)]}' { "log": [ { "commit": "b9311277c28fab089ee0816307035c3a5f546bcf", "tree": "16cc3ce94be533f607ffacf8f4057c4cdeac489a", "parents": [ "665c7a83a1a64582c4393dd45430c848150a10dd" ], "author": { "name": "Anders Kaseorg", "email": "andersk@MIT.EDU", "time": "Sun Jul 15 17:14:25 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Thu Jul 19 08:58:56 2012 -0700" }, "message": "fifo: Do not restart open() if it already found a partner\n\ncommit 05d290d66be6ef77a0b962ebecf01911bd984a78 upstream.\n\nIf a parent and child process open the two ends of a fifo, and the\nchild immediately exits, the parent may receive a SIGCHLD before its\nopen() returns. In that case, we need to make sure that open() will\nreturn successfully after the SIGCHLD handler returns, instead of\nthrowing EINTR or being restarted. Otherwise, the restarted open()\nwould incorrectly wait for a second partner on the other end.\n\nThe following test demonstrates the EINTR that was wrongly thrown from\nthe parent’s open(). Change .sa_flags \u003d 0 to .sa_flags \u003d SA_RESTART\nto see a deadlock instead, in which the restarted open() waits for a\nsecond reader that will never come. (On my systems, this happens\npretty reliably within about 5 to 500 iterations. Others report that\nit manages to loop ~forever sometimes; YMMV.)\n\n #include \u003csys/stat.h\u003e\n #include \u003csys/types.h\u003e\n #include \u003csys/wait.h\u003e\n #include \u003cfcntl.h\u003e\n #include \u003csignal.h\u003e\n #include \u003cstdio.h\u003e\n #include \u003cstdlib.h\u003e\n #include \u003cunistd.h\u003e\n\n #define CHECK(x) do if ((x) \u003d\u003d -1) {perror(#x); abort();} while(0)\n\n void handler(int signum) {}\n\n int main()\n {\n struct sigaction act \u003d {.sa_handler \u003d handler, .sa_flags \u003d 0};\n CHECK(sigaction(SIGCHLD, \u0026act, NULL));\n CHECK(mknod(\"fifo\", S_IFIFO | S_IRWXU, 0));\n for (;;) {\n int fd;\n pid_t pid;\n putc(\u0027.\u0027, stderr);\n CHECK(pid \u003d fork());\n if (pid \u003d\u003d 0) {\n CHECK(fd \u003d open(\"fifo\", O_RDONLY));\n _exit(0);\n }\n CHECK(fd \u003d open(\"fifo\", O_WRONLY));\n CHECK(close(fd));\n CHECK(waitpid(pid, NULL, 0));\n }\n }\n\nThis is what I suspect was causing the Git test suite to fail in\nt9010-svn-fe.sh:\n\n\thttp://bugs.debian.org/678852\n\nSigned-off-by: Anders Kaseorg \u003candersk@mit.edu\u003e\nReviewed-by: Jonathan Nieder \u003cjrnieder@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "e2ea6a3e16a36e26af184f781ede7a12f6cff8b5", "tree": "368ca1b2ff510ee878e351bb8b7384066dcab9ff", "parents": [ "175063bccaefa0b6a4934ffaf3a9639a4a7616d0" ], "author": { "name": "Jeff Moyer", "email": "jmoyer@redhat.com", "time": "Thu Jul 12 09:43:14 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Thu Jul 19 08:58:55 2012 -0700" }, "message": "block: fix infinite loop in __getblk_slow\n\ncommit 91f68c89d8f35fe98ea04159b9a3b42d0149478f upstream.\n\nCommit 080399aaaf35 (\"block: don\u0027t mark buffers beyond end of disk as\nmapped\") exposed a bug in __getblk_slow that causes mount to hang as it\nloops infinitely waiting for a buffer that lies beyond the end of the\ndisk to become uptodate.\n\nThe problem was initially reported by Torsten Hilbrich here:\n\n https://lkml.org/lkml/2012/6/18/54\n\nand also reported independently here:\n\n http://www.sysresccd.org/forums/viewtopic.php?f\u003d13\u0026t\u003d4511\n\nand then Richard W.M. Jones and Marcos Mello noted a few separate\nbugzillas also associated with the same issue. This patch has been\nconfirmed to fix:\n\n https://bugzilla.redhat.com/show_bug.cgi?id\u003d835019\n\nThe main problem is here, in __getblk_slow:\n\n for (;;) {\n struct buffer_head * bh;\n int ret;\n\n bh \u003d __find_get_block(bdev, block, size);\n if (bh)\n return bh;\n\n ret \u003d grow_buffers(bdev, block, size);\n if (ret \u003c 0)\n return NULL;\n if (ret \u003d\u003d 0)\n free_more_memory();\n }\n\n__find_get_block does not find the block, since it will not be marked as\nmapped, and so grow_buffers is called to fill in the buffers for the\nassociated page. I believe the for (;;) loop is there primarily to\nretry in the case of memory pressure keeping grow_buffers from\nsucceeding. However, we also continue to loop for other cases, like the\nblock lying beond the end of the disk. So, the fix I came up with is to\nonly loop when grow_buffers fails due to memory allocation issues\n(return value of 0).\n\nThe attached patch was tested by myself, Torsten, and Rich, and was\nfound to resolve the problem in call cases.\n\nSigned-off-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\nReported-and-Tested-by: Torsten Hilbrich \u003ctorsten.hilbrich@secunet.com\u003e\nTested-by: Richard W.M. Jones \u003crjones@redhat.com\u003e\nReviewed-by: Josh Boyer \u003cjwboyer@redhat.com\u003e\n[ Jens is on vacation, taking this directly - Linus ]\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "dfd45e89bc8040c45ad5df89fb01d7d0138ec953", "tree": "b1431982b85ae956972e5f4ac0ef265f3c1f8227", "parents": [ "763c71b1319c56272e42cf6ada6994131f0193a7" ], "author": { "name": "Dave Jones", "email": "davej@redhat.com", "time": "Fri Jul 13 13:35:36 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Thu Jul 19 08:58:54 2012 -0700" }, "message": "Remove easily user-triggerable BUG from generic_setlease\n\ncommit 8d657eb3b43861064d36241e88d9d61c709f33f0 upstream.\n\nThis can be trivially triggered from userspace by passing in something unexpected.\n\n kernel BUG at fs/locks.c:1468!\n invalid opcode: 0000 [#1] SMP\n RIP: 0010:generic_setlease+0xc2/0x100\n Call Trace:\n __vfs_setlease+0x35/0x40\n fcntl_setlease+0x76/0x150\n sys_fcntl+0x1c6/0x810\n system_call_fastpath+0x1a/0x1f\n\nSigned-off-by: Dave Jones \u003cdavej@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "bb2b649805527af1d823f871e7e3a95c4e6a018e", "tree": "b247be4fc93d60ee914b2f78a02e97956682dc7e", "parents": [ "7ad71f960f0f6e06cbded278809674afc515036a" ], "author": { "name": "Luis Henriques", "email": "luis.henriques@canonical.com", "time": "Wed Jul 11 14:02:10 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:45 2012 -0700" }, "message": "ocfs2: fix NULL pointer dereference in __ocfs2_change_file_space()\n\ncommit a4e08d001f2e50bb8b3c4eebadcf08e5535f02ee upstream.\n\nAs ocfs2_fallocate() will invoke __ocfs2_change_file_space() with a NULL\nas the first parameter (file), it may trigger a NULL pointer dereferrence\ndue to a missing check.\n\nAddresses http://bugs.launchpad.net/bugs/1006012\n\nSigned-off-by: Luis Henriques \u003cluis.henriques@canonical.com\u003e\nReported-by: Bret Towe \u003cmagnade@gmail.com\u003e\nTested-by: Bret Towe \u003cmagnade@gmail.com\u003e\nCc: Sunil Mushran \u003csunil.mushran@oracle.com\u003e\nAcked-by: Joel Becker \u003cjlbec@evilplan.org\u003e\nAcked-by: Mark Fasheh \u003cmfasheh@suse.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "0fa627b15c842095b5147f381fa1943a6a46bb01", "tree": "4293cc0156d458583829aefff1eee7daa56df5fe", "parents": [ "7a08b440fa93e036968102597c8a2ab809a9bdc4" ], "author": { "name": "Bob Liu", "email": "lliubbo@gmail.com", "time": "Wed Jul 11 14:02:35 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:45 2012 -0700" }, "message": "fs: ramfs: file-nommu: add SetPageUptodate()\n\ncommit fea9f718b3d68147f162ed2d870183ce5e0ad8d8 upstream.\n\nThere is a bug in the below scenario for !CONFIG_MMU:\n\n 1. create a new file\n 2. mmap the file and write to it\n 3. read the file can\u0027t get the correct value\n\nBecause\n\n sys_read() -\u003e generic_file_aio_read() -\u003e simple_readpage() -\u003e clear_page()\n\nwhich causes the page to be zeroed.\n\nAdd SetPageUptodate() to ramfs_nommu_expand_for_mapping() so that\ngeneric_file_aio_read() do not call simple_readpage().\n\nSigned-off-by: Bob Liu \u003clliubbo@gmail.com\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nCc: Greg Ungerer \u003cgerg@uclinux.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "2c07f25ea7800adb36cd8da9b58c4ecd3fc3d064", "tree": "b312e3b679b544de20569f8e31dd1469e8a72be1", "parents": [ "5318edefb61eddf91d4c4a089644fcee3ccfda62" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue Jun 12 15:24:40 2012 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:42 2012 -0700" }, "message": "splice: fix racy pipe-\u003ebuffers uses\n\ncommit 047fe3605235888f3ebcda0c728cb31937eadfe6 upstream.\n\nDave Jones reported a kernel BUG at mm/slub.c:3474! triggered\nby splice_shrink_spd() called from vmsplice_to_pipe()\n\ncommit 35f3d14dbbc5 (pipe: add support for shrinking and growing pipes)\nadded capability to adjust pipe-\u003ebuffers.\n\nProblem is some paths don\u0027t hold pipe mutex and assume pipe-\u003ebuffers\ndoesn\u0027t change for their duration.\n\nFix this by adding nr_pages_max field in struct splice_pipe_desc, and\nuse it in place of pipe-\u003ebuffers where appropriate.\n\nsplice_shrink_spd() loses its struct pipe_inode_info argument.\n\nReported-by: Dave Jones \u003cdavej@redhat.com\u003e\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Jens Axboe \u003caxboe@kernel.dk\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Tom Herbert \u003ctherbert@google.com\u003e\nTested-by: Dave Jones \u003cdavej@redhat.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n[bwh: Backported to 3.2:\n - Adjust context in vmsplice_to_pipe()\n - Update one more call to splice_shrink_spd(), from skb_splice_bits()]\nSigned-off-by: Ben Hutchings \u003cben@decadent.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "f183282bb88ffa944449cf3a24a649c754d9e7af", "tree": "efcdf28280b9d42d895f83de42a7b0847269638e", "parents": [ "ee92389156c2cdb45b94866186a4174858b820cd" ], "author": { "name": "Stanislav Kinsbursky", "email": "skinsbursky@parallels.com", "time": "Mon Jun 25 16:40:10 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:40 2012 -0700" }, "message": "NFS: hard-code init_net for NFS callback transports\n\nupstream commit 12918b10d59e975fd5241eef03ef9e6d5ea3dcfe.\n\nIn case of destroying mount namespace on child reaper exit, nsproxy is zeroed\nto the point already. So, dereferencing of it is invalid.\nThis patch hard-code \"init_net\" for all network namespace references for NFS\ncallback services. This will be fixed with proper NFS callback\ncontainerization.\n\nSigned-off-by: Stanislav Kinsbursky \u003cskinsbursky@parallels.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "ee92389156c2cdb45b94866186a4174858b820cd", "tree": "ce025a20a1c679a511d174df4a0818f8146cbf0a", "parents": [ "10762419cafd82a9a3a6f68bef54c29f1af75842" ], "author": { "name": "Stanislav Kinsbursky", "email": "skinsbursky@parallels.com", "time": "Mon Jun 25 16:40:09 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:39 2012 -0700" }, "message": "SUNRPC: move per-net operations from svc_destroy()\n\nupstream commit 786185b5f8abefa6a8a16695bb4a59c164d5a071.\n\nThe idea is to separate service destruction and per-net operations,\nbecause these are two different things and the mix looks ugly.\n\nNotes:\n\n1) For NFS server this patch looks ugly (sorry for that). But these\nplace will be rewritten soon during NFSd containerization.\n\n2) LockD per-net counter increase int lockd_up() was moved prior to\nmake_socks() to make lockd_down_net() call safe in case of error.\n\nSigned-off-by: Stanislav Kinsbursky \u003cskinsbursky@parallels.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "10762419cafd82a9a3a6f68bef54c29f1af75842", "tree": "45e0a1689e4ffa4c94fc09c9c72d454fab1e6639", "parents": [ "0bbc9d1b4b011e83ba65852b1d652561c7f562f1" ], "author": { "name": "Stanislav Kinsbursky", "email": "skinsbursky@parallels.com", "time": "Mon Jun 25 16:40:08 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:39 2012 -0700" }, "message": "SUNRPC: new svc_bind() routine introduced\n\nupstream commit 9793f7c88937e7ac07305ab1af1a519225836823.\n\nThis new routine is responsible for service registration in a specified\nnetwork context.\n\nThe idea is to separate service creation from per-net operations.\n\nNote also: since registering service with svc_bind() can fail, the\nservice will be destroyed and during destruction it will try to\nunregister itself from rpcbind. In this case unregistration has to be\nskipped.\n\nSigned-off-by: Stanislav Kinsbursky \u003cskinsbursky@parallels.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "0bbc9d1b4b011e83ba65852b1d652561c7f562f1", "tree": "0a6e76d8b7e1d473ca92f5358ff2c5fab7f16ba1", "parents": [ "0e924ae7ac646cf8aa95a3bc5671d19f920417c5" ], "author": { "name": "Stanislav Kinsbursky", "email": "skinsbursky@parallels.com", "time": "Mon Jun 25 16:40:07 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:39 2012 -0700" }, "message": "Lockd: pass network namespace to creation and destruction routines\n\nupstream commit e3f70eadb7dddfb5a2bb9afff7abfc6ee17a29d0.\n\nv2: dereference of most probably already released nlm_host removed in\nnlmclnt_done() and reclaimer().\n\nThese routines are called from locks reclaimer() kernel thread. This thread\nworks in \"init_net\" network context and currently relays on persence on lockd\nthread and it\u0027s per-net resources. Thus lockd_up() and lockd_down() can\u0027t relay\non current network context. So let\u0027s pass corrent one into them.\n\nSigned-off-by: Stanislav Kinsbursky \u003cskinsbursky@parallels.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "b6855f6d78cee5d72adc8f07be2646dd83c171c3", "tree": "ba67ce52faf0e783dc649f013a0a16fd83b16450", "parents": [ "b6d7c70709e324e074d3a69249980bba1529fd01" ], "author": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Tue Jun 12 11:17:01 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:26 2012 -0700" }, "message": "eCryptfs: Properly check for O_RDONLY flag before doing privileged open\n\ncommit 9fe79d7600497ed8a95c3981cbe5b73ab98222f0 upstream.\n\nIf the first attempt at opening the lower file read/write fails,\neCryptfs will retry using a privileged kthread. However, the privileged\nretry should not happen if the lower file\u0027s inode is read-only because a\nread/write open will still be unsuccessful.