)]}' { "log": [ { "commit": "a0ccda839a08c3484f450571bc5063602dacbc4d", "tree": "eb91a1fe01262d1e7b3247f1bbe2302b6fa459ff", "parents": [ "0eb2f608f122f43160d74d1ba3db88f0d5f787e7" ], "author": { "name": "Tony Espy", "email": "espy@canonical.com", "time": "Wed Feb 27 17:34:33 2013 -0500" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Sat Apr 27 01:43:07 2013 -0500" }, "message": "Expose virtual PID from child namespace in /proc/PID/status.\n\nThe virtual PID is used by Ubuntu Touch application manager code ( which\nruns in the Android namespace ), to communicate with code that runs inside the\nUbuntu container.\n\nThis patch is not from upstream. It was proposed on the lxc-users\nmailing list:\n\nhttp://www.mail-archive.com/lxc-users@lists.sourceforge.net/msg01840.html\n\nChange-Id: If63652002c227827fb524e03a1d0210fc2b7d4b1\nOriginal-Author: Greg Kurz \u003cgkurz@fr.ibm.com\u003e\nSigned-off-by: Greg Kurz \u003cgkurz@fr.ibm.com\u003e\nSigned-off-by: Tony Espy \u003cespy@canonical.com\u003e\nSigned-off-by: Ricardo Salveti de Araujo \u003cricardo.salveti@canonical.com\u003e\n" }, { "commit": "ac17b61d6c208da6f9ade18a97375030bd158a17", "tree": "e3a0b2d0d82c7f41d3bfbd6808e2495dc5b17931", "parents": [ "b71ead3349787b8b732127b791720858059969a1" ], "author": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Wed Sep 05 18:37:48 2012 -0500" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Wed Sep 05 18:37:48 2012 -0500" }, "message": "Revert \"switch do_fsync() to fget_light()\"\n\nThis reverts commit d2efa9bda0ad5c7fc8bf39780b73fee294892396.\n" }, { "commit": "9f1253577230eb23b75b149d35c27c9e86b0161e", "tree": "6e976b2bac0335a93d149bdc6ee0f8aa7a556daa", "parents": [ "2b1591a8f0fd8a85253470d7b5441fd00f23d4a9" ], "author": { "name": "Jesper Juhl", "email": "jj@chaosbits.net", "time": "Mon Jan 17 00:09:38 2011 +0100" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Tue Aug 28 21:48:23 2012 -0500" }, "message": "Kill off warning: ‘inline’ is not at beginning of declaration\n\nFix a bunch of\n\twarning: ‘inline’ is not at beginning of declaration\nmessages when building a \u0027make allyesconfig\u0027 kernel with -Wextra.\n\nThese warnings are trivial to kill, yet rather annoying when building with\n-Wextra.\nThe more we can cut down on pointless crap like this the better (IMHO).\n\nA previous patch to do this for a \u0027allnoconfig\u0027 build has already been\nmerged. This just takes the cleanup a little further.\n\nSigned-off-by: Jesper Juhl \u003cjj@chaosbits.net\u003e\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "d2efa9bda0ad5c7fc8bf39780b73fee294892396", "tree": "ba57605c8e6de333794a40982ab92ef472ee0883", "parents": [ "147e1d478e1cd0c7a692711d8460a9257b13ecc6" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Apr 19 23:52:50 2012 -0400" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Mon Aug 27 22:36:53 2012 -0500" }, "message": "switch do_fsync() to fget_light()\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "cd85296a84785342c9bd6a7493f9db2746a47548", "tree": "0ead62f47125f601ebcc106256c32f8391286bce", "parents": [ "73d02e1bb50bf3e59d7be12920de848631a8dbab" ], "author": { "name": "Al Viro", "email": "viro@ZenIV.linux.org.uk", "time": "Sun May 29 13:46:08 2011 +0100" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Mon May 14 18:22:14 2012 -0500" }, "message": "cifs/ubifs: Fix shrinker API change fallout\n\nCommit 1495f230fa77 (\"vmscan: change shrinker API by passing\nshrink_control struct\") changed the API of -\u003eshrink(), but missed ubifs\nand cifs instances.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n\nConflicts:\n\n\tfs/cifs/cifsacl.c\ndrewis: checkout --ours (referenced fuction does not exist in current cifs)\n" }, { "commit": "73d02e1bb50bf3e59d7be12920de848631a8dbab", "tree": "c44e677be5f2cc1c3850a24c474baad707099a99", "parents": [ "5a9905ef6165547e982e3c830dc469f9851d425b" ], "author": { "name": "Ying Han", "email": "yinghan@google.com", "time": "Tue May 24 17:12:27 2011 -0700" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Mon May 14 18:22:13 2012 -0500" }, "message": "vmscan: change shrinker API by passing shrink_control struct\n\nChange each shrinker\u0027s API by consolidating the existing parameters into\nshrink_control struct. This will simplify any further features added w/o\ntouching each file of shrinker.\n\n[akpm@linux-foundation.org: fix build]\n[akpm@linux-foundation.org: fix warning]\n[kosaki.motohiro@jp.fujitsu.com: fix up new shrinker API]\n[akpm@linux-foundation.org: fix xfs warning]\n[akpm@linux-foundation.org: update gfs2]\nSigned-off-by: Ying Han \u003cyinghan@google.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nAcked-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Dave Hansen \u003cdave@linux.vnet.ibm.com\u003e\nCc: Steven Whitehouse \u003cswhiteho@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n\nConflicts:\n\n\tdrivers/staging/zcache/zcache.c\n" }, { "commit": "5a9905ef6165547e982e3c830dc469f9851d425b", "tree": "fa9483d136a77d69125f911e84ffd56ae88c94e9", "parents": [ "df04067ea413337ad3f445718d3fc2b9f04fb47e" ], "author": { "name": "Ying Han", "email": "yinghan@google.com", "time": "Tue May 24 17:12:26 2011 -0700" }, "committer": { "name": "Andrew Sutherland", "email": "dr3wsuth3rland@gmail.com", "time": "Mon May 14 18:22:12 2012 -0500" }, "message": "vmscan: change shrink_slab() interfaces by passing shrink_control\n\nConsolidate the existing parameters to shrink_slab() into a new\nshrink_control struct. This is needed later to pass the same struct to\nshrinkers.\n\nSigned-off-by: Ying Han \u003cyinghan@google.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nAcked-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Dave Hansen \u003cdave@linux.vnet.