2014 -0400"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 23 00:32:55 2014 -0400"
      },
      "message": "rcuwalk: recheck mount_lock after mountpoint crossing attempts\n\nWe can get false negative from __lookup_mnt() if an unrelated vfsmount\ngets moved.  In that case legitimize_mnt() is guaranteed to fail,\nand we will fall back to non-RCU walk... unless we end up running\ninto a hard error on a filesystem object we wouldn\u0027t have reached\nif not for that false negative.  IOW, delaying that check until\nthe end of pathname resolution is wrong - we should recheck right\nafter we attempt to cross the mountpoint.  We don\u0027t need to recheck\nunless we see d_mountpoint() being true - in that case even if\nwe have just raced with mount/umount, we can simply go on as if\nwe\u0027d come at the moment when the sucker wasn\u0027t a mountpoint; if we\nrun into a hard error as the result, it was a legitimate outcome.\n__lookup_mnt() returning NULL is different in that respect, since\nit might\u0027ve happened due to operation on completely unrelated\nmountpoint.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "e825196d48d2b89a6ec3a8eff280098d2a78207e",
      "tree": "719a3ab8be442c786a9d5b02276ad8aaf989075a",
      "parents": [
        "99aea68134f3c2a27b4d463c91cfa298c3efaccf"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 23 00:28:40 2014 -0400"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 23 00:28:40 2014 -0400"
      },
      "message": "make prepend_name() work correctly when called with negative *buflen\n\nIn all callchains leading to prepend_name(), the value left in *buflen\nis eventually discarded unused if prepend_name() has returned a negative.\nSo we are free to do what prepend() does, and subtract from *buflen\n*before* checking for underflow (which turns into checking the sign\nof subtraction result, of course).\n\nCc: stable@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "99aea68134f3c2a27b4d463c91cfa298c3efaccf",
      "tree": "6b57fe890caaa2ff026a9d54a42f9ac95b96fee9",
      "parents": [
        "d7a15f8d0777955986a2ab00ab181795cab14b01"
      ],
      "author": {
        "name": "Eric Biggers",
        "email": "ebiggers3@gmail.com",
        "time": "Sun Mar 16 15:47:48 2014 -0500"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 23 00:03:12 2014 -0400"
      },
      "message": "vfs: Don\u0027t let __fdget_pos() get FMODE_PATH files\n\nCommit bd2a31d522344 (\"get rid of fget_light()\") introduced the\n__fdget_pos() function, which returns the resulting file pointer and\nfdput flags combined in an \u0027unsigned long\u0027.  However, it also changed the\nbehavior to return files with FMODE_PATH set, which shouldn\u0027t happen\nbecause read(), write(), lseek(), etc. aren\u0027t allowed on such files.\nThis commit restores the old behavior.\n\nThis regression actually had no effect on read() and write() since\nFMODE_READ and FMODE_WRITE are not set on file descriptors opened with\nO_PATH, but it did cause lseek() on a file descriptor opened with O_PATH\nto fail with ESPIPE rather than EBADF.\n\nSigned-off-by: Eric Biggers \u003cebiggers3@gmail.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "d7a15f8d0777955986a2ab00ab181795cab14b01",
      "tree": "7c0d0639d278669cf81571a4c8b7aa0357d47db3",
      "parents": [
        "774868c7094d35b4518be3d0e654de000a5d11fc"
      ],
      "author": {
        "name": "Eric Biggers",
        "email": "ebiggers3@gmail.com",
        "time": "Sun Mar 16 14:24:08 2014 -0500"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Sun Mar 23 00:03:12 2014 -0400"
      },
      "message": "vfs: atomic f_pos access in llseek()\n\nCommit 9c225f2655e36a4 (\"vfs: atomic f_pos accesses as per POSIX\") changed\nseveral system calls to use fdget_pos() instead of fdget(), but missed\nsys_llseek().  Fix it.\n\nSigned-off-by: Eric Biggers \u003cebiggers3@gmail.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "774868c7094d35b4518be3d0e654de000a5d11fc",
      "tree": "dca59caccf93e19ddb184ec5abdc78e959d550ab",
      "parents": [
        "084c6c5013af3c62f1c344435214496f5ac999f2",
        "7c144bfbf89d612f0c62b7e71056fb687b068455"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 22 09:13:59 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 22 09:13:59 2014 -0700"
      },