\n\nThe check for determining if the open should be retried was intended to\nbe based on the access mode of the lower file\u0027s open flags being\nO_RDONLY, but the check was incorrectly performed. This would cause the\nopen to be retried by the privileged kthread, resulting in a second\nfailed open of the lower file. This patch corrects the check to\ndetermine if the open request should be handled by the privileged\nkthread.\n\nSigned-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\nReported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nAcked-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "b6d7c70709e324e074d3a69249980bba1529fd01", "tree": "42baeff62ad963da452231a44da38a70893556ec", "parents": [ "6dfa530c0b33f1ccd2d8b103a2dea6ec559de66c" ], "author": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Mon Jun 11 10:21:34 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:26 2012 -0700" }, "message": "eCryptfs: Fix lockdep warning in miscdev operations\n\ncommit 60d65f1f07a7d81d3eb3b91fc13fca80f2fdbb12 upstream.\n\nDon\u0027t grab the daemon mutex while holding the message context mutex.\nAddresses this lockdep warning:\n\n ecryptfsd/2141 is trying to acquire lock:\n (\u0026ecryptfs_msg_ctx_arr[i].mux){+.+.+.}, at: [\u003cffffffffa029c213\u003e] ecryptfs_miscdev_read+0x143/0x470 [ecryptfs]\n\n but task is already holding lock:\n (\u0026(*daemon)-\u003emux){+.+...}, at: [\u003cffffffffa029c2ec\u003e] ecryptfs_miscdev_read+0x21c/0x470 [ecryptfs]\n\n which lock already depends on the new lock.\n\n the existing dependency chain (in reverse order) is:\n\n -\u003e #1 (\u0026(*daemon)-\u003emux){+.+...}:\n [\u003cffffffff810a3b8d\u003e] lock_acquire+0x9d/0x220\n [\u003cffffffff8151c6da\u003e] __mutex_lock_common+0x5a/0x4b0\n [\u003cffffffff8151cc64\u003e] mutex_lock_nested+0x44/0x50\n [\u003cffffffffa029c5d7\u003e] ecryptfs_send_miscdev+0x97/0x120 [ecryptfs]\n [\u003cffffffffa029b744\u003e] ecryptfs_send_message+0x134/0x1e0 [ecryptfs]\n [\u003cffffffffa029a24e\u003e] ecryptfs_generate_key_packet_set+0x2fe/0xa80 [ecryptfs]\n [\u003cffffffffa02960f8\u003e] ecryptfs_write_metadata+0x108/0x250 [ecryptfs]\n [\u003cffffffffa0290f80\u003e] ecryptfs_create+0x130/0x250 [ecryptfs]\n [\u003cffffffff811963a4\u003e] vfs_create+0xb4/0x120\n [\u003cffffffff81197865\u003e] do_last+0x8c5/0xa10\n [\u003cffffffff811998f9\u003e] path_openat+0xd9/0x460\n [\u003cffffffff81199da2\u003e] do_filp_open+0x42/0xa0\n [\u003cffffffff81187998\u003e] do_sys_open+0xf8/0x1d0\n [\u003cffffffff81187a91\u003e] sys_open+0x21/0x30\n [\u003cffffffff81527d69\u003e] system_call_fastpath+0x16/0x1b\n\n -\u003e #0 (\u0026ecryptfs_msg_ctx_arr[i].mux){+.+.+.}:\n [\u003cffffffff810a3418\u003e] __lock_acquire+0x1bf8/0x1c50\n [\u003cffffffff810a3b8d\u003e] lock_acquire+0x9d/0x220\n [\u003cffffffff8151c6da\u003e] __mutex_lock_common+0x5a/0x4b0\n [\u003cffffffff8151cc64\u003e] mutex_lock_nested+0x44/0x50\n [\u003cffffffffa029c213\u003e] ecryptfs_miscdev_read+0x143/0x470 [ecryptfs]\n [\u003cffffffff811887d3\u003e] vfs_read+0xb3/0x180\n [\u003cffffffff811888ed\u003e] sys_read+0x4d/0x90\n [\u003cffffffff81527d69\u003e] system_call_fastpath+0x16/0x1b\n\nSigned-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "6dfa530c0b33f1ccd2d8b103a2dea6ec559de66c", "tree": "df4f53ead7bb20b2ca77c173cd5ed8eb5d155762", "parents": [ "6d442d8f7e1a6920aad3d3f3ea544f6f8e6898b8" ], "author": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Mon Jun 11 09:24:11 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:26 2012 -0700" }, "message": "eCryptfs: Gracefully refuse miscdev file ops on inherited/passed files\n\ncommit 8dc6780587c99286c0d3de747a2946a76989414a upstream.\n\nFile operations on /dev/ecryptfs would BUG() when the operations were\nperformed by processes other than the process that originally opened the\nfile. This could happen with open files inherited after fork() or file\ndescriptors passed through IPC mechanisms. Rather than calling BUG(), an\nerror code can be safely returned in most situations.\n\nIn ecryptfs_miscdev_release(), eCryptfs still needs to handle the\nrelease even if the last file reference is being held by a process that\ndidn\u0027t originally open the file. ecryptfs_find_daemon_by_euid() will not\nbe successful, so a pointer to the daemon is stored in the file\u0027s\nprivate_data. The private_data pointer is initialized when the miscdev\nfile is opened and only used when the file is released.\n\nhttps://launchpad.net/bugs/994247\n\nSigned-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\nReported-by: Sasha Levin \u003clevinsasha928@gmail.com\u003e\nTested-by: Sasha Levin \u003clevinsasha928@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "47e38c598d1a3bf7d01c540c46d2039708337622", "tree": "15da175bffe684b17ba0dccda3966d0760fdb6a3", "parents": [ "f632881de16f8c3133cd1b0866937f50fa2e9156" ], "author": { "name": "Junxiao Bi", "email": "junxiao.bi@oracle.com", "time": "Wed Jun 27 17:09:55 2012 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:24 2012 -0700" }, "message": "ocfs2: clear unaligned io flag when dio fails\n\ncommit 3e5d3c35a68c9a933bdbdd8685bd1a205b57e806 upstream.\n\nThe unaligned io flag is set in the kiocb when an unaligned\ndio is issued, it should be cleared even when the dio fails,\nor it may affect the following io which are using the same\nkiocb.\n\nSigned-off-by: Junxiao Bi \u003cjunxiao.bi@oracle.com\u003e\nSigned-off-by: Joel Becker \u003cjlbec@evilplan.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "881aa809b005753a64fe9409c1d229912cc925ce", "tree": "05511e50d03ffe9a6ddd5719b8bf19ef654e6610", "parents": [ "30716eeec35ce79a2140b4814bd417e02da08450" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 07 10:17:00 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:23 2012 -0700" }, "message": "vfs: make O_PATH file descriptors usable for \u0027fchdir()\u0027\n\ncommit 332a2e1244bd08b9e3ecd378028513396a004a24 upstream.\n\nWe already use them for openat() and friends, but fchdir() also wants to\nbe able to use O_PATH file descriptors. This should make it comparable\nto the O_SEARCH of Solaris. In particular, O_PATH allows you to access\n(not-quite-open) a directory you don\u0027t have read persmission to, only\nexecute permission.\n\nNoticed during development of multithread support for ksh93.\n\nReported-by: ольга крыжановская \u003colga.kryzhanovska@gmail.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "d5b8efa118c7bf1b88e0f6f0185fbdd3c0ee67ed", "tree": "3cddec5b63740d5b01a86a7b70b706bd722afcc2", "parents": [ "2f0d20928b7360267ec80bb393971752f2351463" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Wed Jun 20 14:35:28 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:09 2012 -0700" }, "message": "NFS: Force the legacy idmapper to be single threaded\n\ncommit b1027439dff844675f6c0df97a1b1d190791a699 upstream.\n\nIt was initially coded under the assumption that there would only be one\nrequest at a time, so use a lock to enforce this requirement..\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "21017faf87a93117ca7a14aa8f0dd2f315fdeb08", "tree": "dbdd62ebc0e69df1992f2c14cba75fce37a2bef1", "parents": [ "a7c60136e5bf96bc58a27ab9822a2e48ed768429" ], "author": { "name": "Konstantin Khlebnikov", "email": "khlebnikov@openvz.org", "time": "Wed Jun 20 12:53:01 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:08 2012 -0700" }, "message": "mm: correctly synchronize rss-counters at exit/exec\n\ncommit 4fe7efdbdfb1c7e7a7f31decfd831c0f31d37091 upstream.\n\ndo_exit() and exec_mmap() call sync_mm_rss() before mm_release() does\nput_user(clear_child_tid) which can update task-\u003erss_stat and thus make\nmm-\u003erss_stat inconsistent. This triggers the \"BUG:\" printk in check_mm().\n\nLet\u0027s fix this bug in the safest way, and optimize/cleanup this later.\n\nReported-by: Markus Trippelsdorf \u003cmarkus@trippelsdorf.de\u003e\nSigned-off-by: Konstantin Khlebnikov \u003ckhlebnikov@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "dbc904343caccba0ce47a51379c4a743da504a8a", "tree": "ae28a633804949dd3f485b2538c1c4f1defddac0", "parents": [ "becbf4ebbb41acee5c0dcf2851af6bd84964fc00" ], "author": { "name": "Chris Mason", "email": "chris.mason@fusionio.com", "time": "Mon Jul 02 15:29:53 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:04:05 2012 -0700" }, "message": "Btrfs: run delayed directory updates during log replay\n\ncommit b6305567e7d31b0bec1b8cb9ec0cadd7f7086f5f upstream.\n\nWhile we are resolving directory modifications in the\ntree log, we are triggering delayed metadata updates to\nthe filesystem btrees.\n\nThis commit forces the delayed updates to run so the\nreplay code can find any modifications done. It stops\nus from crashing because the directory deleltion replay\nexpects items to be removed immediately from the tree.\n\nSigned-off-by: Chris Mason \u003cchris.mason@fusionio.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "f4367f246e048e7fac50769d1d1447306a984f4e", "tree": "9f75608b0be4ad61a6f54f510f8949c389ba7bab", "parents": [ "9471356d7e58fbc3e1d307d911432261398abb56" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Mon Jul 02 07:24:25 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:03:54 2012 -0700" }, "message": "cifs: when server doesn\u0027t set CAP_LARGE_READ_X, cap default rsize at MaxBufferSize\n\ncommit ec01d738a1691dfc85b96b9f796020267a7be577 upstream.\n\nWhen the server doesn\u0027t advertise CAP_LARGE_READ_X, then MS-CIFS states\nthat you must cap the size of the read at the client\u0027s MaxBufferSize.\nUnfortunately, testing with many older servers shows that they often\ncan\u0027t service a read larger than their own MaxBufferSize.\n\nSince we can\u0027t assume what the server will do in this situation, we must\nbe conservative here for the default. When the server can\u0027t do large\nreads, then assume that it can\u0027t satisfy any read larger than its\nMaxBufferSize either.\n\nLuckily almost all modern servers can do large reads, so this won\u0027t\naffect them. This is really just for older win9x and OS/2 era servers.\nAlso, note that this patch just governs the default rsize. The admin can\nalways override this if he so chooses.\n\nReported-by: David H. Durgee \u003cdhdurgee@acm.org\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steven French \u003csfrench@w500smf.none\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "9471356d7e58fbc3e1d307d911432261398abb56", "tree": "170a8f29a922861154c4e9884fb08fee0406b38d", "parents": [ "0f9c37b38800c35e83f43949d93bffefff5b43af" ], "author": { "name": "Suresh Jayaraman", "email": "sjayaraman@suse.com", "time": "Tue Jun 12 07:15:50 2012 +0530" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:03:52 2012 -0700" }, "message": "cifs: fix parsing of password mount option\n\ncommit e73f843a3235a19de38359c91586e9eadef12238 upstream.\n\nThe double delimiter check that allows a comma in the password parsing code is\nunconditional. We set \"tmp_end\" to the end of the string and we continue to\ncheck for double delimiter. In the case where the password doesn\u0027t contain a\ncomma we end up setting tmp_end to NULL and eventually setting \"options\" to\n\"end\". This results in the premature termination of the options string and hence\nthe values of UNCip and UNC are being set to NULL. This results in mount failure\nwith \"Connecting to DFS root not implemented yet\" error.\n\nThis error is usually not noticable as we have password as the last option in\nthe superblock mountdata. But when we call expand_dfs_referral() from\ncifs_mount() and try to compose mount options for the submount, the resulting\nmountdata will be of the form\n\n \",ver\u003d1,user\u003dfoo,pass\u003dbar,ip\u003dx.x.x.x,unc\u003d\\\\server\\share\"\n\nand hence results in the above error. This bug has been seen with older NAS\nservers running Samba 3.0.24.\n\nFix this by moving the double delimiter check inside the conditional loop.\n\nChanges since -v1\n\n - removed the wrong strlen() micro optimization.\n\nSigned-off-by: Suresh Jayaraman \u003csjayaraman@suse.com\u003e\nAcked-by: Sachin Prabhu \u003csprabhu@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "4836ee563d65bb492f907cbe267a5761b9693e4d", "tree": "a511d2b95441b7e36c9c8ea2ef70a36a0c7161f0", "parents": [ "b3b9f9cd546c4b3d72b0a95ba4ca29c840fce28e" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Jun 27 21:23:07 2012 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:03:51 2012 -0700" }, "message": "udf: Fortify loading of sparing table\n\ncommit 1df2ae31c724e57be9d7ac00d78db8a5dabdd050 upstream.\n\nAdd sanity checks when loading sparing table from disk to avoid accessing\nunallocated memory or writing to it.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "b3b9f9cd546c4b3d72b0a95ba4ca29c840fce28e", "tree": "51efd28037725f043eb319413e0aebecea312683", "parents": [ "132a45d8bdc699d831dcc51cfd4db7238d183b42" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Jun 27 20:20:22 2012 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:03:51 2012 -0700" }, "message": "udf: Avoid run away loop when partition table length is corrupted\n\ncommit adee11b2085bee90bd8f4f52123ffb07882d6256 upstream.\n\nCheck provided length of partition table so that (possibly maliciously)\ncorrupted partition table cannot cause accessing data beyond current buffer.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "132a45d8bdc699d831dcc51cfd4db7238d183b42", "tree": "c0d4cc02244f2671e11ea0a4db0cf37bc63e052b", "parents": [ "615fb8b018599cfe16b466b1a1c98cf1c3ecaf0e" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Jun 27 20:08:44 2012 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:03:51 2012 -0700" }, "message": "udf: Use \u0027ret\u0027 instead of abusing \u0027i\u0027 in udf_load_logicalvol()\n\ncommit cb14d340ef1737c24125dd663eff77734a482d47 upstream.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "615fb8b018599cfe16b466b1a1c98cf1c3ecaf0e", "tree": "6fcd1e66ea44c39f678743441c97a2c41b9a2e6b", "parents": [ "a4ab3b05fe0320bad00a1ba320d6fd4b9cd29643" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Wed Jun 20 12:52:57 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Mon Jul 16 09:03:51 2012 -0700" }, "message": "nilfs2: ensure proper cache clearing for gc-inodes\n\ncommit fbb24a3a915f105016f1c828476be11aceac8504 upstream.