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e422ffb446f5308b187191dd3fe1dfdacc738536", "tree": "f28e08d24ec3fdf33bd027323055a376cc0657c6", "parents": [ "9f4ef2a6ba3dd7c2b0ca2da7016d85b4366af18c" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Sat Dec 18 12:46:40 2010 +1100" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Wed Feb 22 00:27:47 2012 -0500" }, "message": "fs: fsync optimisations\n\nOptimise fsync by adding a datasync parameter to sync_inode_metadata to\nDTRT with writing back the inode (-\u003ewrite_inode in theory should have a\ndatasync parameter too perhaps, but that\u0027s for another time).\n\nAlso, implement the metadata sync optimally rather than reusing the\nnormal data writeback path. This means less useless moving the inode around the\nwriteback lists, and less dropping and retaking of inode_lock, and avoiding\nthe data writeback call with nr_pages \u003d\u003d 0.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "9f4ef2a6ba3dd7c2b0ca2da7016d85b4366af18c", "tree": "257a368fbc580a7b0fe31c1bbedce65ea29db221", "parents": [ "ebe5e8d8fc682e77b48815353e82b23fea1059f4" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Sat Dec 18 12:46:39 2010 +1100" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Wed Feb 22 00:27:36 2012 -0500" }, "message": "fs: ext2 inode sync fix\n\nThere is a big fuckup with inode metadata writeback (I suspect in a lot\nof filesystems, but I\u0027ve only dared to look at a couple as yet).\n\next2 relies on -\u003ewrite_inode being called from sync_inode_metadata in\nfsync in order to sync the inode. However I_DIRTY_SYNC gets cleared\nafter a call to this guy, and it doesn\u0027t actually write back and wait\non the inode block unless it is called for sync. This means that write_inode\nfrom background writeback can kill the inode dirty bits without the data\ngetting to disk. Fsync will subsequently miss it.\n\nThe fix is for -\u003ewrite_inode to dirty the buffers/cache, and then -\u003efsync to\nwrite back the dirty data. In the full filesystem sync case, buffercache\nwriteback in the generic code will write back the dirty data. Other\nfilesystems could use -\u003esync_fs for this.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "ebe5e8d8fc682e77b48815353e82b23fea1059f4", "tree": "b049b313cd990cf80fb0c36390928885eee038f2", "parents": [ "2683189064bf4569e10ace4a7a7774bfa956dbcf" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Sat Dec 18 12:46:38 2010 +1100" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Wed Feb 22 00:27:27 2012 -0500" }, "message": "fs: preserve inode dirty bits on failed metadata writeback\n\nOtherwise we think the inode is clean even if syncing failed.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "2683189064bf4569e10ace4a7a7774bfa956dbcf", "tree": "0419e5272683636037cf7fc09d0c308034966c0b", "parents": [ "740cd5cd68e136f03b305adf7a65c6e1ac156147" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Sat Dec 18 12:46:37 2010 +1100" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Wed Feb 22 00:27:07 2012 -0500" }, "message": "fs: introduce inode writeback helpers\n\nInode dirty state cannot be securely tested without participating properly\nin the inode writeback protocol. Some filesystems need to check this state,\nso break out the code into helpers and make them available.\n\nThis could also be used to reduce strange interactions between background\nwriteback and fsync. Currently if we fsync a single page in a file, the\nentire file gets requeued to the back of the background IO list, even if\nit is due for writeout and has a large number of pages. That\u0027s left for\na later time.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "740cd5cd68e136f03b305adf7a65c6e1ac156147", "tree": "759b7e75e76ec07bc287793a6d3c3d5206e0193f", "parents": [ "2e5ce40906acab7449c404bd0af0456aab21cf62" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Sat Dec 18 12:46:36 2010 +1100" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Wed Feb 22 00:26:58 2012 -0500" }, "message": "fs: simple fsync race fix\n\nIt is incorrect to test inode dirty bits without participating in the inode\nwriteback protocol. Inode writeback sets I_SYNC and clears I_DIRTY_?, then\nwrites out the particular bits, then clears I_SYNC when it is done. BTW. it\nmay not completely write all pages out, so I_DIRTY_PAGES would get set\nagain.\n\nThis is a standard pattern used throughout the kernel\u0027s writeback caches\n(I_SYNC ~\u003d I_WRITEBACK, if that makes it clearer).\n\nAnd so it is not possible to determine an inode\u0027s dirty status just by\nchecking I_DIRTY bits. Especially not for the purpose of data integrity\nsyncs.\n\nMissing the check for these bits means that fsync can complete while\nwriteback to the inode is underway. Inode writeback functions get this\nright, so call into them rather than try to shortcut things by testing\ndirty state improperly.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "2e5ce40906acab7449c404bd0af0456aab21cf62", "tree": "ba2d398cc7b6195a8f7fe4479db14ee0c52a0864", "parents": [ "d470147c4d1c67067c56567162c9636d0145264d" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Sat Dec 18 12:46:35 2010 +1100" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Wed Feb 22 00:26:48 2012 -0500" }, "message": "fs: mark_inode_dirty barrier fix\n\nFilesystems appear to be using -\u003edirty_inode, expecting that the dirtying\noperating is done and visible to all CPUs (eg. setting private inode dirty\nbits, without any barriers themselves). So release the dirty \"critical\nsection\" with a barrier before calling -\u003edirty_inode.\n\nCost is not significantly changed, because we\u0027re just moving the barrier.\nThose filesystems that do use -\u003edirty_inode should have to care slightly\nless about barriers, which is a good thing.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "91d12c145d84aae336ee874147cf378a859fae20", "tree": "990e81af9974f7208bb4703f10e1ea65a3f32e2a", "parents": [ "66d3b0219ffdf18f6ca7ed0c874588418040f9d0" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Apr 18 17:29:14 2011 -0400" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Sun Feb 19 15:16:09 2012 -0500" }, "message": "ext4: check for ext[23] file system features when mounting as ext[23]\n\nProvide better emulation for ext[23] mode by enforcing that the file\nsystem does not have any unsupported file system features as defined\nby ext[23] when emulating the ext[23] file system driver when\nCONFIG_EXT4_USE_FOR_EXT23 is defined.