      "message": "Merge branch \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf fixes from Ingo Molnar:\n \"Two oneliner \u0027perf bench\u0027 tooling fixes\"\n\n* \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  perf bench: Fix NULL pointer dereference in \"perf bench all\"\n  perf bench numa: Make no args mean \u0027run all tests\u0027\n"
    },
    {
      "commit": "084c6c5013af3c62f1c344435214496f5ac999f2",
      "tree": "363a3271b5110d55998b83c7e1fcb3fafc35db4b",
      "parents": [
        "08edb33c4e1b810011f21d7705811b7b9a0535f0",
        "f656d46bbb2d3ecdb71c998ccf657dccea8d19a9"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 22 09:13:13 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Mar 22 09:13:13 2014 -0700"
      },
      "message": "Merge tag \u0027fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc\n\nPull ARM SoC fixes from Olof Johansson:\n \"Only two patches this time, one to fix ethernet probe order on at91\n  (better fix with proper device aliasing will be done for 3.15, this is\n  stop-gap), and one update to MAINTAINERS due to Freescale moving their\n  repo to kernel.org\"\n\n* tag \u0027fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:\n  ARM: at91: fix network interface ordering for sama5d36\n  MAINTAINERS: update IMX kernel git tree\n"
    },
    {
      "commit": "08edb33c4e1b810011f21d7705811b7b9a0535f0",
      "tree": "60180e188c9c9dfc5055cbaf2e1eccaadcaa8e22",
      "parents": [
        "708f04d2abf4e90abee61d9ffb1f165038017ecf",
        "004e5cf743086990e5fc04a14437b3966d7fa9a2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:32:28 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:32:28 2014 -0700"
      },
      "message": "Merge branch \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes from Dave Airlie:\n \"Some final few intel fixes, all regressions, all stable cc, and one\n  exynos oops fixer.\n\n  The biggest is probably the intel display error irqs one, but it seems\n  to fix a few crashes on startup, and one use after free in drm core\"\n\n* \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux:\n  drm/exynos: Fix (more) freeing issues in exynos_drm_drv.c\n  drm/i915: Disable stolen memory when DMAR is active\n  Revert \"drm/i915: don\u0027t touch the VDD when disabling the panel\"\n  drm: Fix use-after-free in the shadow-attache exit code\n  drm/i915: Don\u0027t enable display error interrupts from the start\n  drm/i915: Fix scanline counter fixup on BDW\n  drm/i915: Add a workaround for HSW scanline counter weirdness\n  drm/i915: Fix PSR programming\n"
    },
    {
      "commit": "708f04d2abf4e90abee61d9ffb1f165038017ecf",
      "tree": "c9a25407c8011c821da54ea51a4965c0e2a4f44d",
      "parents": [
        "11d4616bd07f38d496bd489ed8fad1dc4d928823"
      ],
      "author": {
        "name": "Dave Jones",
        "email": "davej@redhat.com",
        "time": "Thu Mar 20 15:03:58 2014 -0600"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:32:06 2014 -0700"
      },
      "message": "block: free q-\u003eflush_rq in blk_init_allocated_queue error paths\n\nCommit 7982e90c3a57 (\"block: fix q-\u003eflush_rq NULL pointer crash on\ndm-mpath flush\") moved an allocation to blk_init_allocated_queue(), but\nneglected to free that allocation on the error paths that follow.\n\nSigned-off-by: Dave Jones \u003cdavej@fedoraproject.org\u003e\nAcked-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@fb.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "11d4616bd07f38d496bd489ed8fad1dc4d928823",
      "tree": "764239232ca4beac8797f274baf924e497f39151",
      "parents": [
        "477cc48484ea0bfdee9e8319259b35a0ec03f332"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:11:17 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:11:17 2014 -0700"
      },
      "message": "futex: revert back to the explicit waiter counting code\n\nSrikar Dronamraju reports that commit b0c29f79ecea (\"futexes: Avoid\ntaking the hb-\u003elock if there\u0027s nothing to wake up\") causes java threads\ngetting stuck on futexes when runing specjbb on a power7 numa box.\n\nThe cause appears to be that the powerpc spinlocks aren\u0027t using the same\nticket lock model that we use on x86 (and other) architectures, which in\nturn result in the \"spin_is_locked()\" test in hb_waiters_pending()\noccasionally reporting an unlocked spinlock even when there are pending\nwaiters.\n\nSo this reinstates Davidlohr Bueso\u0027s original explicit waiter counting\ncode, which I had convinced Davidlohr to drop in favor of figuring out\nthe pending waiters by just using the existing state of the spinlock and\nthe wait queue.\n\nReported-and-tested-by: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\nOriginal-code-by: Davidlohr Bueso \u003cdavidlohr@hp.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "477cc48484ea0bfdee9e8319259b35a0ec03f332",