\n\nA gc-inode is a pseudo inode used to buffer the blocks to be moved by\ngarbage collection.\n\nBlock caches of gc-inodes must be cleared every time a garbage collection\nfunction (nilfs_clean_segments) completes. Otherwise, stale blocks\nbuffered in the caches may be wrongly reused in successive calls of the GC\nfunction.\n\nFor user files, this is not a problem because their gc-inodes are\ndistinguished by a checkpoint number as well as an inode number. They\nnever buffer different blocks if either an inode number, a checkpoint\nnumber, or a block offset differs.\n\nHowever, gc-inodes of sufile, cpfile and DAT file can store different data\nfor the same block offset. Thus, the nilfs_clean_segments function can\nmove incorrect block for these meta-data files if an old block is cached.\nI found this is really causing meta-data corruption in nilfs.\n\nThis fixes the issue by ensuring cache clear of gc-inodes and resolves\nreported GC problems including checkpoint file corruption, b-tree\ncorruption, and the following warning during GC.\n\n nilfs_palloc_freev: entry number 307234 already freed.\n ...\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\nTested-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "231afb7c7ded9bf798b71efdf576615e91c367af", "tree": "152a14dd70998400a0c66516ed4aa08a43864e4e", "parents": [ "152a4f421da62406d8836aa2c9e53b7b3e405e59" ], "author": { "name": "Matthew Garrett", "email": "mjg@redhat.com", "time": "Sun Jun 17 17:05:25 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Jun 22 11:36:57 2012 -0700" }, "message": "hfsplus: fix bless ioctl when used with hardlinks\n\ncommit 7dea9665fee828fb56db3bae5b9685d9fa006d33 upstream.\n\nHFS+ doesn\u0027t really implement hard links - instead, hardlinks are indicated\nby a magic file type which refers to an indirect node in a hidden\ndirectory. The spec indicates that stat() should return the inode number\nof the indirect node, but it turns out that this doesn\u0027t satisfy the\nfirmware when it\u0027s looking for a bootloader - it wants the catalog ID of\nthe hardlink file instead. Fix up this case.\n\nSigned-off-by: Matthew Garrett \u003cmjg@redhat.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "152a4f421da62406d8836aa2c9e53b7b3e405e59", "tree": "37e0d51d101ed42a6b5ceef7687b40860abcc130", "parents": [ "d64cbbc9603861015cd616327f63b10394c0418a" ], "author": { "name": "Janne Kalliomäki", "email": "janne@tuxera.com", "time": "Sun Jun 17 17:05:24 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Jun 22 11:36:57 2012 -0700" }, "message": "hfsplus: fix overflow in sector calculations in hfsplus_submit_bio\n\ncommit a6dc8c04218eb752ff79cdc24a995cf51866caed upstream.\n\nThe variable io_size was unsigned int, which caused the wrong sector number\nto be calculated after aligning it. This then caused mount to fail with big\nvolumes, as backup volume header information was searched from a\nwrong sector.\n\nSigned-off-by: Janne Kalliomäki \u003cjanne@tuxera.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "d946d96cd89b13fc354b6f7ed2c4de3e72c87d8e", "tree": "1755a0c65e7be1598a5e7ee7fbddee75ca4c335e", "parents": [ "5e74cfe1feb56745cecce815adf6fd2c7dfb6a58" ], "author": { "name": "J. Bruce Fields", "email": "bfields@redhat.com", "time": "Tue Jun 12 08:28:48 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Jun 22 11:36:54 2012 -0700" }, "message": "nfsd4: BUG_ON(!is_spin_locked()) no good on UP kernels\n\ncommit bc2df47a408f2d64cf81bcfd0f6e3e14c84cb0ab upstream.\n\nMost frequent symptom was a BUG triggering in expire_client, with the\nserver locking up shortly thereafter.\n\nIntroduced by 508dc6e110c6dbdc0bbe84298ccfe22de7538486 \"nfsd41:\nfree_session/free_client must be called under the client_lock\".\n\nCc: Benny Halevy \u003cbhalevy@tonian.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "5e74cfe1feb56745cecce815adf6fd2c7dfb6a58", "tree": "9057b07d5378285f5d650e0f0f1582915317ac7b", "parents": [ "5f1d81e9fb6c05e8ab89a610bb63cdc8f0f66d81" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Jun 08 10:58:09 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Jun 22 11:36:54 2012 -0700" }, "message": "NFSv4: Fix unnecessary delegation returns in nfs4_do_open\n\ncommit 2d0dbc6ae8a5194aaecb9cfffb9053f38fce8b86 upstream.\n\nWhile nfs4_do_open() expects the fmode argument to be restricted to\ncombinations of FMODE_READ and FMODE_WRITE, both nfs4_atomic_open()\nand nfs4_proc_create will pass the nfs_open_context-\u003emode,\nwhich contains the full fmode_t.\n\nThis patch ensures that nfs4_do_open strips the other fmode_t bits,\nfixing a problem in which the nfs4_do_open call would result in an\nunnecessary delegation return.\n\nReported-by: Fred Isaman \u003ciisaman@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "9abcb7517f13aa54152bee6370538b8f56893349", "tree": "a15e24c622f773c714a72f3b1113054cfd9f7d89", "parents": [ "d5b9a38383178758ddf671b7a5551afab4e504b2" ], "author": { "name": "Pavel Shilovsky", "email": "piastry@etersoft.ru", "time": "Thu May 10 19:49:38 2012 +0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 17 11:21:29 2012 -0700" }, "message": "fuse: fix stat call on 32 bit platforms\n\ncommit 45c72cd73c788dd18c8113d4a404d6b4a01decf1 upstream.\n\nNow we store attr-\u003eino at inode-\u003ei_ino, return attr-\u003eino at the\nfirst time and then return inode-\u003ei_ino if the attribute timeout\nisn\u0027t expired. That\u0027s wrong on 32 bit platforms because attr-\u003eino\nis 64 bit and inode-\u003ei_ino is 32 bit in this case.\n\nFix this by saving 64 bit ino in fuse_inode structure and returning\nit every time we call getattr. Also squash attr-\u003eino into inode-\u003ei_ino\nexplicitly.\n\nSigned-off-by: Pavel Shilovsky \u003cpiastry@etersoft.ru\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "4e050dfccceeb07641b8cf50618eb71e5208a6dc", "tree": "c4ab8a5a76896d58c1197edf7537f070d51b2bec", "parents": [ "e82c95f9f0cb025da8e1af6df1c29080a78d0f84" ], "author": { "name": "Josef Bacik", "email": "josef@redhat.com", "time": "Wed May 23 16:10:14 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 17 11:21:24 2012 -0700" }, "message": "Btrfs: fall back to non-inline if we don\u0027t have enough space\n\ncommit 2adcac1a7331d93a17285804819caa96070b231f upstream.\n\nIf cow_file_range_inline fails with ENOSPC we abort the transaction which\nisn\u0027t very nice. This really shouldn\u0027t be happening anyways but there\u0027s no\nsense in making it a horrible error when we can easily just go allocate\nnormal data space for this stuff. Thanks,\n\nSigned-off-by: Josef Bacik \u003cjosef@redhat.com\u003e\nAcked-by: Chris Mason \u003cchris.mason@fusionio.com\u003e\nCc: Alexandre Oliva \u003coliva@lsd.ic.unicamp.br\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "c573b3798f5fd9e24f9ec23a39c4915c6e024faf", "tree": "a62c1612ed42b718b5e803240210635d41abe735", "parents": [ "d913c02b0a172d5dca6280da5b17a407d69bbce4" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Jun 07 18:56:06 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 17 11:21:23 2012 -0700" }, "message": "ext4: fix the free blocks calculation for ext3 file systems w/ uninit_bg\n\ncommit b0dd6b70f0fda17ae9762fbb72d98e40a4f66556 upstream.\n\nExt3 filesystems that are converted to use as many ext4 file system\nfeatures as possible will enable uninit_bg to speed up e2fsck times.\nThese file systems will have a native ext3 layout of inode tables and\nblock allocation bitmaps (as opposed to ext4\u0027s flex_bg layout).\nUnfortunately, in these cases, when first allocating a block in an\nuninitialized block group, ext4 would incorrectly calculate the number\nof free blocks in that block group, and then errorneously report that\nthe file system was corrupt:\n\nEXT4-fs error (device vdd): ext4_mb_generate_buddy:741: group 30, 32254 clusters in bitmap, 32258 in gd\n\nThis problem can be reproduced via:\n\n mke2fs -q -t ext4 -O ^flex_bg /dev/vdd 5g\n mount -t ext4 /dev/vdd /mnt\n fallocate -l 4600m /mnt/test\n\nThe problem was caused by a bone headed mistake in the check to see if a\nparticular metadata block was part of the block group.\n\nMany thanks to Kees Cook for finding and bisecting the buggy commit\nwhich introduced this bug (commit fd034a84e1, present since v3.2).\n\nReported-by: Sander Eikelenboom \u003clinux@eikelenboom.it\u003e\nReported-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nTested-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "6ab997e0b5440eee1bc76bb9d83aa86e3966adb2", "tree": "b980792fe1f7e9f2709d6f7a1e1ce089490b99a9", "parents": [ "dadce2efa18e448bb42678efdc72a819bceaa9b7" ], "author": { "name": "Tao Ma", "email": "boyu.mt@taobao.com", "time": "Thu Jun 07 19:04:19 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:19 2012 +0900" }, "message": "ext4: don\u0027t set i_flags in EXT4_IOC_SETFLAGS\n\ncommit b22b1f178f6799278d3178d894f37facb2085765 upstream.\n\nCommit 7990696 uses the ext4_{set,clear}_inode_flags() functions to\nchange the i_flags automatically but fails to remove the error setting\nof i_flags. So we still have the problem of trashing state flags.\nFix this by removing the assignment.\n\nSigned-off-by: Tao Ma \u003cboyu.mt@taobao.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "cf9ab4c62be7837c2f007cd51ab3604ca0620070", "tree": "81b5d80733c87e81e8c153d11e937db2ce28d8c2", "parents": [ "cd08f77d07306f1294a78f13de5f55348c21df5b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jun 04 11:00:45 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:18 2012 +0900" }, "message": "vfs: Fix /proc/\u003ctid\u003e/fdinfo/\u003cfd\u003e file handling\n\ncommit 0640113be25d283e0ff77a9f041e1242182387f0 upstream.\n\nCyrill Gorcunov reports that I broke the fdinfo files with commit\n30a08bf2d31d (\"proc: move fd symlink i_mode calculations into\ntid_fd_revalidate()\"), and he\u0027s quite right.\n\nThe tid_fd_revalidate() function is not just used for the \u003ctid\u003e/fd\nsymlinks, it\u0027s also used for the \u003ctid\u003e/fdinfo/\u003cfd\u003e files, and the\npermission model for those are different.\n\nSo do the dynamic symlink permission handling just for symlinks, making\nthe fdinfo files once more appear as the proper regular files they are.\n\nOf course, Al Viro argued (probably correctly) that we shouldn\u0027t do the\nsymlink permission games at all, and make the symlinks always just be\nthe normal \u0027lrwxrwxrwx\u0027. That would have avoided this issue too, but\nsince somebody noticed that the permissions had changed (which was the\nreason for that original commit 30a08bf2d31d in the first place), people\ndo apparently use this feature.\n\n[ Basically, you can use the symlink permission data as a cheap \"fdinfo\"\n replacement, since you see whether the file is open for reading and/or\n writing by just looking at st_mode of the symlink. So the feature\n does make sense, even if the pain it has caused means we probably\n shouldn\u0027t have done it to begin with. ]\n\nReported-and-tested-by: Cyrill Gorcunov \u003cgorcunov@openvz.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "f940c5366eaaea9630ba643a036c2df321f66db3", "tree": "b41af7952ab645e2e8525ba0ed15df2149f78116", "parents": [ "d2e926777a97fc29ebfa409584ed4e790f2e68bf" ], "author": { "name": "Salman Qazi", "email": "sqazi@google.com", "time": "Thu May 31 23:52:14 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:16 2012 +0900" }, "message": "ext4: remove mb_groups before tearing down the buddy_cache\n\ncommit 95599968d19db175829fb580baa6b68939b320fb upstream.\n\nWe can\u0027t have references held on pages in the s_buddy_cache while we are\ntrying to truncate its pages and put the inode. All the pages must be\ngone before we reach clear_inode. This can only be gauranteed if we\ncan prevent new users from grabbing references to s_buddy_cache\u0027s pages.\n\nThe original bug can be reproduced and the bug fix can be verified by:\n\nwhile true; do mount -t ext4 /dev/ram0 /export/hda3/ram0; \\\n\tumount /export/hda3/ram0; done \u0026\n\nwhile true; do cat /proc/fs/ext4/ram0/mb_groups; done\n\nSigned-off-by: Salman Qazi \u003csqazi@google.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "d2e926777a97fc29ebfa409584ed4e790f2e68bf", "tree": "bd452ee3b42c2aa64cca4ffa7ca0b70dbb56336c", "parents": [ "5720f8de251bbf23fc9f4191d6ed5734110b7d67" ], "author": { "name": "Salman Qazi", "email": "sqazi@google.com", "time": "Thu May 31 23:51:27 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:16 2012 +0900" }, "message": "ext4: add ext4_mb_unload_buddy in the error path\n\ncommit 02b7831019ea4e7994968c84b5826fa8b248ffc8 upstream.\n\next4_free_blocks fails to pair an ext4_mb_load_buddy with a matching\next4_mb_unload_buddy when it fails a memory allocation.\n\nSigned-off-by: Salman Qazi \u003csqazi@google.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "5720f8de251bbf23fc9f4191d6ed5734110b7d67", "tree": "b4cdfad6bd419c59d552aee6d14ba3765867f310", "parents": [ "d7286a55eb1088df8f8d8c1bde34e0342c63804c" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu May 31 23:46:01 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:16 2012 +0900" }, "message": "ext4: don\u0027t trash state flags in EXT4_IOC_SETFLAGS\n\ncommit 79906964a187c405db72a3abc60eb9b50d804fbc upstream.\n\nIn commit 353eb83c we removed i_state_flags with 64-bit longs, But\nwhen handling the EXT4_IOC_SETFLAGS ioctl, we replace i_flags\ndirectly, which trashes the state flags which are stored in the high\n32-bits of i_flags on 64-bit platforms. So use the the\next4_{set,clear}_inode_flags() functions which use atomic bit\nmanipulation functions instead.\n\nReported-by: Tao Ma \u003cboyu.mt@taobao.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "d7286a55eb1088df8f8d8c1bde34e0342c63804c", "tree": "7407ebb8f6f2e7dcaeae8078936017dd963a1003", "parents": [ "c40a4bd1e80db2b3af7b064b6eece0e689be60a9" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed May 30 23:00:16 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:16 2012 +0900" }, "message": "ext4: add missing save_error_info() to ext4_error()\n\ncommit f3fc0210c0fc91900766c995f089c39170e68305 upstream.