\n\nThis causes the file system type information in /proc/mounts to be\ncorrect for the automatically mounted root file system. This also\nmeans that \"mount -t ext2 /dev/sda /mnt\" will fail if /dev/sda\ncontains an ext3 or ext4 file system, just as one would expect if the\noriginal ext2 file system driver were in use.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "e18263f084152ed9c45e7d38a773876bcaeceb6a", "tree": "548d0ad2fd84a396f813a538f601ddeda3e616f5", "parents": [ "07f15da3d3d145982c883cf635527fa1d2175f4a", "4b7a6d2528bfb625cc359d89ac16439b0ec744ea" ], "author": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Sun Jun 05 00:44:09 2011 -0400" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Sun Jun 05 00:44:09 2011 -0400" }, "message": "merge 2.6.38.8\n" }, { "commit": "ad106f8c9b2493de31e4b560b9aac000aa81ba10", "tree": "cd64f8c718d8832887d339bc5573e375f39caf51", "parents": [ "abe87a0b0ba5a0cddfab30bbafc6e2b121537b15" ], "author": { "name": "Lukas Czerner", "email": "lczerner@redhat.com", "time": "Fri May 20 13:49:04 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:49 2011 +0900" }, "message": "ext4: Use schedule_timeout_interruptible() for waiting in lazyinit thread\n\ncommit 4ed5c033c11b33149d993734a6a8de1016e8f03f upstream.\n\nIn order to make lazyinit eat approx. 10% of io bandwidth at max, we\nare sleeping between zeroing each single inode table. For that purpose\nwe are using timer which wakes up thread when it expires. It is set\nvia add_timer() and this may cause troubles in the case that thread\nhas been woken up earlier and in next iteration we call add_timer() on\nstill running timer hence hitting BUG_ON in add_timer(). We could fix\nthat by using mod_timer() instead however we can use\nschedule_timeout_interruptible() for waiting and hence simplifying\nthings a lot.\n\nThis commit exchange the old \"waiting mechanism\" with simple\nschedule_timeout_interruptible(), setting the time to sleep. Hence we\ndo not longer need li_wait_daemon waiting queue and others, so get rid\nof it.\n\nAddresses-Red-Hat-Bugzilla: #699708\n\nSigned-off-by: Lukas Czerner \u003clczerner@redhat.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0ab692407602f96ad71d62f9c551fdba4529c803", "tree": "e1f79869f67905c1ff9affe6e2e5abed8ed1ccbb", "parents": [ "6844274a3ca81e5643b53c9844796f3dc11b5518" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu May 26 14:26:35 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:48 2011 +0900" }, "message": "NFSv4.1: Fix the handling of NFS4ERR_SEQ_MISORDERED errors\n\ncommit 444f72fe7e7b5f4db34cee933fa3546ebb8e9122 upstream.\n\nCurrently, the call to nfs4_schedule_session_recovery() will actually just\nresult in a test of the lease when what we really want is to force a\nsession reset.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "6844274a3ca81e5643b53c9844796f3dc11b5518", "tree": "cfadfb6f2919cc670e80d0b254c84d678a5df34e", "parents": [ "a0a238f7cb684c0ea76a5817d1c92e2c7b0b7bd2" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu May 26 14:26:35 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:47 2011 +0900" }, "message": "NFSv4: Handle expired stateids when the lease is still valid\n\ncommit 0ced63d1a245ac11241a5d37932e6d04d9c8040d upstream.\n\nCurrently, if the server returns NFS4ERR_EXPIRED in reply to a READ or\nWRITE, but the RENEW test determines that the lease is still active, we\nfail to recover and end up looping forever in a READ/WRITE + RENEW death\nspiral.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "c02eace342342b3156426e7a5ffe1ddc93552c51", "tree": "cc36596dfb623fd4e7e47594c8a70f0650f592dd", "parents": [ "b93e342044640789e34a48893c120875afbcb07d" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Fri May 06 17:08:56 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:38 2011 +0900" }, "message": "UBIFS: fix a rare memory leak in ro to rw remounting path\n\ncommit eaeee242c531cd4b0a4a46e8b5dd7ef504380c42 upstream.\n\nWhen re-mounting from R/O mode to R/W mode and the LEB count in the superblock\nis not up-to date, because for the underlying UBI volume became larger, we\nre-write the superblock. We allocate RAM for these purposes, but never free it.\nSo this is a memory leak, although very rare one.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "b93e342044640789e34a48893c120875afbcb07d", "tree": "fb33518f767f77aa8d20df85512216375531bbb0", "parents": [ "d0f7955df4d1fac108e5edfd6a9ebe9135b1597f" ], "author": { "name": "Tyler Hicks", "email": "tyhicks@linux.vnet.ibm.com", "time": "Tue May 17 00:50:33 2011 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:31 2011 +0900" }, "message": "eCryptfs: Allow 2 scatterlist entries for encrypted filenames\n\ncommit 8d08dab786ad5cc2aca2bf870de370144b78c85a upstream.\n\nThe buffers allocated while encrypting and decrypting long filenames can\nsometimes straddle two pages. In this situation, virt_to_scatterlist()\nwill return -ENOMEM, causing the operation to fail and the user will get\nscary error messages in their logs:\n\nkernel: ecryptfs_write_tag_70_packet: Internal error whilst attempting\nto convert filename memory to scatterlist; expected rc \u003d 1; got rc \u003d\n[-12]. block_aligned_filename_size \u003d [272]\nkernel: ecryptfs_encrypt_filename: Error attempting to generate tag 70\npacket; rc \u003d [-12]\nkernel: ecryptfs_encrypt_and_encode_filename: Error attempting to\nencrypt filename; rc \u003d [-12]\nkernel: ecryptfs_lookup: Error attempting to encrypt and encode\nfilename; rc \u003d [-12]\n\nThe solution is to allow up to 2 scatterlist entries to be used.\n\nSigned-off-by: Tyler Hicks \u003ctyhicks@linux.vnet.