      "tree": "95a45bf408c3d7dec964855f67287cfcd162ad65",
      "parents": [
        "7e09e738afd21ef99f047425fc0b0c9be8b03254",
        "87291347c49dc40aa339f587b209618201c2e527"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:09:30 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:09:30 2014 -0700"
      },
      "message": "Merge tag \u0027trace-fixes-v3.14-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull trace fix from Steven Rostedt:\n \"Vaibhav Nagarnaik discovered that since 3.10 a clean-up patch made the\n  array index in the trace event format bogus.\n\n  He supplied an elegant solution that uses __stringify() and also\n  removes the need for the event_storage and event_storage_mutex and\n  also cuts off a few K of overhead from the trace events\"\n\n* tag \u0027trace-fixes-v3.14-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n  tracing: Fix array size mismatch in format string\n"
    },
    {
      "commit": "7e09e738afd21ef99f047425fc0b0c9be8b03254",
      "tree": "423cb848a7d6875509b99720d4afcfa208bac078",
      "parents": [
        "3fb725c48b93c0a152174b6dbbc1029b5e734c7b"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hughd@google.com",
        "time": "Thu Mar 20 21:52:17 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Mar 20 22:09:09 2014 -0700"
      },
      "message": "mm: fix swapops.h:131 bug if remap_file_pages raced migration\n\nAdd remove_linear_migration_ptes_from_nonlinear(), to fix an interesting\nlittle include/linux/swapops.h:131 BUG_ON(!PageLocked) found by trinity:\nindicating that remove_migration_ptes() failed to find one of the\nmigration entries that was temporarily inserted.\n\nThe problem comes from remap_file_pages()\u0027s switch from vma_interval_tree\n(good for inserting the migration entry) to i_mmap_nonlinear list (no good\nfor locating it again); but can only be a problem if the remap_file_pages()\nrange does not cover the whole of the vma (zap_pte() clears the range).\n\nremove_migration_ptes() needs a file_nonlinear method to go down the\ni_mmap_nonlinear list, applying linear location to look for migration\nentries in those vmas too, just in case there was this race.\n\nThe file_nonlinear method does need rmap_walk_control.arg to do this;\nbut it never needed vma passed in - vma comes from its own iteration.\n\nReported-and-tested-by: Dave Jones \u003cdavej@redhat.com\u003e\nReported-and-tested-by: Sasha Levin \u003csasha.levin@oracle.com\u003e\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "b74d3feccc3f5e6e4cf98cf88daa2e0cdbc4ca62",
      "tree": "651a6736040bd5084916ed9c461817fb0cbbe52b",
      "parents": [
        "f518338b16038beeb73e155e60d0f70beb9379f4",
        "f9b8c4c8baded129535d82d74df8e87a7a369f54"
      ],
      "author": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 20 17:29:02 2014 -0400"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 20 17:29:02 2014 -0400"
      },
      "message": "Merge branch \u0027fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch\n\nJesse Gross says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nOpen vSwitch\n\nFour small fixes for net/3.14. I realize that these are late in the\ncycle - just got back from vacation.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "f518338b16038beeb73e155e60d0f70beb9379f4",
      "tree": "247f3590b7e0b750cdaee366c854ae684275ca45",
      "parents": [
        "65886f439ab0fdc2dff20d1fa87afb98c6717472"
      ],
      "author": {
        "name": "Nicolas Dichtel",
        "email": "nicolas.dichtel@6wind.com",
        "time": "Wed Mar 19 17:47:51 2014 +0100"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Mar 20 16:24:28 2014 -0400"
      },
      "message": "ip6mr: fix mfc notification flags\n\nCommit 812e44dd1829 (\"ip6mr: advertise new mfc entries via rtnl\") reuses the\nfunction ip6mr_fill_mroute() to notify mfc events.\nBut this function was used only for dump and thus was always setting the\nflag NLM_F_MULTI, which is wrong in case of a single notification.\n\nLibraries like libnl will wait forever for NLMSG_DONE.\n\nCC: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\nAcked-by: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    }
  ],
  "next": "65886f439ab0fdc2dff20d1fa87afb98c6717472"
}