\n\nThe ext4_error() function is missing a call to save_error_info().\nSince this is the function which marks the file system as containing\nan error, this oversight (which was introduced in 2.6.36) is quite\nsignificant, and should be backported to older stable kernels with\nhigh urgency.\n\nReported-by: Ken Sumrall \u003cksumrall@google.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: ksumrall@google.com\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "c40a4bd1e80db2b3af7b064b6eece0e689be60a9", "tree": "4dfa6bac93d96e90a109736d1d0425ed3d86c72c", "parents": [ "a9ea4481033197d29c606a935b24336a96cb3082" ], "author": { "name": "Andreas Dilger", "email": "adilger@dilger.ca", "time": "Mon May 28 17:02:25 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:16 2012 +0900" }, "message": "ext4: disallow hard-linked directory in ext4_lookup\n\ncommit 7e936b737211e6b54e34b71a827e56b872e958d8 upstream.\n\nA hard-linked directory to its parent can cause the VFS to deadlock,\nand is a sign of a corrupted file system. So detect this case in\next4_lookup(), before the rmdir() lockup scenario can take place.\n\nSigned-off-by: Andreas Dilger \u003cadilger@dilger.ca\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "a9ea4481033197d29c606a935b24336a96cb3082", "tree": "d1236a43ca42a124c272e9fb5ca3a15bbd573d90", "parents": [ "3c7f096206323ae4c9ad3db62fece0b19df21532" ], "author": { "name": "Haogang Chen", "email": "haogangchen@gmail.com", "time": "Mon May 28 14:21:55 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:16 2012 +0900" }, "message": "ext4: fix potential integer overflow in alloc_flex_gd()\n\ncommit 967ac8af4475ce45474800709b12137aa7634c77 upstream.\n\nIn alloc_flex_gd(), when flexbg_size is large, kmalloc size would\noverflow and flex_gd-\u003egroups would point to a buffer smaller than\nexpected, causing OOB accesses when it is used.\n\nNote that in ext4_resize_fs(), flexbg_size is calculated using\nsbi-\u003es_log_groups_per_flex, which is read from the disk and only bounded\nto [1, 31]. The patch returns NULL for too large flexbg_size.\n\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Haogang Chen \u003chaogangchen@gmail.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "3c7f096206323ae4c9ad3db62fece0b19df21532", "tree": "04c1ab9e49145719e32813f954cd8a7c159b29af", "parents": [ "2c693b1d2fd1789a8b52f3644d41547b1ba9a26e" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon May 28 14:17:25 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:15 2012 +0900" }, "message": "ext4: force ro mount if ext4_setup_super() fails\n\ncommit 7e84b6216467b84cd332c8e567bf5aa113fd2f38 upstream.\n\nIf ext4_setup_super() fails i.e. due to a too-high revision,\nthe error is logged in dmesg but the fs is not mounted RO as\nindicated.\n\nTested by:\n\n# mkfs.ext4 -r 4 /dev/sdb6\n# mount /dev/sdb6 /mnt/test\n# dmesg | grep \"too high\"\n[164919.759248] EXT4-fs (sdb6): revision level too high, forcing read-only mode\n# grep sdb6 /proc/mounts\n/dev/sdb6 /mnt/test2 ext4 rw,seclabel,relatime,data\u003dordered 0 0\n\nReviewed-by: Andreas Dilger \u003cadilger@whamcloud.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "2c693b1d2fd1789a8b52f3644d41547b1ba9a26e", "tree": "145a3893edbde6a3b87beb5b1699f7012784530e", "parents": [ "978041f2a946a4616f7cdb60f6d453ffa83298c1" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Mon May 28 14:16:57 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:15 2012 +0900" }, "message": "ext4: fix potential NULL dereference in ext4_free_inodes_counts()\n\ncommit bb3d132a24cd8bf5e7773b2d9f9baa58b07a7dae upstream.\n\nThe ext4_get_group_desc() function returns NULL on error, and\next4_free_inodes_count() function dereferences it without checking.\nThere is a check on the next line, but it\u0027s too late.\n\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "c8c0b91569f01d7713c843245c538b51f733646a", "tree": "c4af2b9289b3b063a0098f406c5ff97db1d07020", "parents": [ "15de0eade174f9aade57a4080f1bac1a36e45e83" ], "author": { "name": "Dmitry Kasatkin", "email": "dmitry.kasatkin@intel.com", "time": "Tue May 29 11:02:21 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:12 2012 +0900" }, "message": "vfs: increment iversion when a file is truncated\n\ncommit 799243a389bde0de10fa21ca1ca453d2fe538b85 upstream.\n\nWhen a file is truncated with truncate()/ftruncate() and then closed,\niversion is not updated. This patch uses ATTR_SIZE flag as an indication\nto increment iversion.\n\nMimi said:\n\nOn fput(), i_version is used to detect and flag files that have changed\nand need to be re-measured in the IMA measurement policy. When a file\nis truncated with truncate()/ftruncate() and then closed, i_version is\nnot updated. As a result, although the file has changed, it will not be\nre-measured and added to the IMA measurement list on subsequent access.\n\nSigned-off-by: Dmitry Kasatkin \u003cdmitry.kasatkin@intel.com\u003e\nAcked-by: Mimi Zohar \u003czohar@us.ibm.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "15de0eade174f9aade57a4080f1bac1a36e45e83", "tree": "3351b21436f75bbd3cfc83cfde55cff18da3384f", "parents": [ "75df3ae2060eb00b909ac65fc4d2798c4e53969a" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue May 29 22:03:48 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:12 2012 +0900" }, "message": "vfs: umount_tree() might be called on subtree that had never made it\n\ncommit 63d37a84ab6004c235314ffd7a76c5eb28c2fae0 upstream.\n\n__mnt_make_shortterm() in there undoes the effect of __mnt_make_longterm()\nwe\u0027d done back when we set -\u003emnt_ns non-NULL; it should not be done to\nvfsmounts that had never gone through commit_tree() and friends. Kudos to\nlczerner for catching that one...\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "27b0a5338ea769a35f1c8346cd07a3c938c919e7", "tree": "cf256d2fecb028dd7db205d9e2604c882a5882c0", "parents": [ "c780682d113ef97164359897f4463585b4c27524" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon May 28 11:36:28 2012 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:09 2012 +0900" }, "message": "NFSv4: Map NFS4ERR_SHARE_DENIED into an EACCES error instead of EIO\n\ncommit fb13bfa7e1bcfdcfdece47c24b62f1a1cad957e9 upstream.\n\nIf a file OPEN is denied due to a share lock, the resulting\nNFS4ERR_SHARE_DENIED is currently mapped to the default EIO.\nThis patch adds a more appropriate mapping, and brings Linux\ninto line with what Solaris 10 does.\n\nSee https://bugzilla.kernel.org/show_bug.cgi?id\u003d43286\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "c780682d113ef97164359897f4463585b4c27524", "tree": "3a8e1278191eb864849c8fbce606e7ca2d830af4", "parents": [ "2b06dfb55b78e57734a696324ce7d593b1c084be" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Mon May 14 22:45:28 2012 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:09 2012 +0900" }, "message": "NFS: kmalloc() doesn\u0027t return an ERR_PTR()\n\ncommit 5abc03cd919535c61b813f2319cb38326a41e810 upstream.\n\nObviously we should check for NULL here instead of IS_ERR().\n\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "27ab30eb965b9d26c5811f5505d95c8ede08c580", "tree": "2b74dd981ce60b10b4876f4f027f712315b08507", "parents": [ "444b7bce3152cc6ef1fb599331d053c9c40eb374" ], "author": { "name": "Shirish Pargaonkar", "email": "shirishpargaonkar@gmail.com", "time": "Mon May 21 09:20:12 2012 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:07 2012 +0900" }, "message": "cifs: fix oops while traversing open file list (try #4)\n\ncommit 2c0c2a08bed7a3b791f88d09d16ace56acb3dd98 upstream.\n\nWhile traversing the linked list of open file handles, if the identfied\nfile handle is invalid, a reopen is attempted and if it fails, we\nresume traversing where we stopped and cifs can oops while accessing\ninvalid next element, for list might have changed.\n\nSo mark the invalid file handle and attempt reopen if no\nvalid file handle is found in rest of the list.\nIf reopen fails, move the invalid file handle to the end of the list\nand start traversing the list again from the begining.\nRepeat this four times before giving up and returning an error if\nfile reopen keeps failing.\n\nSigned-off-by: Shirish Pargaonkar \u003cshirishpargaonkar@gmail.com\u003e\nReviewed-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "444b7bce3152cc6ef1fb599331d053c9c40eb374", "tree": "56798496e5e3af94f5e13667fe9a974cd6ddf63f", "parents": [ "784c9700444d2ebbace0a1d3a3ac48c3d5852164" ], "author": { "name": "Shirish Pargaonkar", "email": "shirishpargaonkar@gmail.com", "time": "Tue May 15 10:19:16 2012 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:07 2012 +0900" }, "message": "cifs: Include backup intent search flags during searches {try #2)\n\ncommit 2608bee744a92d60d15ff4e6e0b913d8b406aedd upstream.\n\nAs observed and suggested by Tushar Gosavi...\n\n---------\nreaddir calls these function to send TRANS2_FIND_FIRST and\nTRANS2_FIND_NEXT command to the server. The current cifs module is\nnot specifying CIFS_SEARCH_BACKUP_SEARCH flag while sending these\ncommand when backupuid/backupgid is specified. This can be resolved\nby specifying CIFS_SEARCH_BACKUP_SEARCH flag.\n---------\n\nReported-and-Tested-by: Tushar Gosavi \u003ctugosavi@in.ibm.com\u003e\nSigned-off-by: Shirish Pargaonkar \u003cshirishpargaonkar@gmail.com\u003e\nAcked-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "3634d6fec1f80012110962d58dbd396aaa99d960", "tree": "c48557e7ef092600ac0c3c2e73c234f49a615375", "parents": [ "3876c722319d9b8463b9e82394d81f6d0c45021e" ], "author": { "name": "Sasha Levin", "email": "levinsasha928@gmail.com", "time": "Tue May 29 15:06:15 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:06 2012 +0900" }, "message": "mm: fix NULL ptr deref when walking hugepages\n\ncommit 08fa29d916c6e271ad13978cd993e7238c68db97 upstream.\n\nA missing validation of the value returned by find_vma() could cause a\nNULL ptr dereference when walking the pagetable.\n\nThis is triggerable from usermode by a simple user by trying to read a\npage info out of /proc/pid/pagemap which doesn\u0027t exist.\n\nIntroduced by commit 025c5b2451e4 (\"thp: optimize away unnecessary page\ntable locking\").\n\nSigned-off-by: Sasha Levin \u003clevinsasha928@gmail.com\u003e\nReviewed-by: Naoya Horiguchi \u003cn-horiguchi@ah.jp.nec.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "edb302ba464c6151ddb24ba4fb18c8e0a0ef9f7c", "tree": "e9494bf0cf0fb50600a65a9db23d240c2cdca9be", "parents": [ "f18a511703b6667edee4703fdd8b73ca18e232a4" ], "author": { "name": "Boaz Harrosh", "email": "bharrosh@panasas.com", "time": "Wed May 16 14:22:21 2012 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sun Jun 10 00:36:05 2012 +0900" }, "message": "exofs: Fix CRASH on very early IO errors.\n\ncommit 6abe4a87f7bc7978705c386dbba0ca0c7790b3ec upstream.\n\nIf at exofs_fill_super() we had an early termination\ndo to any error, like an IO error while reading the\nsuper-block. We would crash inside exofs_free_sbi().\n\nThis is because sbi-\u003eoc.numdevs was set to 1, before\nwe actually have a device table at all.\n\nFix it by moving the sbi-\u003eoc.numdevs \u003d 1 to after the\nallocation of the device table.\n\nReported-by: Johannes Schild \u003cJSchild@gmx.de\u003e\n\nSigned-off-by: Boaz Harrosh \u003cbharrosh@panasas.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "64126fa12b904d5dae0b7e33fac540c46faf67c5", "tree": "3d0d92f25324a19775f6f8949b7c4e35e412454f", "parents": [ "4e1468e1f16dcd141bb256fd7320687b5b8bc975" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 21 16:06:20 2012 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri Jun 01 15:18:14 2012 +0800" }, "message": "vfs: make AIO use the proper rw_verify_area() area helpers\n\ncommit a70b52ec1aaeaf60f4739edb1b422827cb6f3893 upstream.\n\nWe had for some reason overlooked the AIO interface, and it didn\u0027t use\nthe proper rw_verify_area() helper function that checks (for example)\nmandatory locking on the file, and that the size of the access doesn\u0027t\ncause us to overflow the provided offset limits etc.\n\nInstead, AIO did just the security_file_permission() thing (that\nrw_verify_area() also does) directly.\n\nThis fixes it to do all the proper helper functions, which not only\nmeans that now mandatory file locking works with AIO too, we can\nactually remove lines of code.\n\nReported-by: Manish Honap \u003cmanish_honap_vit@yahoo.co.in\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n" }, { "commit": "14e931a264498fbd4baef07ee0644e347252393b", "tree": "c202cd976f478fed3d7d786521ab3e32abdc0f9e", "parents": [ "a2ae9787568ac50978c03ce67bfb79ad2e100cca", "05c69d298c96703741cac9a5cbbf6c53bd55a6e2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 19 10:12:17 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 19 10:12:17 2012 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.dk/linux-block\n\nPull block layer fixes from Jens Axboe:\n \"A few small, but important fixes. Most of them are marked for stable\n as well\n\n - Fix failure to release a semaphore on error path in mtip32xx.\n - Fix crashable condition in bio_get_nr_vecs().\n - Don\u0027t mark end-of-disk buffers as mapped, limit it to i_size.\n - Fix for build problem with CONFIG_BLOCK\u003dn on arm at least.\n - Fix for a buffer overlow on UUID partition printing.\n - Trivial removal of unused variables in dac960.\"\n\n* \u0027for-linus\u0027 of git://git.kernel.dk/linux-block:\n block: fix buffer overflow when printing partition UUIDs\n Fix blkdev.h build errors when BLOCK\u003dn\n bio allocation failure due to bio_get_nr_vecs()\n block: don\u0027t mark buffers beyond end of disk as mapped\n mtip32xx: release the semaphore on an error path\n dac960: Remove unused variables from DAC960_CreateProcEntries()\n" }, { "commit": "73f1f5dd3ee3ec6e20768d831d9759b0330fad0e", "tree": "22f8498c47467f468dce6c3bca0eb1f3a0c9ade0", "parents": [ "30a08bf2d31d275c6fc71dd1811342777e95c831", "93c2d656c7120e29de8df5bc17bb2a97664104e9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 18 15:56:25 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 18 15:56:25 2012 -0700" }, "message": "Merge branch \u0027akpm\u0027 (Andrew\u0027s patch-bomb)\n\nMerge misc fixes from Andrew Morton.