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "2f026bcd6c57ca6cae4032fd0748abf6fd24c19c", "tree": "d8c4dee668f39dfd81f2853cb97667182061f8b1", "parents": [ "ea1e05a4044e3ef488ee43098f89b4a34c034837" ], "author": { "name": "Timo Warns", "email": "Warns@pre-sense.de", "time": "Thu May 19 09:24:17 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:13 2011 +0900" }, "message": "Fix for buffer overflow in ldm_frag_add not sufficient\n\ncommit cae13fe4cc3f24820ffb990c09110626837e85d4 upstream.\n\nAs Ben Hutchings discovered [1], the patch for CVE-2011-1017 (buffer\noverflow in ldm_frag_add) is not sufficient. The original patch in\ncommit c340b1d64000 (\"fs/partitions/ldm.c: fix oops caused by corrupted\npartition table\") does not consider that, for subsequent fragments,\npreviously allocated memory is used.\n\n[1] http://lkml.org/lkml/2011/5/6/407\n\nReported-by: Ben Hutchings \u003cben@decadent.org.uk\u003e\nSigned-off-by: Timo Warns \u003cwarns@pre-sense.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "f1ed16123ae7b2b87b20d9bf60d179f635dba3d5", "tree": "745e4ff52e323ee2beb5a0f9d6546372f9865dc1", "parents": [ "8574041ad2ecac425ea510f2db86eeed27a6decb" ], "author": { "name": "Steve French", "email": "sfrench@us.ibm.com", "time": "Thu May 26 18:38:54 2011 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:34:12 2011 +0900" }, "message": "When mandatory encryption on share, fail mount\n\ncommit 6848b7334b24b47aa3d0e70342ff839ffa95d5fa upstream.\n\n When mandatory encryption is configured in samba server on a\n share (smb.conf parameter \"smb encrypt \u003d mandatory\") the\n server will hang up the tcp session when we try to send\n the first frame after the tree connect if it is not a\n QueryFSUnixInfo, this causes cifs mount to hang (it must\n be killed with ctl-c). Move the QueryFSUnixInfo call\n earlier in the mount sequence, and check whether the SetFSUnixInfo\n fails due to mandatory encryption so we can return a sensible\n error (EACCES) on mount.\n\n In a future patch (for 2.6.40) we will support mandatory\n encryption.\n\nReviewed-by: Pavel Shilovsky \u003cpiastry@etersoft.ru\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "af2d3c5a7d5bf8305a6930f866b865284e3db3f4", "tree": "0be2c1fba2f4e5bc31668b55b7c0f1e0ec695e41", "parents": [ "271f27554f70be8444aaf632dca5f91d33816753" ], "author": { "name": "Yang Ruirui", "email": "ruirui.r.yang@tieto.com", "time": "Sat Apr 16 19:17:48 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:53 2011 +0900" }, "message": "ext4: release page cache in ext4_mb_load_buddy error path\n\ncommit 26626f1172fb4f3f323239a6a5cf4e082643fa46 upstream.\n\nAdd missing page_cache_release in the error path of ext4_mb_load_buddy\n\nSigned-off-by: Yang Ruirui \u003cruirui.r.yang@tieto.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "271f27554f70be8444aaf632dca5f91d33816753", "tree": "a811411dcb91b1f93a5a44e0b15398f0f17087b5", "parents": [ "1ddeaea5338e280bf6ad194838a040ff2a13a9eb" ], "author": { "name": "Ted Ts\u0027o", "email": "tytso@mit.edu", "time": "Sat Apr 30 13:17:11 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:52 2011 +0900" }, "message": "jbd: fix fsync() tid wraparound bug\n\ncommit d9b01934d56a96d9f4ae2d6204d4ea78a36f5f36 upstream.\n\nIf an application program does not make any changes to the indirect\nblocks or extent tree, i_datasync_tid will not get updated. If there\nare enough commits (i.e., 2**31) such that tid_geq()\u0027s calculations\nwrap, and there isn\u0027t a currently active transaction at the time of\nthe fdatasync() call, this can end up triggering a BUG_ON in\nfs/jbd/commit.c:\n\n\tJ_ASSERT(journal-\u003ej_running_transaction !\u003d NULL);\n\nIt\u0027s pretty rare that this can happen, since it requires the use of\nfdatasync() plus *very* frequent and excessive use of fsync(). But\nwith the right workload, it can.\n\nWe fix this by replacing the use of tid_geq() with an equality test,\nsince there\u0027s only one valid transaction id that is valid for us to\nstart: namely, the currently running transaction (if it exists).\n\nReported-by: Martin_Zielinski@McAfee.com\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1ddeaea5338e280bf6ad194838a040ff2a13a9eb", "tree": "dc7032b9556b5ed84c7341306ba8e4fc1e8a2562", "parents": [ "d5994d14b3697d22117b2e3c91317c4bff87803a" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Thu May 05 13:59:35 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:52 2011 +0900" }, "message": "jbd: Fix forever sleeping process in do_get_write_access()\n\ncommit 2842bb20eed2e25cde5114298edc62c8883a1d9a upstream.\n\nIn do_get_write_access() we wait on BH_Unshadow bit for buffer to get\nfrom shadow state. The waking code in journal_commit_transaction() has\na bug because it does not issue a memory barrier after the buffer is moved\nfrom the shadow state and before wake_up_bit() is called. Thus a waitqueue\ncheck can happen before the buffer is actually moved from the shadow state\nand waiting process may never be woken. Fix the problem by issuing proper\nbarrier.\n\nReported-by: Tao Ma \u003cboyu.mt@taobao.com\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "d5994d14b3697d22117b2e3c91317c4bff87803a", "tree": "1baf739c646158202510e9390151237840696ed6", "parents": [ "811dfd3f3d8304443b9dc5a74a953f1b907e6129" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Apr 27 18:20:44 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:52 2011 +0900" }, "message": "ext3: Fix fs corruption when make_indexed_dir() fails\n\ncommit 86c4f6d85595cd7da635dc6985d27bfa43b1ae10 upstream.\n\nWhen make_indexed_dir() fails (e.g. because of ENOSPC) after it has allocated\nblock for index tree root, we did not properly mark all changed buffers dirty.\nThis lead to only some of these buffers being written out and thus effectively\ncorrupting the directory.\n\nFix the issue by marking all changed data dirty even in the error failure case.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1db58ae0a6e032378326412bd446ea23591ff50c", "tree": "192eff2166620883814ebfa182e2b5f56d22b746", "parents": [ "aa0ff541cfa5935a5d52f3807c8a4993c46a2799" ], "author": { "name": "Lukas Czerner", "email": "lczerner@redhat.com", "time": "Fri May 20 13:55:29 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:48 2011 +0900" }, "message": "ext4: fix possible use-after-free in ext4_remove_li_request()\n\ncommit 1bb933fb1fa8e4cb337a0d5dfd2ff4c0dc2073e8 upstream.