\n\n* emailed from Andrew Morton \u003cakpm@linux-foundation.org\u003e: (4 patches)\n frv: delete incorrect task prototypes causing compile fail\n slub: missing test for partial pages flush work in flush_all()\n fs, proc: fix ABBA deadlock in case of execution attempt of map_files/ entries\n drivers/rtc/rtc-pl031.c: configure correct wday for 2000-01-01\n" }, { "commit": "30a08bf2d31d275c6fc71dd1811342777e95c831", "tree": "541e4d439b73accdaa48bf48fc77a1915555439b", "parents": [ "3d9944978e0bb6c98b901949cb7a22256e48b23d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 18 11:32:15 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 18 14:06:17 2012 -0700" }, "message": "proc: move fd symlink i_mode calculations into tid_fd_revalidate()\n\nInstead of doing the i_mode calculations at proc_fd_instantiate() time,\nmove them into tid_fd_revalidate(), which is where the other inode state\n(notably uid/gid information) is updated too.\n\nOtherwise we\u0027ll end up with stale i_mode information if an fd is re-used\nwhile the dentry still hangs around. Not that anything really *cares*\n(symlink permissions don\u0027t really matter), but Tetsuo Handa noticed that\nthe owner read/write bits don\u0027t always match the state of the\nreadability of the file descriptor, and we _used_ to get this right a\nlong time ago in a galaxy far, far away.\n\nBesides, aside from fixing an ugly detail (that has apparently been this\nway since commit 61a28784028e: \"proc: Remove the hard coded inode\nnumbers\" in 2006), this removes more lines of code than it adds. And it\njust makes sense to update i_mode in the same place we update i_uid/gid.\n\nAl Viro correctly points out that we could just do the inode fill in the\ninode iops -\u003egetattr() function instead. However, that does require\nsomewhat slightly more invasive changes, and adds yet *another* lookup\nof the file descriptor. We need to do the revalidate() for other\nreasons anyway, and have the file descriptor handy, so we might as well\nfill in the information at this point.\n\nReported-by: Tetsuo Handa \u003cpenguin-kernel@i-love.sakura.ne.jp\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nAcked-by: Eric Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "eb94cd96e05d6c65a07937e66a04ea265c1b767d", "tree": "0432ec8db57798b5212fa2200fe1293b01bd3361", "parents": [ "c0a5f4a05af588a0f9951f8d24e2564b09501918" ], "author": { "name": "Cyrill Gorcunov", "email": "gorcunov@openvz.org", "time": "Thu May 17 17:03:25 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 17 18:00:51 2012 -0700" }, "message": "fs, proc: fix ABBA deadlock in case of execution attempt of map_files/ entries\n\nmap_files/ entries are never supposed to be executed, still curious\nminds might try to run them, which leads to the following deadlock\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\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ INFO: possible circular locking dependency detected ]\n 3.4.0-rc4-24406-g841e6a6 #121 Not tainted\n -------------------------------------------------------\n bash/1556 is trying to acquire lock:\n (\u0026sb-\u003es_type-\u003ei_mutex_key#8){+.+.+.}, at: do_lookup+0x267/0x2b1\n\n but task is already holding lock:\n (\u0026sig-\u003ecred_guard_mutex){+.+.+.}, at: prepare_bprm_creds+0x2d/0x69\n\n which lock already depends on the new lock.\n\n the existing dependency chain (in reverse order) is:\n\n -\u003e #1 (\u0026sig-\u003ecred_guard_mutex){+.+.+.}:\n validate_chain+0x444/0x4f4\n __lock_acquire+0x387/0x3f8\n lock_acquire+0x12b/0x158\n __mutex_lock_common+0x56/0x3a9\n mutex_lock_killable_nested+0x40/0x45\n lock_trace+0x24/0x59\n proc_map_files_lookup+0x5a/0x165\n __lookup_hash+0x52/0x73\n do_lookup+0x276/0x2b1\n walk_component+0x3d/0x114\n do_last+0xfc/0x540\n path_openat+0xd3/0x306\n do_filp_open+0x3d/0x89\n do_sys_open+0x74/0x106\n sys_open+0x21/0x23\n tracesys+0xdd/0xe2\n\n -\u003e #0 (\u0026sb-\u003es_type-\u003ei_mutex_key#8){+.+.+.}:\n check_prev_add+0x6a/0x1ef\n validate_chain+0x444/0x4f4\n __lock_acquire+0x387/0x3f8\n lock_acquire+0x12b/0x158\n __mutex_lock_common+0x56/0x3a9\n mutex_lock_nested+0x40/0x45\n do_lookup+0x267/0x2b1\n walk_component+0x3d/0x114\n link_path_walk+0x1f9/0x48f\n path_openat+0xb6/0x306\n do_filp_open+0x3d/0x89\n open_exec+0x25/0xa0\n do_execve_common+0xea/0x2f9\n do_execve+0x43/0x45\n sys_execve+0x43/0x5a\n stub_execve+0x6c/0xc0\n\nThis is because prepare_bprm_creds grabs task-\u003esignal-\u003ecred_guard_mutex\nand when do_lookup happens we try to grab task-\u003esignal-\u003ecred_guard_mutex\nagain in lock_trace.\n\nFix it using plain ptrace_may_access() helper in proc_map_files_lookup()\nand in proc_map_files_readdir() instead of lock_trace(), the caller must\nbe CAP_SYS_ADMIN granted anyway.\n\nSigned-off-by: Cyrill Gorcunov \u003cgorcunov@openvz.org\u003e\nReported-by: Sasha Levin \u003clevinsasha928@gmail.com\u003e\nCc: Konstantin Khlebnikov \u003ckhlebnikov@openvz.org\u003e\nCc: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Dave Jones \u003cdavej@redhat.com\u003e\nCc: Vasiliy Kulikov \u003csegoon@openwall.com\u003e\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "dfae359f083fac3a884e10b46ebe0f262a9bd97a", "tree": "25a547a4cb2b1cdac99766c62442c1f73bc50582", "parents": [ "39d6411b7df566715138c8c9f7fa00227f4ae75b", "531c8ff0d472295f5ef5d1bd306115c81a84889e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 16 14:22:38 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 16 14:22:38 2012 -0700" }, "message": "Merge git://git.samba.org/sfrench/cifs-2.6\n\nPull CIFS fix from Jeff Layton\n\n* git://git.samba.org/sfrench/cifs-2.6:\n cifs: fix misspelling of \"forcedirectio\"\n" }, { "commit": "531c8ff0d472295f5ef5d1bd306115c81a84889e", "tree": "1b7c638d73d60bba5c27094566e5d07e3e6b249b", "parents": [ "36be50515fe2aef61533b516fa2576a2c7fe7664" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed May 16 07:12:26 2012 -0400" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Wed May 16 11:26:25 2012 -0500" }, "message": "cifs: fix misspelling of \"forcedirectio\"\n\n...and add a \"directio\" synonym since that\u0027s what the manpage has\nalways advertised.\n\nAcked-by: Sachin Prabhu \u003csprabhu@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "9ff00d58a915b6747ba2e843ab2d04c712b4dc32", "tree": "3c4bed740dae7b6115b89387d766962d6dad7cd6", "parents": [ "36be50515fe2aef61533b516fa2576a2c7fe7664", "b027274d2e3a332683b73f15e5cea79c240bc9a3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 13 11:33:09 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 13 11:33:09 2012 -0700" }, "message": "Merge tag \u0027for-linus-3.4-20120513\u0027 of git://git.infradead.org/linux-mtd\n\nPull three MTD fixes from David Woodhouse:\n - Fix a lock ordering deadlock in JFFS2\n - Fix an oops in the dataflash driver, triggered by a dummy call to test\n whether it has OTP functionality.\n - Fix request_mem_region() failure on amsdelta NAND driver.\n\n* tag \u0027for-linus-3.4-20120513\u0027 of git://git.infradead.org/linux-mtd:\n mtd: ams-delta: fix request_mem_region() failure\n jffs2: Fix lock acquisition order bug in gc path\n mtd: fix oops in dataflash driver\n" }, { "commit": "f908ee9463b09ddd05e1c1a0111132212dc05fac", "tree": "af80b227ec2049aa23fc3a8867e70fd685a4d34a", "parents": [ "080399aaaf3531f5b8761ec0ac30ff98891e8686" ], "author": { "name": "Bernd Schubert", "email": "bernd.schubert@itwm.fraunhofer.de", "time": "Fri May 11 16:36:44 2012 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri May 11 16:45:12 2012 +0200" }, "message": "bio allocation failure due to bio_get_nr_vecs()\n\nThe number of bio_get_nr_vecs() is passed down via bio_alloc() to\nbvec_alloc_bs(), which fails the bio allocation if\nnr_iovecs \u003e BIO_MAX_PAGES. For the underlying caller this causes an\nunexpected bio allocation failure.\nLimiting to queue_max_segments() is not sufficient, as max_segments\nalso might be very large.\n\nbvec_alloc_bs(gfp_mask, nr_iovecs, ) \u003d\u003e NULL when nr_iovecs \u003e BIO_MAX_PAGES\nbio_alloc_bioset(gfp_mask, nr_iovecs, ...)\nbio_alloc(GFP_NOIO, nvecs)\nxfs_alloc_ioend_bio()\n\nSigned-off-by: Bernd Schubert \u003cbernd.schubert@itwm.fraunhofer.de\u003e\nCc: stable@kernel.org\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "080399aaaf3531f5b8761ec0ac30ff98891e8686", "tree": "28ba8f41bb67a58992bc426dac7002983bc64e8f", "parents": [ "a09ba13eefb155a00d8d50008a0c0a2406985ddd" ], "author": { "name": "Jeff Moyer", "email": "jmoyer@redhat.com", "time": "Fri May 11 16:34:10 2012 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri May 11 16:42:14 2012 +0200" }, "message": "block: don\u0027t mark buffers beyond end of disk as mapped\n\nHi,\n\nWe have a bug report open where a squashfs image mounted on ppc64 would\nexhibit errors due to trying to read beyond the end of the disk. It can\neasily be reproduced by doing the following:\n\n[root@ibm-p750e-02-lp3 ~]# ls -l install.img\n-rw-r--r-- 1 root root 142032896 Apr 30 16:46 install.img\n[root@ibm-p750e-02-lp3 ~]# mount -o loop ./install.img /mnt/test\n[root@ibm-p750e-02-lp3 ~]# dd if\u003d/dev/loop0 of\u003d/dev/null\ndd: reading `/dev/loop0\u0027: Input/output error\n277376+0 records in\n277376+0 records out\n142016512 bytes (142 MB) copied, 0.9465 s, 150 MB/s\n\nIn dmesg, you\u0027ll find the following:\n\nsquashfs: version 4.0 (2009/01/31) Phillip Lougher\n[ 43.106012] attempt to access beyond end of device\n[ 43.106029] loop0: rw\u003d0, want\u003d277410, limit\u003d277408\n[ 43.106039] Buffer I/O error on device loop0, logical block 138704\n[ 43.106053] attempt to access beyond end of device\n[ 43.106057] loop0: rw\u003d0, want\u003d277412, limit\u003d277408\n[ 43.106061] Buffer I/O error on device loop0, logical block 138705\n[ 43.106066] attempt to access beyond end of device\n[ 43.106070] loop0: rw\u003d0, want\u003d277414, limit\u003d277408\n[ 43.106073] Buffer I/O error on device loop0, logical block 138706\n[ 43.106078] attempt to access beyond end of device\n[ 43.106081] loop0: rw\u003d0, want\u003d277416, limit\u003d277408\n[ 43.106085] Buffer I/O error on device loop0, logical block 138707\n[ 43.106089] attempt to access beyond end of device\n[ 43.106093] loop0: rw\u003d0, want\u003d277418, limit\u003d277408\n[ 43.106096] Buffer I/O error on device loop0, logical block 138708\n[ 43.106101] attempt to access beyond end of device\n[ 43.106104] loop0: rw\u003d0, want\u003d277420, limit\u003d277408\n[ 43.106108] Buffer I/O error on device loop0, logical block 138709\n[ 43.106112] attempt to access beyond end of device\n[ 43.106116] loop0: rw\u003d0, want\u003d277422, limit\u003d277408\n[ 43.106120] Buffer I/O error on device loop0, logical block 138710\n[ 43.106124] attempt to access beyond end of device\n[ 43.106128] loop0: rw\u003d0, want\u003d277424, limit\u003d277408\n[ 43.106131] Buffer I/O error on device loop0, logical block 138711\n[ 43.106135] attempt to access beyond end of device\n[ 43.106139] loop0: rw\u003d0, want\u003d277426, limit\u003d277408\n[ 43.106143] Buffer I/O error on device loop0, logical block 138712\n[ 43.106147] attempt to access beyond end of device\n[ 43.106151] loop0: rw\u003d0, want\u003d277428, limit\u003d277408\n[ 43.106154] Buffer I/O error on device loop0, logical block 138713\n[ 43.106158] attempt to access beyond end of device\n[ 43.106162] loop0: rw\u003d0, want\u003d277430, limit\u003d277408\n[ 43.106166] attempt to access beyond end of device\n[ 43.106169] loop0: rw\u003d0, want\u003d277432, limit\u003d277408\n...\n[ 43.106307] attempt to access beyond end of device\n[ 43.106311] loop0: rw\u003d0, want\u003d277470, limit\u003d2774\n\nSquashfs manages to read in the end block(s) of the disk during the\nmount operation. Then, when dd reads the block device, it leads to\nblock_read_full_page being called with buffers that are beyond end of\ndisk, but are marked as mapped. Thus, it would end up submitting read\nI/O against them, resulting in the errors mentioned above. I fixed the\nproblem by modifying init_page_buffers to only set the buffer mapped if\nit fell inside of i_size.\n\nCheers,\nJeff\n\nSigned-off-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\nAcked-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n\n--\n\nChanges from v1-\u003ev2: re-used max_block, as suggested by Nick Piggin.\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "7c283324da366a3e6ffaad4352a51a3c71fcae17", "tree": "b4d02c340a17ac5134889c4eec7dcd06d407e3ca", "parents": [ "9e5869f8d70d94850cf86163c57ba8d4daa29924", "17ff3c1fa4c7bd0c38d751715033023ebf32fc96" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 10 15:17:24 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 10 15:17:24 2012 -0700" }, "message": "Merge branch \u0027akpm\u0027 (Andrew\u0027s patch-bomb)\n\nMerge misc fixes from Andrew Morton.\n\n* emailed from Andrew Morton \u003cakpm@linux-foundation.org\u003e: (8 patches)\n MAINTAINERS: add maintainer for LED subsystem\n mm: nobootmem: fix sign extend problem in __free_pages_memory()\n drivers/leds: correct __devexit annotations\n memcg: free spare array to avoid memory leak\n namespaces, pid_ns: fix leakage on fork() failure\n hugetlb: prevent BUG_ON in hugetlb_fault() -\u003e hugetlb_cow()\n mm: fix division by 0 in percpu_pagelist_fraction()\n proc/pid/pagemap: correctly report non-present ptes and holes between vmas\n" }, { "commit": "16fbdce62d9c89b794e303f4a232e4749b77e9ac", "tree": "b0088af87f70ceeef0fbe55bc2200b4f1b9aa816", "parents": [ "bc46f9375a286d05f84a9464efc2b7f1f5614ff4" ], "author": { "name": "Konstantin Khlebnikov", "email": "khlebnikov@openvz.org", "time": "Thu May 10 13:01:43 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 10 15:06:44 2012 -0700" }, "message": "proc/pid/pagemap: correctly report non-present ptes and holes between vmas\n\nReset the current pagemap-entry if the current pte isn\u0027t present, or if\ncurrent vma is over. Otherwise pagemap reports last entry again and\nagain.\n\nNon-present pte reporting was broken in commit 092b50bacd1c (\"pagemap:\nintroduce data structure for pagemap entry\")\n\nReporting for holes was broken in commit 5aaabe831eb5 (\"pagemap: avoid\nsplitting thp when reading /proc/pid/pagemap\")\n\nSigned-off-by: Konstantin Khlebnikov \u003ckhlebnikov@openvz.org\u003e\nReported-by: Pavel Emelyanov \u003cxemul@parallels.com\u003e\nCc: Naoya Horiguchi \u003cn-horiguchi@ah.jp.nec.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "48a5730e5b71201e226ff06e245bf308feba5f10", "tree": "2018cd2924ed13f736032beec587858c5e688ce1", "parents": [ "7ee94d97aafacf5a019b3578e0eae6daa2e2bcd5" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Mon Apr 30 17:36:21 2012 +0300" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Wed May 09 15:16:22 2012 -0500" }, "message": "cifs: fix revalidation test in cifs_llseek()\n\nThis test is always true so it means we revalidate the length every\ntime, which generates more network traffic. When it is SEEK_SET or\nSEEK_CUR, then we don\u0027t need to revalidate.\n\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nReviewed-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "226bb7df3d22bcf4a1c0fe8206c80cc427498eae", "tree": "9e876ce5c9a8699ecacccf9dc3ee4a9c436015c3", "parents": [ "7a84477c4acebf6299b6a8bd6a1d5894eb838ffa" ], "author": { "name": "Josh Cartwright", "email": "joshc@linux.com", "time": "Thu Mar 29 19:34:53 2012 -0400" }, "committer": { "name": "David Woodhouse", "email": "David.Woodhouse@intel.com", "time": "Mon May 07 20:30:14 2012 +0100" }, "message": "jffs2: Fix lock acquisition order bug in gc path\n\nThe locking policy is such that the erase_complete_block spinlock is\nnested within the alloc_sem mutex. This fixes a case in which the\nacquisition order was erroneously reversed. This issue was caught by\nthe following lockdep splat:\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\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ INFO: possible circular locking dependency detected ]\n 3.0.5 #1\n -------------------------------------------------------\n jffs2_gcd_mtd6/299 is trying to acquire lock:\n (\u0026c-\u003ealloc_sem){+.+.+.}, at: [\u003cc01f7714\u003e] jffs2_garbage_collect_pass+0x314/0x890\n\n but task is already holding lock:\n (\u0026(\u0026c-\u003eerase_completion_lock)-\u003erlock){+.+...}, at: [\u003cc01f7708\u003e] jffs2_garbage_collect_pass+0x308/0x890\n\n which lock already depends on the new lock.\n\n the existing dependency chain (in reverse order) is:\n\n -\u003e #1 (\u0026(\u0026c-\u003eerase_completion_lock)-\u003erlock){+.+...}:\n [\u003cc008bec4\u003e] validate_chain+0xe6c/0x10bc\n [\u003cc008c660\u003e] __lock_acquire+0x54c/0xba4\n [\u003cc008d240\u003e] lock_acquire+0xa4/0x114\n [\u003cc046780c\u003e] _raw_spin_lock+0x3c/0x4c\n [\u003cc01f744c\u003e] jffs2_garbage_collect_pass+0x4c/0x890\n [\u003cc01f937c\u003e] jffs2_garbage_collect_thread+0x1b4/0x1cc\n [\u003cc0071a68\u003e] kthread+0x98/0xa0\n [\u003cc000f264\u003e] kernel_thread_exit+0x0/0x8\n\n -\u003e #0 (\u0026c-\u003ealloc_sem){+.+.+.}:\n [\u003cc008ad2c\u003e] print_circular_bug+0x70/0x2c4\n [\u003cc008c08c\u003e] validate_chain+0x1034/0x10bc\n [\u003cc008c660\u003e] __lock_acquire+0x54c/0xba4\n [\u003cc008d240\u003e] lock_acquire+0xa4/0x114\n [\u003cc0466628\u003e] mutex_lock_nested+0x74/0x33c\n [\u003cc01f7714\u003e] jffs2_garbage_collect_pass+0x314/0x890\n [\u003cc01f937c\u003e] jffs2_garbage_collect_thread+0x1b4/0x1cc\n [\u003cc0071a68\u003e] kthread+0x98/0xa0\n [\u003cc000f264\u003e] kernel_thread_exit+0x0/0x8\n\n other info that might help us debug this:\n\n Possible unsafe locking scenario:\n\n CPU0 CPU1\n ---- ----\n lock(\u0026(\u0026c-\u003eerase_completion_lock)-\u003erlock);\n lock(\u0026c-\u003ealloc_sem);\n lock(\u0026(\u0026c-\u003eerase_completion_lock)-\u003erlock);\n lock(\u0026c-\u003ealloc_sem);\n\n *** DEADLOCK ***\n\n 1 lock held by jffs2_gcd_mtd6/299:\n #0: (\u0026(\u0026c-\u003eerase_completion_lock)-\u003erlock){+.+...}, at: [\u003cc01f7708\u003e] jffs2_garbage_collect_pass+0x308/0x890\n\n stack backtrace:\n [\u003cc00155dc\u003e] (unwind_backtrace+0x0/0x100) from [\u003cc0463dc0\u003e] (dump_stack+0x20/0x24)\n [\u003cc0463dc0\u003e] (dump_stack+0x20/0x24) from [\u003cc008ae84\u003e] (print_circular_bug+0x1c8/0x2c4)\n [\u003cc008ae84\u003e] (print_circular_bug+0x1c8/0x2c4) from [\u003cc008c08c\u003e] (validate_chain+0x1034/0x10bc)\n [\u003cc008c08c\u003e] (validate_chain+0x1034/0x10bc) from [\u003cc008c660\u003e] (__lock_acquire+0x54c/0xba4)\n [\u003cc008c660\u003e] (__lock_acquire+0x54c/0xba4) from [\u003cc008d240\u003e] (lock_acquire+0xa4/0x114)\n [\u003cc008d240\u003e] (lock_acquire+0xa4/0x114) from [\u003cc0466628\u003e] (mutex_lock_nested+0x74/0x33c)\n [\u003cc0466628\u003e] (mutex_lock_nested+0x74/0x33c) from [\u003cc01f7714\u003e] (jffs2_garbage_collect_pass+0x314/0x890)\n [\u003cc01f7714\u003e] (jffs2_garbage_collect_pass+0x314/0x890) from [\u003cc01f937c\u003e] (jffs2_garbage_collect_thread+0x1b4/0x1cc)\n [\u003cc01f937c\u003e] (jffs2_garbage_collect_thread+0x1b4/0x1cc) from [\u003cc0071a68\u003e] (kthread+0x98/0xa0)\n [\u003cc0071a68\u003e] (kthread+0x98/0xa0) from [\u003cc000f264\u003e] (kernel_thread_exit+0x0/0x8)\n\nThis was introduce in \u002781cfc9f jffs2: Fix serious write stall due to erase\u0027.\n\nCc: stable@kernel.org [2.6.37+]\nSigned-off-by: Josh Cartwright \u003cjoshc@linux.com\u003e\nSigned-off-by: Artem Bityutskiy \u003cartem.bityutskiy@linux.intel.com\u003e\nSigned-off-by: David Woodhouse \u003cDavid.Woodhouse@intel.com\u003e\n" }, { "commit": "271fd5d7286eb931142402c170943d14640bb922", "tree": "dc60c1623f63a7db588de4c3c7362a9ac7abf56a", "parents": [ "ce7e5d2d19bc371e1b67826bfbc79bbcbaa9772f", "b9fab919b748c7b39c19ff236ed6c5682c266dde" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 06 10:20:07 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 06 10:20:07 2012 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs\n\nPull btrfs fixes from Chris Mason:\n \"The big ones here are a memory leak we introduced in rc1, and a\n scheduling while atomic if the transid on disk doesn\u0027t match the\n transid we expected. This happens for corrupt blocks, or out of date\n disks.\n\n It also fixes up the ioctl definition for our ioctl to resolve logical\n inode numbers. The __u32 was a merging error and doesn\u0027t match what\n we ship in the progs.\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:\n Btrfs: avoid sleeping in verify_parent_transid while atomic\n Btrfs: fix crash in scrub repair code when device is missing\n btrfs: Fix mismatching struct members in ioctl.h\n Btrfs: fix page leak when allocing extent buffers\n Btrfs: Add properly locking around add_root_to_dirty_list\n" }, { "commit": "b9fab919b748c7b39c19ff236ed6c5682c266dde", "tree": "49e5a6f8041a7f0a9be0c1a39cd9088e3faa1df2", "parents": [ "ea9947b4395fa34666086b2fa6f686e94903e047" ], "author": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Sun May 06 07:23:47 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Sun May 06 07:23:47 2012 -0400" }, "message": "Btrfs: avoid sleeping in verify_parent_transid while atomic\n\nverify_parent_transid needs to lock the extent range to make\nsure no IO is underway, and so it can safely clear the\nuptodate bits if our checks fail.\n\nBut, a few callers are using it with spinlocks held. Most\nof the time, the generation numbers are going to match, and\nwe don\u0027t want to switch to a blocking lock just for the error\ncase. This adds an atomic flag to verify_parent_transid,\nand changes it to return EAGAIN if it needs to block to\nproperly verifiy things.\n\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "6f24f892871acc47b40dd594c63606a17c714f77", "tree": "b01432358955a328da347e7f05c096d2ca4366ce", "parents": [ "f756beba940ca21755396851521463d494893566" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri May 04 12:09:39 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 04 17:11:24 2012 -0700" }, "message": "hfsplus: Fix potential buffer overflows\n\nCommit ec81aecb2966 (\"hfs: fix a potential buffer overflow\") fixed a few\npotential buffer overflows in the hfs filesystem. But as Timo Warns\npointed out, these changes also need to be made on the hfsplus\nfilesystem as well.\n\nReported-by: Timo Warns \u003cwarns@pre-sense.de\u003e\nAcked-by: WANG Cong \u003camwang@redhat.com\u003e\nCc: Alexey Khoroshilov \u003ckhoroshilov@ispras.ru\u003e\nCc: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nCc: Sage Weil \u003csage@newdream.net\u003e\nCc: Eugene Teo \u003ceteo@redhat.com\u003e\nCc: Roman Zippel \u003czippel@linux-m68k.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nCc: Dave Anderson \u003canderson@redhat.com\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c6de1687f57893df31fd5266ead3479418230a06", "tree": "70aee3407e518f55377da4ec105bf13f491b47e8", "parents": [ "a03a09b2245b4aa255760796e0c0d05b5d79b2a8", "d8f2799b105a24bb0bbd3380a0d56e6348484058" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 04 15:34:21 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 04 15:34:21 2012 -0700" }, "message": "Merge git://git.samba.org/sfrench/cifs-2.6\n\nPull CIFS fixes from Steve French.\n\n* git://git.samba.org/sfrench/cifs-2.6:\n fs/cifs: fix parsing of dfs referrals\n cifs: make sure we ignore the credentials\u003d and cred\u003d options\n [CIFS] Update cifs version to 1.78\n cifs - check S_AUTOMOUNT in revalidate\n cifs: add missing initialization of server-\u003ereq_lock\n cifs: don\u0027t cap ra_pages at the same level as default_backing_dev_info\n CIFS: Fix indentation in cifs_show_options\n" }, { "commit": "ea9947b4395fa34666086b2fa6f686e94903e047", "tree": "086ab5150fc7f37b2070ab1200bf8b5275c85f06", "parents": [ "d04b1debc92535453df2494d0b019edf0bb91003" ], "author": { "name": "Stefan Behrens", "email": "sbehrens@giantdisaster.de", "time": "Fri May 04 15:16:07 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri May 04 15:16:07 2012 -0400" }, "message": "Btrfs: fix crash in scrub repair code when device is missing\n\nFix that when scrub tries to repair an I/O or checksum error and one of\nthe devices containing the mirror is missing, it crashes in bio_add_page\nbecause the bdev is a NULL pointer for missing devices.\n\nReported-by: Marco L. Crociani \u003cmarco.crociani@gmail.com\u003e\nSigned-off-by: Stefan Behrens \u003csbehrens@giantdisaster.de\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "d04b1debc92535453df2494d0b019edf0bb91003", "tree": "386e5ce40a3d4c211bafc20da9480eb6783f6ab4", "parents": [ "17de39ac17bf99b8bf0d819d13668d5048836efc" ], "author": { "name": "Alexander Block", "email": "ablock84@googlemail.com", "time": "Fri May 04 15:16:06 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri May 04 15:16:06 2012 -0400" }, "message": "btrfs: Fix mismatching struct members in ioctl.h\n\nFix the size members of btrfs_ioctl_ino_path_args and\nbtrfs_ioctl_logical_ino_args. The user space btrfs-progs utilities used\n__u64 and the kernel headers used __u32 before.\n\nSigned-off-by: Alexander Block \u003cablock84@googlemail.com\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "17de39ac17bf99b8bf0d819d13668d5048836efc", "tree": "6afd6d7659ad9d4d46aecc24e359c026bae7c7f7", "parents": [ "e5846fc665d1c3dd32d877febe7402ccd583b8a1" ], "author": { "name": "Josef Bacik", "email": "josef@redhat.com", "time": "Fri May 04 15:16:06 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri May 04 15:16:06 2012 -0400" }, "message": "Btrfs: fix page leak when allocing extent buffers\n\nIf we happen to alloc a extent buffer and then alloc a page and notice that\npage is already attached to an extent buffer, we will only unlock it and\nfree our existing eb. Any pages currently attached to that eb will be\nproperly freed, but we don\u0027t do the page_cache_release() on the page where\nwe noticed the other extent buffer which can cause us to leak pages and I\nhope cause the weird issues we\u0027ve been seeing in this area. Thanks,\n\nSigned-off-by: Josef Bacik \u003cjosef@redhat.com\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "e5846fc665d1c3dd32d877febe7402ccd583b8a1", "tree": "162d4487dd5cb573036507821d92d5c0c559d56b", "parents": [ "dc7fdde39e4962b1a88741f7eba2a6b3be1285d8" ], "author": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Thu May 03 12:08:48 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri May 04 15:14:11 2012 -0400" }, "message": "Btrfs: Add properly locking around add_root_to_dirty_list\n\nadd_root_to_dirty_list happens once at the very beginning of the\ntransaction, but it is still racey.\n\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "d8f2799b105a24bb0bbd3380a0d56e6348484058", "tree": "170d00ed1bb01443fb1b1695f9fcea3adb3d1f10", "parents": [ "a557b97616c49d81e09c8439831d4c4f13ef4050" ], "author": { "name": "Stefan Metzmacher", "email": "metze@samba.org", "time": "Fri May 04 00:19:28 2012 +0200" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Thu May 03 22:47:39 2012 -0500" }, "message": "fs/cifs: fix parsing of dfs referrals\n\nThe problem was that the first referral was parsed more than once\nand so the caller tried the same referrals multiple times.\n\nThe problem was introduced partly by commit\n066ce6899484d9026acd6ba3a8dbbedb33d7ae1b,\nwhere \u0027ref +\u003d le16_to_cpu(ref-\u003eSize);\u0027 got lost,\nbut that was also wrong...\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Stefan Metzmacher \u003cmetze@samba.org\u003e\nTested-by: Björn Jacke \u003cbj@sernet.de\u003e\nReviewed-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "e419b4cc585680940bc42f8ca8a071d6023fb1bb", "tree": "8fce0f12b7b2a0fdca7a937af137910011efa783", "parents": [ "ac001e76546523ec2ef05b2f7001d8fdc588d069" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 03 10:16:43 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 03 14:01:40 2012 -0700" }, "message": "vfs: make word-at-a-time accesses handle a non-existing page\n\nIt turns out that there are more cases than CONFIG_DEBUG_PAGEALLOC that\ncan have holes in the kernel address space: it seems to happen easily\nwith Xen, and it looks like the AMD gart64 code will also punch holes\ndynamically.