\n\nWe need to take reference to the s_li_request after we take a mutex,\nbecause it might be freed since then, hence result in accessing old\nalready freed memory. Also we should protect the whole\next4_remove_li_request() because ext4_li_info might be in the process of\nbeing freed in ext4_lazyinit_thread().\n\nSigned-off-by: Lukas Czerner \u003clczerner@redhat.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "c63356c3ed12f1242ec7fb4b14a6a28b0e0c72a2", "tree": "4cf027c29ae40a5c2c9a11bb38788f546db425bf", "parents": [ "5fb46ae7b8c51b05a12c6a66108e8d398c20ee09" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Thu Apr 21 20:54:46 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:43 2011 +0900" }, "message": "block: don\u0027t block events on excl write for non-optical devices\n\ncommit d4dc210f69bcb0b4bef5a83b1c323817be89bad1 upstream.\n\nDisk event code automatically blocks events on excl write. This is\nprimarily to avoid issuing polling commands while burning is in\nprogress. This behavior doesn\u0027t fit other types of devices with\nremoveable media where polling commands don\u0027t have adverse side\neffects and door locking usually doesn\u0027t exist.\n\nThis patch introduces new genhd flag which controls the auto-blocking\nbehavior and uses it to enable auto-blocking only on optical devices.\n\nNote for stable: 2.6.38 and later only\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nReported-by: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "9d7e15702edac354221e82cfd64e21b99590e81c", "tree": "b82c87c1287147e7cd61ce93f13c7e0761bbda8f", "parents": [ "1b42653c2a64cd81723b19d011639c07c54909e6" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Mon May 23 13:26:07 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:41 2011 +0900" }, "message": "block: move bd_set_size() above rescan_partitions() in __blkdev_get()\n\ncommit 7e69723fef8771a9d57bd27d36281d756130b4b5 upstream.\n\n02e352287a4 (block: rescan partitions on invalidated devices on\n-ENOMEDIA too) relocated partition rescan above explicit bd_set_size()\nto simplify condition check. As rescan_partitions() does its own bdev\nsize setting, this doesn\u0027t break anything; however,\nrescan_partitions() prints out the following messages when adjusting\nbdev size, which can be confusing.\n\n sda: detected capacity change from 0 to 146815737856\n sdb: detected capacity change from 0 to 146815737856\n\nThis patch restores the original order and remove the warning\nmessages.\n\nstable: Please apply together with 02e352287a4 (block: rescan\n partitions on invalidated devices on -ENOMEDIA too).\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nReported-by: Tony Luck \u003ctony.luck@gmail.com\u003e\nTested-by: Tony Luck \u003ctony.luck@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n\n" }, { "commit": "1b42653c2a64cd81723b19d011639c07c54909e6", "tree": "2fd5fade8fcd9d90dfb8121ebc352a26f3885218", "parents": [ "05d0b1401071aff5e886533e424e1c7b4421a3a1" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Apr 29 10:15:20 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Jun 03 10:33:41 2011 +0900" }, "message": "block: rescan partitions on invalidated devices on -ENOMEDIA too\n\ncommit 02e352287a40bd456eb78df705bf888bc3161d3f upstream.\n\n__blkdev_get() doesn\u0027t rescan partitions if disk-\u003efops-\u003eopen() fails,\nwhich leads to ghost partition devices lingering after medimum removal\nis known to both the kernel and userland. The behavior also creates a\nsubtle inconsistency where O_NONBLOCK open, which doesn\u0027t fail even if\nthere\u0027s no medium, clears the ghots partitions, which is exploited to\nwork around the problem from userland.\n\nFix it by updating __blkdev_get() to issue partition rescan after\n-ENOMEDIA too.\n\nThis was reported in the following bz.\n\n https://bugzilla.kernel.org/show_bug.cgi?id\u003d13029\n\nStable: 2.6.38\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nReported-by: David Zeuthen \u003czeuthen@gmail.com\u003e\nReported-by: Martin Pitt \u003cmartin.pitt@ubuntu.com\u003e\nReported-by: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nTested-by: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "07f15da3d3d145982c883cf635527fa1d2175f4a", "tree": "c2b0959f8a167a9e0632540e4400e7dffc68e9ba", "parents": [ "b9c8360b42534c6762a2782dca07511b932cf4c3", "1c6dfdb912ee90ce3d20db20fec032e06a4315dd" ], "author": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Sun May 22 22:13:52 2011 -0400" }, "committer": { "name": "cayniarb", "email": "charlton.danielk@gmail.com", "time": "Sun May 22 22:13:52 2011 -0400" }, "message": "merge remote branch master of git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-2.6.38.y.git\n" }, { "commit": "abe43277b747dc3820678a5a919ca758013017f3", "tree": "d7e28dad24ddb8eed155570c14a1529268cd8d9d", "parents": [ "8f18c3721794569b3ac31f30f5e3f0ac983033a5" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue May 17 15:28:21 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:36 2011 -0700" }, "message": "cifs: fix cifsConvertToUCS() for the mapchars case\n\ncommit 11379b5e33950048ad66825da7f462b0d0da9d73 upstream.\n\nAs Metze pointed out, commit 84cdf74e broke mapchars option:\n\n Commit \"cifs: fix unaligned accesses in cifsConvertToUCS\"\n (84cdf74e8096a10dd6acbb870dd404b92f07a756) does multiple steps\n in just one commit (moving the function and changing it without\n testing).\n\n put_unaligned_le16(temp, \u0026target[j]); is never called for any\n codepoint the goes via the \u0027default\u0027 switch statement. As a result\n we put just zero (or maybe uninitialized) bytes into the target\n buffer.\n\nHis proposed patch looks correct, but doesn\u0027t apply to the current head\nof the tree. This patch should also fix it.