\n\nActually hitting that case is still very unlikely, so just do the\naccess, and take an exception and fix it up for the very unlikely case\nof it being a page-crosser with no next page.\n\nAnd hey, this abstraction might even help other architectures that have\nother issues with unaligned word accesses than the possible missing next\npage. IOW, this could do the byte order magic too.\n\nPeter Anvin fixed a thinko in the shifting for the exception case.\n\nReported-and-tested-by: Jana Saout \u003cjana@saout.de\u003e\nCc: Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a557b97616c49d81e09c8439831d4c4f13ef4050", "tree": "11824355379fb4272353c4e7e240acd6fabbba02", "parents": [ "f966424e9935900e34cace8116d37aa70cff23d0" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed May 02 14:02:40 2012 -0400" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Thu May 03 13:50:01 2012 -0500" }, "message": "cifs: make sure we ignore the credentials\u003d and cred\u003d options\n\nOlder mount.cifs programs passed this on to the kernel after parsing\nthe file. Make sure the kernel ignores that option.\n\nShould fix:\n\n https://bugzilla.kernel.org/show_bug.cgi?id\u003d43195\n\nCc: Sachin Prabhu \u003csprabhu@redhat.com\u003e\nReported-by: Ronald \u003cronald645@gmail.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "f966424e9935900e34cace8116d37aa70cff23d0", "tree": "3e02c49ef9b23ff7370d808b2226ad772a1acaa8", "parents": [ "936ad9094462578953042d3395b973f1c9e6fa95" ], "author": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Wed May 02 11:58:19 2012 -0500" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Thu May 03 13:50:01 2012 -0500" }, "message": "[CIFS] Update cifs version to 1.78\n\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "936ad9094462578953042d3395b973f1c9e6fa95", "tree": "22d52c9a2489768a9d70d95ef1a99c78185698c3", "parents": [ "58fa015f611b51e1f501b048bc5ac263c78852f0" ], "author": { "name": "Ian Kent", "email": "raven@themaw.net", "time": "Wed May 02 07:19:09 2012 -0400" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Thu May 03 13:49:47 2012 -0500" }, "message": "cifs - check S_AUTOMOUNT in revalidate\n\nWhen revalidating a dentry, if the inode wasn\u0027t known to be a dfs\nentry when the dentry was instantiated, such as when created via\n-\u003ereaddir(), the DCACHE_NEED_AUTOMOUNT flag needs to be set on the\ndentry in -\u003ed_revalidate().\n\nThe false return from cifs_d_revalidate(), due to the inode now\nbeing marked with the S_AUTOMOUNT flag, might not invalidate the\ndentry if there is a concurrent unlazy path walk. This is because\nthe dentry reference count will be at least 2 in this case causing\nd_invalidate() to return EBUSY. So the asumption that the dentry\nwill be discarded then correctly instantiated via -\u003elookup() might\nnot hold.\n\nSigned-off-by: Ian Kent \u003craven@themaw.net\u003e\nReviewed-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nCc: Steve French \u003csmfrench@gmail.com\u003e\nCc: linux-cifs@vger.kernel.org\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "529acf58981440eefeaf1451387e2a0aa4825c12", "tree": "ffcf408493a34ab1e0e1530dca83aa279708560c", "parents": [ "b821861b905a79f71746945237968c3382d99adc", "3617e5031b3acec04efaa36566a8111ac8f07325" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 02 08:17:57 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 02 08:17:57 2012 -0700" }, "message": "Merge tag \u0027nfs-for-3.4-4\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs\n\nPull NFS client bugfixes from Trond Myklebust:\n - Fixes for the NFSv4 security negotiation\n - Use the correct hostname when mounting from a private namespace\n - NFS net namespace bugfixes for the pipefs filesystem\n - NFSv4 GETACL bugfixes\n - IPv6 bugfix for NFSv4 referrals\n\n* tag \u0027nfs-for-3.4-4\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs:\n NFSv4.1: Use the correct hostname in the client identifier string\n SUNRPC: RPC client must use the current utsname hostname string\n NFS: get module in idmap PipeFS notifier callback\n NFS: Remove unused function nfs_lookup_with_sec()\n NFS: Honor the authflavor set in the clone mount data\n NFS: Fix following referral mount points with different security\n NFS: Do secinfo as part of lookup\n NFS: Handle exceptions coming out of nfs4_proc_fs_locations()\n NFS: Fix SECINFO_NO_NAME\n SUNRPC: traverse clients tree on PipeFS event\n SUNRPC: set per-net PipeFS superblock before notification\n SUNRPC: skip clients with program without PipeFS entries\n SUNRPC: skip dead but not buried clients on PipeFS events\n Avoid beyond bounds copy while caching ACL\n Avoid reading past buffer when calling GETACL\n fix page number calculation bug for block layout decode buffer\n NFSv4.1 fix page number calculation bug for filelayout decode buffers\n pnfs-obj: Remove unused variable from objlayout_get_deviceinfo()\n nfs4: fix referrals on mounts that use IPv6 addrs\n" }, { "commit": "58fa015f611b51e1f501b048bc5ac263c78852f0", "tree": "b4a8723d1ddb9e943c5954dd65fe9e3e3c1089e5", "parents": [ "8f71465c19ffefbfd0da3c1f5dc172b4bce05e93" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue May 01 17:41:16 2012 -0400" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Tue May 01 22:29:51 2012 -0500" }, "message": "cifs: add missing initialization of server-\u003ereq_lock\n\nCc: Pavel Shilovsky \u003cpiastryyy@gmail.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "8f71465c19ffefbfd0da3c1f5dc172b4bce05e93", "tree": "c27bb25b91b148e5977ea29132dc98ccea89f725", "parents": [ "156d17905e783d057061b3b56a9b3befec064e47" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue May 01 17:41:49 2012 -0400" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Tue May 01 22:27:54 2012 -0500" }, "message": "cifs: don\u0027t cap ra_pages at the same level as default_backing_dev_info\n\nWhile testing, I\u0027ve found that even when we are able to negotiate a\nmuch larger rsize with the server, on-the-wire reads often end up being\ncapped at 128k because of ra_pages being capped at that level.\n\nLifting this restriction gave almost a twofold increase in sequential\nread performance on my craptactular KVM test rig with a 1M rsize.\n\nI think this is safe since the actual ra_pages that the VM requests\nis run through max_sane_readahead() prior to submitting the I/O. Under\nmemory pressure we should end up with large readahead requests being\nsuppressed anyway.\n\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "156d17905e783d057061b3b56a9b3befec064e47", "tree": "83df60eb473d8b502e6f496cffdc67bc9ad2c55a", "parents": [ "b821861b905a79f71746945237968c3382d99adc" ], "author": { "name": "Sachin Prabhu", "email": "sprabhu@redhat.com", "time": "Wed Apr 25 12:10:14 2012 +0100" }, "committer": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Tue May 01 22:19:43 2012 -0500" }, "message": "CIFS: Fix indentation in cifs_show_options\n\nTrivial patch which fixes a misplaced tab in cifs_show_options().\n\nSigned-off-by: Sachin Prabhu \u003csprabhu@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\n" }, { "commit": "8a7dc4b04b22be285ea5bef7d01a02f91e30d562", "tree": "20a8c483fbbee01c7b972dd4bd45b6753fa2d92f", "parents": [ "fbf717629ff03f110f177d244933276423f99a27" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@xenotime.net", "time": "Mon Apr 30 12:25:31 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 30 12:28:48 2012 -0700" }, "message": "nfsd: fix nfs4recover.c printk format warning\n\nFix printk format warnings -- both items are size_t,\nso use %zu to print them.\n\nfs/nfsd/nfs4recover.c:580:3: warning: format \u0027%lu\u0027 expects type \u0027long unsigned int\u0027, but argument 3 has type \u0027size_t\u0027\nfs/nfsd/nfs4recover.c:580:3: warning: format \u0027%lu\u0027 expects type \u0027long unsigned int\u0027, but argument 4 has type \u0027unsigned int\u0027\n\nSigned-off-by: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nCc: \"J. Bruce Fields\" \u003cbfields@fieldses.org\u003e\nCc: linux-nfs@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3617e5031b3acec04efaa36566a8111ac8f07325", "tree": "20fac2901263ee8602e2c211dd971705c58974f5", "parents": [ "cbbb34498f8b2b26cbdc79532c8a2ee5cd1e756a" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Apr 30 12:04:58 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Apr 30 12:04:58 2012 -0400" }, "message": "NFSv4.1: Use the correct hostname in the client identifier string\n\nWe need to use the hostname of the process that created the nfs_client.\nThat hostname is now stored in the rpc_client-\u003ecl_nodename.\n\nAlso remove the utsname()-\u003edomainname component. There is no reason\nto include the NIS/YP domainname in a client identifier string.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "64f371bc3107e69efce563a3d0f0e6880de0d537", "tree": "7eac8ef3bf7a6cc8f9e147b9bf341b14fc6ae7f3", "parents": [ "9883035ae7edef3ec62ad215611cb8e17d6a1a5d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 29 13:30:08 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 29 13:30:08 2012 -0700" }, "message": "autofs: make the autofsv5 packet file descriptor use a packetized pipe\n\nThe autofs packet size has had a very unfortunate size problem on x86:\nbecause the alignment of \u0027u64\u0027 differs in 32-bit and 64-bit modes, and\nbecause the packet data was not 8-byte aligned, the size of the autofsv5\npacket structure differed between 32-bit and 64-bit modes despite\nlooking otherwise identical (300 vs 304 bytes respectively).\n\nWe first fixed that up by making the 64-bit compat mode know about this\nproblem in commit a32744d4abae (\"autofs: work around unhappy compat\nproblem on x86-64\"), and that made a 32-bit \u0027systemd\u0027 work happily on a\n64-bit kernel because everything then worked the same way as on a 32-bit\nkernel.\n\nBut it turned out that \u0027automount\u0027 had actually known and worked around\nthis problem in user space, so fixing the kernel to do the proper 32-bit\ncompatibility handling actually *broke* 32-bit automount on a 64-bit\nkernel, because it knew that the packet sizes were wrong and expected\nthose incorrect sizes.\n\nAs a result, we ended up reverting that compatibility mode fix, and\nthus breaking systemd again, in commit fcbf94b9dedd.\n\nWith both automount and systemd doing a single read() system call, and\nverifying that they get *exactly* the size they expect but using\ndifferent sizes, it seemed that fixing one of them inevitably seemed to\nbreak the other. At one point, a patch I seriously considered applying\nfrom Michael Tokarev did a \"strcmp()\" to see if it was automount that\nwas doing the operation. Ugly, ugly.\n\nHowever, a prettier solution exists now thanks to the packetized pipe\nmode. By marking the communication pipe as being packetized (by simply\nsetting the O_DIRECT flag), we can always just write the bigger packet\nsize, and if user-space does a smaller read, it will just get that\npartial end result and the extra alignment padding will simply be thrown\naway.\n\nThis makes both automount and systemd happy, since they now get the size\nthey asked for, and the kernel side of autofs simply no longer needs to\ncare - it could pad out the packet arbitrarily.\n\nOf course, if there is some *other* user of autofs (please, please,\nplease tell me it ain\u0027t so - and we haven\u0027t heard of any) that tries to\nread the packets with multiple writes, that other user will now be\nbroken - the whole point of the packetized mode is that one system call\ngets exactly one packet, and you cannot read a packet in pieces.\n\nTested-by: Michael Tokarev \u003cmjt@tls.msk.ru\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nCc: Ian Kent \u003craven@themaw.net\u003e\nCc: Thomas Meyer \u003cthomas@m3y3r.de\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9883035ae7edef3ec62ad215611cb8e17d6a1a5d", "tree": "ab4afff1603e0f1e85e349b8a1fdb8415cc457cf", "parents": [ "de9e24eda331bbefb9195a4d646c786bdcbba7d4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 29 13:12:42 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 29 13:12:42 2012 -0700" }, "message": "pipes: add a \"packetized pipe\" mode for writing\n\nThe actual internal pipe implementation is already really about\nindividual packets (called \"pipe buffers\"), and this simply exposes that\nas a special packetized mode.\n\nWhen we are in the packetized mode (marked by O_DIRECT as suggested by\nAlan Cox), a write() on a pipe will not merge the new data with previous\nwrites, so each write will get a pipe buffer of its own. The pipe\nbuffer is then marked with the PIPE_BUF_FLAG_PACKET flag, which in turn\nwill tell the reader side to break the read at that boundary (and throw\naway any partial packet contents that do not fit in the read buffer).\n\nEnd result: as long as you do writes less than PIPE_BUF in size (so that\nthe pipe doesn\u0027t have to split them up), you can now treat the pipe as a\npacket interface, where each read() system call will read one packet at\na time. You can just use a sufficiently big read buffer (PIPE_BUF is\nsufficient, since bigger than that doesn\u0027t guarantee atomicity anyway),\nand the return value of the read() will naturally give you the size of\nthe packet.\n\nNOTE! We do not support zero-sized packets, and zero-sized reads and\nwrites to a pipe continue to be no-ops. Also note that big packets will\ncurrently be split at write time, but that the size at which that\nhappens is not really specified (except that it\u0027s bigger than PIPE_BUF).\nCurrently that limit is the system page size, but we might want to\nexplicitly support bigger packets some day.\n\nThe main user for this is going to be the autofs packet interface,\nallowing us to stop having to care so deeply about exact packet sizes\n(which have had bugs with 32/64-bit compatibility modes). But user\nspace can create packetized pipes with \"pipe2(fd, O_DIRECT)\", which will\nfail with an EINVAL on kernels that do not support this interface.