\n\nReported-by: Stefan Metzmacher \u003cmetze@samba.org\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "8f18c3721794569b3ac31f30f5e3f0ac983033a5", "tree": "ccf16fe39cb967681aa6d98830cbde64ce84d5b4", "parents": [ "3ea80459be84832670d6888ca06b3aaffe32e6a4" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue Apr 05 15:02:37 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:36 2011 -0700" }, "message": "cifs: clean up various nits in unicode routines (try #2)\n\ncommit 581ade4d1c025eb10421eda0d0c0a2f04447d7c5 upstream.\n\nMinor revision to the original patch. Don\u0027t abuse the __le16 variable\non the stack by casting it to wchar_t and handing it off to char2uni.\nDeclare an actual wchar_t on the stack instead. This fixes a valid\nsparse warning.\n\nFix the spelling of UNI_ASTERISK. Eliminate the unneeded len_remaining\nvariable in cifsConvertToUCS.\n\nAlso, as David Howells points out. We were better off making\ncifsConvertToUCS *not* use put_unaligned_le16 since it means that we\ncan\u0027t optimize the mapped characters at compile time. Switch them\ninstead to use cpu_to_le16, and simply use put_unaligned to set them\nin the string.\n\nReported-and-acked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ba0bb0c9b289b919dc7b106a26ee05e1516171a8", "tree": "81b2723cc63e0406cf77e865ad46f783351faf52", "parents": [ "758b8cc02fae512d6748a87d1c8656a94b1fecaa" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue May 17 06:40:30 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:27 2011 -0700" }, "message": "cifs: add fallback in is_path_accessible for old servers\n\ncommit 221d1d797202984cb874e3ed9f1388593d34ee22 upstream.\n\nThe is_path_accessible check uses a QPathInfo call, which isn\u0027t\nsupported by ancient win9x era servers. Fall back to an older\nSMBQueryInfo call if it fails with the magic error codes.\n\nReported-and-Tested-by: Sandro Bonazzola \u003csandro.bonazzola@gmail.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "0d88dc66b6ca8011870d03b7fb1e65500a02bc4b", "tree": "6d75a5e5a9656ed2d57ae1359ae7a442b98e6f26", "parents": [ "a7d9222022375081576a7909167869a0bf1ed44c" ], "author": { "name": "Pavel Shilovsky", "email": "piastry@etersoft.ru", "time": "Thu Apr 14 22:00:56 2011 +0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:10 2011 -0700" }, "message": "CIFS: Fix memory over bound bug in cifs_parse_mount_options\n\ncommit 4906e50b37e6f6c264e7ee4237343eb2b7f8d16d upstream.\n\nWhile password processing we can get out of options array bound if\nthe next character after array is delimiter. The patch adds a check\nif we reach the end.\n\nSigned-off-by: Pavel Shilovsky \u003cpiastry@etersoft.ru\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@suse.de\u003e\n\n" }, { "commit": "e277c750ea81076d8dc68694acc7994a024d83b2", "tree": "5bb1a87f3c6f05d348a823be4d1685431bf09bcb", "parents": [ "3dead919f4cf95df59d8df06c1fa559c55571384" ], "author": { "name": "Mikulas Patocka", "email": "mikulas@artax.karlin.mff.cuni.cz", "time": "Mon May 09 13:01:09 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:07 2011 -0700" }, "message": "Don\u0027t lock guardpage if the stack is growing up\n\ncommit a09a79f66874c905af35d5bb5e5f2fdc7b6b894d upstream.\n\nLinux kernel excludes guard page when performing mlock on a VMA with\ndown-growing stack. However, some architectures have up-growing stack\nand locking the guard page should be excluded in this case too.\n\nThis patch fixes lvm2 on PA-RISC (and possibly other architectures with\nup-growing stack). lvm2 calculates number of used pages when locking and\nwhen unlocking and reports an internal error if the numbers mismatch.\n\n[ Patch changed fairly extensively to also fix /proc/\u003cpid\u003e/maps for the\n grows-up case, and to move things around a bit to clean it all up and\n share the infrstructure with the /proc bits.\n\n Tested on ia64 that has both grow-up and grow-down segments - Linus ]\n\nSigned-off-by: Mikulas Patocka \u003cmikulas@artax.karlin.mff.cuni.cz\u003e\nTested-by: Tony Luck \u003ctony.luck@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "f8ee0a32d68faa762ff0c6a5b8510370108e74e5", "tree": "54d17154bd553d9d929db117c264f66ec16ab91a", "parents": [ "ef65b13bc0be7792c8982264209d782f8ee1dd3c" ], "author": { "name": "Timo Warns", "email": "Warns@pre-sense.de", "time": "Fri May 06 13:47:35 2011 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:03 2011 -0700" }, "message": "Validate size of EFI GUID partition entries.\n\ncommit fa039d5f6b126fbd65eefa05db2f67e44df8f121 upstream.\n\nOtherwise corrupted EFI partition tables can cause total confusion.\n\nSigned-off-by: Timo Warns \u003cwarns@pre-sense.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "ef65b13bc0be7792c8982264209d782f8ee1dd3c", "tree": "f4d0e00e1d1cde3fe292771bf20f5f7d5648ea4b", "parents": [ "8004af443bb5537018555febbc93de467461abf6" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Fri Apr 29 06:52:44 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:03 2011 -0700" }, "message": "cifs: handle errors from coalesce_t2\n\ncommit 16541ba11c4f04ffe94b073e301f00b749fb84a1 upstream.\n\ncifs_demultiplex_thread calls coalesce_t2 to try and merge follow-on t2\nresponses into the original mid buffer. coalesce_t2 however can return\nerrors, but the caller doesn\u0027t handle that situation properly. Fix the\nthread to treat such a case as it would a malformed packet. Mark the\nmid as being malformed and issue the callback.\n\nAcked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "8004af443bb5537018555febbc93de467461abf6", "tree": "f4d5fbd6739895b0c503bf315c1276d24aaa9225", "parents": [ "e025b48277c11706feb296d535b15ded500fbfdd" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Fri Apr 29 06:52:43 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:02 2011 -0700" }, "message": "cifs: refactor mid finding loop in cifs_demultiplex_thread\n\ncommit 146f9f65bd13f56665205aed7205d531c810cb35 upstream.\n\n...to reduce the extreme indentation. This should introduce no\nbehavioral changes.