\n\nTested-by: Michael Tokarev \u003cmjt@tls.msk.ru\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nCc: Ian Kent \u003craven@themaw.net\u003e\nCc: Thomas Meyer \u003cthomas@m3y3r.de\u003e\nCc: stable@kernel.org # needed for systemd/autofs interaction fix\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "71dfc5fa5160bb73752f0731539404569a77faca", "tree": "d35919ad0d645b91404ca0d1717e5ffa00e89ae5", "parents": [ "e245d4250d0326cfcf7c816a2081b6ab2ea810be" ], "author": { "name": "Stanislav Kinsbursky", "email": "skinsbursky@parallels.com", "time": "Sat Apr 28 19:32:21 2012 +0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Apr 28 13:22:19 2012 -0400" }, "message": "NFS: get module in idmap PipeFS notifier callback\n\nThis is bug fix.\nNotifier callback is called from SUNRPC module. So before dereferencing NFS\nmodule we have to make sure, that it\u0027s alive.\n\nSigned-off-by: Stanislav Kinsbursky \u003cskinsbursky@parallels.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "f7b006931751f029620ad2f8310ac7a1484fbdb4", "tree": "71120f4c4c51752902317fbf853e3b0316c2adb0", "parents": [ "b990f9b3cb068578b8aefd3a34f8c8555661ef95", "dc7fdde39e4962b1a88741f7eba2a6b3be1285d8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 28 09:30:07 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 28 09:30:07 2012 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs\n\nPull btrfs fixes from Chris Mason:\n \"This has our collection of bug fixes. I missed the last rc because I\n thought our patches were making NFS crash during my xfs test runs.\n Turns out it was an NFS client bug fixed by someone else while I tried\n to bisect it.\n\n All of these fixes are small, but some are fairly high impact. The\n biggest are fixes for our mount -o remount handling, a deadlock due to\n GFP_KERNEL allocations in readdir, and a RAID10 error handling bug.\n\n This was tested against both 3.3 and Linus\u0027 master as of this morning.\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (26 commits)\n Btrfs: reduce lock contention during extent insertion\n Btrfs: avoid deadlocks from GFP_KERNEL allocations during btrfs_real_readdir\n Btrfs: Fix space checking during fs resize\n Btrfs: fix block_rsv and space_info lock ordering\n Btrfs: Prevent root_list corruption\n Btrfs: fix repair code for RAID10\n Btrfs: do not start delalloc inodes during sync\n Btrfs: fix that check_int_data mount option was ignored\n Btrfs: don\u0027t count CRC or header errors twice while scrubbing\n Btrfs: fix btrfs_ioctl_dev_info() crash on missing device\n btrfs: don\u0027t return EINTR\n Btrfs: double unlock bug in error handling\n Btrfs: always store the mirror we read the eb from\n fs/btrfs/volumes.c: add missing free_fs_devices\n btrfs: fix early abort in \u0027remount\u0027\n Btrfs: fix max chunk size check in chunk allocator\n Btrfs: add missing read locks in backref.c\n Btrfs: don\u0027t call free_extent_buffer twice in iterate_irefs\n Btrfs: Make free_ipath() deal gracefully with NULL pointers\n Btrfs: avoid possible use-after-free in clear_extent_bit()\n ...\n" }, { "commit": "fcbf94b9dedd2ce08e798a99aafc94fec8668161", "tree": "bc81982bbcf96538a09103b2c722ac7d2c99bdef", "parents": [ "c629eaf8392b676b4f83c3dc344e66402bfeec92" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 28 08:29:56 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 28 08:29:56 2012 -0700" }, "message": "Revert \"autofs: work around unhappy compat problem on x86-64\"\n\nThis reverts commit a32744d4abae24572eff7269bc17895c41bd0085.\n\nWhile that commit was technically the right thing to do, and made the\nx86-64 compat mode work identically to native 32-bit mode (and thus\nfixing the problem with a 32-bit systemd install on a 64-bit kernel), it\nturns out that the automount binaries had workarounds for this compat\nproblem.\n\nNow, the workarounds are disgusting: doing an \"uname()\" to find out the\narchitecture of the kernel, and then comparing it for the 64-bit cases\nand fixing up the size of the read() in automount for those. And they\nwere confused: it\u0027s not actually a generic 64-bit issue at all, it\u0027s\nvery much tied to just x86-64, which has different alignment for an\n\u0027u64\u0027 in 64-bit mode than in 32-bit mode.\n\nBut the end result is that fixing the compat layer actually breaks the\ncase of a 32-bit automount on a x86-64 kernel.\n\nThere are various approaches to fix this (including just doing a\n\"strcmp()\" on current-\u003ecomm and comparing it to \"automount\"), but I\nthink that I will do the one that teaches pipes about a special \"packet\nmode\", which will allow user space to not have to care too deeply about\nthe padding at the end of the autofs packet.\n\nThat change will make the compat workaround unnecessary, so let\u0027s revert\nit first, and get automount working again in compat mode. The\npacketized pipes will then fix autofs for systemd.\n\nReported-and-requested-by: Michael Tokarev \u003cmjt@tls.msk.ru\u003e\nCc: Ian Kent \u003craven@themaw.net\u003e\nCc: stable@kernel.org # for 3.3\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c629eaf8392b676b4f83c3dc344e66402bfeec92", "tree": "5edd51bb94b7b83ddf3c3c2793c0a6a1ac671e9c", "parents": [ "4bbbf13fd55a9817452533934dee489683e703b8", "28f8881023c9713c303c0feda270929f9384c019" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 27 20:56:54 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 27 20:56:54 2012 -0700" }, "message": "Merge git://git.samba.org/sfrench/cifs-2.6\n\nPull CIFS fixes from Steve French.\n\n* git://git.samba.org/sfrench/cifs-2.6:\n Use correct conversion specifiers in cifs_show_options\n CIFS: Show backupuid/gid in /proc/mounts\n cifs: fix offset handling in cifs_iovec_write\n" }, { "commit": "dc7fdde39e4962b1a88741f7eba2a6b3be1285d8", "tree": "97cd8b1f9d8c0682c64303d45de1fb55925abcdf", "parents": [ "fede766f28dd766d4e8feb321fdb19edb21ef6fb" ], "author": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 14:31:29 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 14:51:05 2012 -0400" }, "message": "Btrfs: reduce lock contention during extent insertion\n\nWe\u0027re spending huge amounts of time on lock contention during\nend_io processing because we unconditionally assume we are overwriting\nan existing extent in the file for each IO.\n\nThis checks to see if we are outside i_size, and if so, it uses a\nless expensive readonly search of the btree to look for existing\nextents.\n\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "fede766f28dd766d4e8feb321fdb19edb21ef6fb", "tree": "86e59f6bf94ad77c2fa1b232faa8f8b63b8675cb", "parents": [ "7654b72417e10e294563496e25211200f9b8b6d3" ], "author": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 14:23:22 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 14:23:22 2012 -0400" }, "message": "Btrfs: avoid deadlocks from GFP_KERNEL allocations during btrfs_real_readdir\n\nBtrfs has an optimization where it will preallocate dentries during\nreaddir to fill in enough information to open the inode without an extra\nlookup.\n\nBut, we\u0027re calling d_alloc, which is doing GFP_KERNEL allocations, and\nthat leads to deadlocks because our readdir code has tree locks held.\n\nFor now, disable this optimization. We\u0027ll fix the gfp mask in the next\nmerge window.\n\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "e245d4250d0326cfcf7c816a2081b6ab2ea810be", "tree": "1424c497bdcd1305137b18201a3086a365595a72", "parents": [ "7e6eb683d260d9325f0d1bd911518d5ed3cb4f0c" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Fri Apr 27 13:27:43 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:10:03 2012 -0400" }, "message": "NFS: Remove unused function nfs_lookup_with_sec()\n\nThis fixes a compiler warning.\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "7e6eb683d260d9325f0d1bd911518d5ed3cb4f0c", "tree": "51b22645b2e55cff65acf9feb0d9f43f44ed70d3", "parents": [ "f05d147f7e3cf0d86b3a4bd5603029a7cb109633" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Fri Apr 27 13:27:42 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:10:03 2012 -0400" }, "message": "NFS: Honor the authflavor set in the clone mount data\n\nThe authflavor is set in an nfs_clone_mount structure and passed to the\nxdev_mount() functions where it was promptly ignored. Instead, use it\nto initialize an rpc_clnt for the cloned server.\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "f05d147f7e3cf0d86b3a4bd5603029a7cb109633", "tree": "e351ef7dd3d541626f5cccef86f50906bfb3a649", "parents": [ "72de53ec4bca39c26709122a8f78bfefe7b6bca4" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Fri Apr 27 13:27:41 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:10:02 2012 -0400" }, "message": "NFS: Fix following referral mount points with different security\n\nI create a new proc_lookup_mountpoint() to use when submounting an NFS\nv4 share. This function returns an rpc_clnt to use for performing an\nfs_locations() call on a referral\u0027s mountpoint.\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "72de53ec4bca39c26709122a8f78bfefe7b6bca4", "tree": "dac32c44281b153df7104037c1e97c11deffae69", "parents": [ "db0a9593d52f935c80085d8993bdcead1ad30b0c" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Fri Apr 27 13:27:40 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:10:02 2012 -0400" }, "message": "NFS: Do secinfo as part of lookup\n\nWhenever lookup sees wrongsec do a secinfo and retry the lookup to find\nattributes of the file or directory, such as \"is this a referral\nmountpoint?\". This also allows me to remove handling -NFS4ERR_WRONSEC\nas part of getattr xdr decoding.\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "db0a9593d52f935c80085d8993bdcead1ad30b0c", "tree": "e0ca32d8d4366042479cb2a87d6fe87998a9849f", "parents": [ "31e4dda4747713de13889f7c79c7aec341fea61b" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Fri Apr 27 13:27:39 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:10:01 2012 -0400" }, "message": "NFS: Handle exceptions coming out of nfs4_proc_fs_locations()\n\nWe don\u0027t want to return -NFS4ERR_WRONGSEC to the VFS because it could\ncause the kernel to oops.\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "31e4dda4747713de13889f7c79c7aec341fea61b", "tree": "9ec536cc095e2e512d1863301dcf37ca92dde8b5", "parents": [ "ea8cfa06795bb30d2ea61f503ef129284492c06a" ], "author": { "name": "Bryan Schumaker", "email": "bjschuma@netapp.com", "time": "Fri Apr 27 13:27:38 2012 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:10:01 2012 -0400" }, "message": "NFS: Fix SECINFO_NO_NAME\n\nI was using the same decoder function for SECINFO and SECINFO_NO_NAME,\nso it was returning an error when it tried to decode an OP_SECINFO_NO_NAME\nheader as OP_SECINFO.\n\nSigned-off-by: Bryan Schumaker \u003cbjschuma@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "5794d21ef4639f0e33440927bb903f9598c21e92", "tree": "d4bdbde63624df6e970751f46fd6fe0dacb8718c", "parents": [ "5a00689930ab975fdd1b37b034475017e460cf2a" ], "author": { "name": "Sachin Prabhu", "email": "sprabhu@redhat.com", "time": "Tue Apr 17 14:36:40 2012 +0100" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 27 14:09:53 2012 -0400" }, "message": "Avoid beyond bounds copy while caching ACL\n\nWhen attempting to cache ACLs returned from the server, if the bitmap\nsize + the ACL size is greater than a PAGE_SIZE but the ACL size itself\nis smaller than a PAGE_SIZE, we can read past the buffer page boundary.\n\nSigned-off-by: Sachin Prabhu \u003csprabhu@redhat.com\u003e\nReported-by: Jian Li \u003cjiali@redhat.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "7654b72417e10e294563496e25211200f9b8b6d3", "tree": "5ca24ea9bb735133c0e09a789d3eb50d9a8b69b4", "parents": [ "1f699d38b6556c393ac80f1c23c2053502a51631" ], "author": { "name": "Daniel J Blueman", "email": "daniel@quora.org", "time": "Fri Apr 27 12:41:46 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 13:55:14 2012 -0400" }, "message": "Btrfs: Fix space checking during fs resize\n\nFix out-of-space checking, addressing a warning and potential resource\nleak when resizing the filesystem down while allocating blocks.\n\nSigned-off-by: Daniel J Blueman \u003cdaniel@quora.org\u003e\nReviewed-by: Josef Bacik \u003cjosef@redhat.com\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "1f699d38b6556c393ac80f1c23c2053502a51631", "tree": "9ba8e3c95cc044461eab17dfba50515535ca2a25", "parents": [ "1daf3540fa77faea2f91d96bcaf07ce48ee827be" ], "author": { "name": "Stefan Behrens", "email": "sbehrens@giantdisaster.de", "time": "Fri Apr 27 12:41:46 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 13:55:14 2012 -0400" }, "message": "Btrfs: fix block_rsv and space_info lock ordering\n\nmay_commit_transaction() calls\n spin_lock(\u0026space_info-\u003elock);\n spin_lock(\u0026delayed_rsv-\u003elock);\nand update_global_block_rsv() calls\n spin_lock(\u0026block_rsv-\u003elock);\n spin_lock(\u0026sinfo-\u003elock);\n\nLockdep complains about this at run time.\nEverywhere except in update_global_block_rsv(), the space_info lock is\nthe outer lock, therefore the locking order in update_global_block_rsv()\nis changed.\n\nSigned-off-by: Stefan Behrens \u003csbehrens@giantdisaster.de\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "1daf3540fa77faea2f91d96bcaf07ce48ee827be", "tree": "0cdec92cca58085352b7ba8f4c3245a91dd0b5a9", "parents": [ "3e74317ad773ba9df36db1fa32848cba41ac4d1a" ], "author": { "name": "Daniel J Blueman", "email": "daniel@quora.org", "time": "Fri Apr 27 12:41:46 2012 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Fri Apr 27 13:55:13 2012 -0400" }, "message": "Btrfs: Prevent root_list corruption\n\nI was seeing root_list corruption on unmount during fs resize in 3.4-rc4; add\ncorrect locking to address this.\n\nSigned-off-by: Daniel J Blueman \u003cdaniel@quora.org\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" } ], "next": "3e74317ad773ba9df36db1fa32848cba41ac4d1a" }