\n\nAcked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "e025b48277c11706feb296d535b15ded500fbfdd", "tree": "1a71fc27c1de0cbb90b5b50d815ba3af3794f59e", "parents": [ "61e16b8c3e8f5657403acabbb5d4eb42e2e6b0a1" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed Apr 27 13:29:49 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:02 2011 -0700" }, "message": "cifs: sanitize length checking in coalesce_t2 (try #3)\n\ncommit 2a2047bc94d0efc316401170c3d078d9edc20dc4 upstream.\n\nThere are a couple of places in this code where these values can wrap or\ngo negative, and that could potentially end up overflowing the buffer.\nEnsure that that doesn\u0027t happen. Do all of the length calculation and\nchecks first, and only perform the memcpy after they pass.\n\nAlso, increase some stack variables to 32 bits to ensure that they don\u0027t\nwrap without being detected.\n\nFinally, change the error codes to be a bit more descriptive of any\nproblems detected. -EINVAL isn\u0027t very accurate.\n\nReported-and-Acked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "61e16b8c3e8f5657403acabbb5d4eb42e2e6b0a1", "tree": "cd9ca6bf3128fac96263cb9c3dcf721070124062", "parents": [ "96bea36718ae2f617dbbab2587acd6c7c0c9d3df" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed Apr 27 13:25:51 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:02 2011 -0700" }, "message": "cifs: check for bytes_remaining going to zero in CIFS_SessSetup\n\ncommit fcda7f4578bbf9717444ca6da8a421d21489d078 upstream.\n\nIt\u0027s possible that when we go to decode the string area in the\nSESSION_SETUP response, that bytes_remaining will be 0. Decrementing it at\nthat point will mean that it can go \"negative\" and wrap. Check for a\nbytes_remaining value of 0, and don\u0027t try to decode the string area if\nthat\u0027s the case.\n\nReported-and-Acked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "96bea36718ae2f617dbbab2587acd6c7c0c9d3df", "tree": "6c8197941ba24373450b9605c4503eff173e263d", "parents": [ "678562e527fd9979f1765ffa1eb34738fc174425" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed Apr 27 13:25:51 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Sat May 21 15:13:01 2011 -0700" }, "message": "cifs: change bleft in decode_unicode_ssetup back to signed type\n\ncommit bfacf2225a955bea9c41c707fc72ba16009674a0 upstream.\n\nThe buffer length checks in this function depend on this value being a\nsigned data type, but 690c522fa converted it to an unsigned type.\n\nAlso, eliminate a problem with the null termination check in the same\nfunction. cifs_strndup_from_ucs handles that situation correctly\nalready, and the existing check could potentially lead to a buffer\noverrun since it increments bleft without checking to see whether it\nfalls off the end of the buffer.\n\nReported-and-Acked-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "e8fb3db949f205177bd4f05f5cd831d3dd57aa4a", "tree": "3ac3086ecf6510b675f695410c0144db6f2f8ae3", "parents": [ "9eaffd03b8735c83dbd83cd7a4a7c2adabcf35ff" ], "author": { "name": "Timo Warns", "email": "Warns@pre-sense.de", "time": "Thu Apr 14 15:21:56 2011 -0700" }, "committer": { "name": "toastcfh", "email": "toastcfh@gmail.com", "time": "Tue May 10 18:17:44 2011 -0400" }, "message": "fs/partitions/ldm.c: fix oops caused by corrupted partition table\n\ncommit c340b1d640001c8c9ecff74f68fd90422ae2448a upstream.\n\nThe kernel automatically evaluates partition tables of storage devices.\nThe code for evaluating LDM partitions (in fs/partitions/ldm.c) contains\na bug that causes a kernel oops on certain corrupted LDM partitions.\nA kernel subsystem seems to crash, because, after the oops, the kernel no\nlonger recognizes newly connected storage devices.\n\nThe patch validates the value of vblk_size.\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Timo Warns \u003cwarns@pre-sense.de\u003e\nCc: Eugene Teo \u003ceugeneteo@kernel.sg\u003e\nCc: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nCc: Richard Russon \u003crich@flatcap.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@suse.de\u003e\n" }, { "commit": "7c72d9c4cc7bf40f49fa9ce878d1e8d9682332ba", "tree": "d4105fedadcf744a182120cb8d3877f409668b84", "parents": [ "0c7f1863c818ab6d84ba6731de664c14e5c6c5f0" ], "author": { "name": "Sachin Prabhu", "email": "sprabhu@redhat.com", "time": "Wed Apr 20 13:09:35 2011 +0100" }, "committer": { "name": "toastcfh", "email": "toastcfh@gmail.com", "time": "Tue May 10 18:16:15 2011 -0400" }, "message": "Open with O_CREAT flag set fails to open existing files on non writable directories\n\ncommit 1574dff8996ab1ed92c09012f8038b5566fce313 upstream.\n\nAn open on a NFS4 share using the O_CREAT flag on an existing file for\nwhich we have permissions to open but contained in a directory with no\nwrite permissions will fail with EACCES.\n\nA tcpdump shows that the client had set the open mode to UNCHECKED which\nindicates that the file should be created if it doesn\u0027t exist and\nencountering an existing flag is not an error. Since in this case the\nfile exists and can be opened by the user, the NFS server is wrong in\nattempting to check create permissions on the parent directory.\n\nThe patch adds a conditional statement to check for create permissions\nonly if the file doesn\u0027t exist.\n\nSigned-off-by: Sachin S. Prabhu \u003csprabhu@redhat.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "023d2bb9bbc68ef536abfbaf2183446f8ab09250", "tree": "27d10bf708fbe32b7d344efbf3a14434000fe221", "parents": [ "f30c4957dfc87de6bce9a9397fa0388630d7e9de" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Mon Apr 25 18:46:31 2011 +0300" }, "committer": { "name": "toastcfh", "email": "toastcfh@gmail.com", "time": "Tue May 10 17:56:37 2011 -0400" }, "message": "UBIFS: seek journal heads to the latest bud in replay\n\ncommit 52c6e6f990669deac3f370f1603815adb55a1dbd upstream.\n\nThis is the second fix of the following symptom:\n\nUBIFS error (pid 34456): could not find an empty LEB\n\nwhich sometimes happens after power cuts when we mount the file-system - UBIFS\nrefuses it with the above error message which comes from the\n\u0027ubifs_rcvry_gc_commit()\u0027 function. I can reproduce this using the integck test\nwith the UBIFS power cut emulation enabled.\n\nAnalysis of the problem.\n\nCurrently UBIFS replay seeks the journal heads to the last _replayed_ bud.\nBut the buds are replayed out-of-order, so the replay basically seeks journal\nheads to the \"random\" bud belonging to this head, and not to the _last_ one.\n\nThe result of this is that the GC head may be seeked to a full LEB with no free\nspace, or very little free space. And \u0027ubifs_rcvry_gc_commit()\u0027 tries to find a\nfully or mostly dirty LEB to match the current GC head (because we need to\ngarbage-collect that dirty LEB at one go, because we do not have @c-\u003egc_lnum).\nSo \u0027ubifs_find_dirty_leb()\u0027 fails and we fall back to finding an empty LEB and\nalso fail. As a result - recovery fails and mounting fails.\n\nThis patch teaches the replay to initialize the GC heads exactly to the latest\nbuds, i.e. the buds which have the largest sequence number in corresponding\nlog reference nodes.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "f30c4957dfc87de6bce9a9397fa0388630d7e9de", "tree": "c0e6097f19bd156a93371d2ed26fd3bb46260147", "parents": [ "43a7385ed9fe8db12c140c47d9b37e336a48f7e1" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Mon Apr 25 18:17:09 2011 +0300" }, "committer": { "name": "toastcfh", "email": "toastcfh@gmail.com", "time": "Tue May 10 17:56:06 2011 -0400" }, "message": "UBIFS: do not free write-buffers when in R/O mode\n\ncommit b50b9f408502a2ea90459ae36ba8cdc9cc005cfe upstream.\n\nCurrently UBIFS has a small optimization - it frees write-buffers when it is\nre-mounted from R/W mode to R/O mode. Of course, when it is mounted R/O, it\ndoes not allocate write-buffers as well.\n\nThis optimization is nice but it leads to subtle problems and complications\nin recovery, which I can reproduce using the integck test. The symptoms are\nthat after a power cut the file-system cannot be mounted if we first mount\nit R/O, and then re-mount R/W - \u0027ubifs_rcvry_gc_commit()\u0027 prints:\n\nUBIFS error (pid 34456): could not find an empty LEB\n\nAnalysis of the problem.\n\nWhen mounting R/W, the reply process sets journal heads to buds [1], but\nwhen mounting R/O - it does not do this, because the write-buffers are not\nallocated. So \u0027ubifs_rcvry_gc_commit()\u0027 works completely differently for the\nsame file-system but for the following 2 cases:\n\n1. mounting R/W after a power cut and recover\n2. mounting R/O after a power cut, re-mounting R/W and run deferred recovery\n\nIn the former case, we have journal heads seeked to the a bud, in the latter\ncase, they are non-seeked (wbuf-\u003elnum \u003d\u003d -1). So in the latter case we do not\ntry to recover the GC LEB by garbage-collecting to the GC head, but we just\ntry to find an empty LEB, and there may be no empty LEBs, so we just fail.\nOn the other hand, in the former case (mount R/W), we are able to make a GC LEB\n(@c-\u003egc_lnum) by garbage-collecting.\n\nThus, let\u0027s remove this small nice optimization and always allocate\nwrite-buffers. This should not make too big difference - we have only 3\nof them, each of max. write unit size, which is usually 2KiB. So this is\nabout 6KiB of RAM for the typical case, and only when mounted R/O.\n\n[1]: Note, currently the replay process is setting (seeking) the journal heads\nto _some_ buds, not necessarily to the buds which had been the journal heads\nbefore the power cut happened. This will be fixed separately.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "c99afaf0385d61a8e1747511904c66838270b02b", "tree": "4960232669661ffea3ca2357082a2c540f061bef", "parents": [ "ba1ba1c0aaa659914bf9d47616453f6378add40f" ], "author": { "name": "Timo Warns", "email": "Warns@pre-sense.de", "time": "Thu Apr 14 15:21:56 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 09 15:06:49 2011 -0700" }, "message": "fs/partitions/ldm.c: fix oops caused by corrupted partition table\n\ncommit c340b1d640001c8c9ecff74f68fd90422ae2448a upstream.\n\nThe kernel automatically evaluates partition tables of storage devices.\nThe code for evaluating LDM partitions (in fs/partitions/ldm.c) contains\na bug that causes a kernel oops on certain corrupted LDM partitions.\nA kernel subsystem seems to crash, because, after the oops, the kernel no\nlonger recognizes newly connected storage devices.\n\nThe patch validates the value of vblk_size.\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Timo Warns \u003cwarns@pre-sense.de\u003e\nCc: Eugene Teo \u003ceugeneteo@kernel.sg\u003e\nCc: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nCc: Richard Russon \u003crich@flatcap.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@suse.de\u003e\n\n" }, { "commit": "6f0a6240e8ce1659574e944eac04e871a9dc1fe7", "tree": "593d99950553e629afd0540628784c18c93826cd", "parents": [ "7b17624a7f6b41678d7452ebd5a525ddc3aaf3d3" ], "author": { "name": "Sachin Prabhu", "email": "sprabhu@redhat.com", "time": "Wed Apr 20 13:09:35 2011 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 09 15:06:48 2011 -0700" }, "message": "Open with O_CREAT flag set fails to open existing files on non writable directories\n\ncommit 1574dff8996ab1ed92c09012f8038b5566fce313 upstream.\n\nAn open on a NFS4 share using the O_CREAT flag on an existing file for\nwhich we have permissions to open but contained in a directory with no\nwrite permissions will fail with EACCES.\n\nA tcpdump shows that the client had set the open mode to UNCHECKED which\nindicates that the file should be created if it doesn\u0027t exist and\nencountering an existing flag is not an error. Since in this case the\nfile exists and can be opened by the user, the NFS server is wrong in\nattempting to check create permissions on the parent directory.\n\nThe patch adds a conditional statement to check for create permissions\nonly if the file doesn\u0027t exist.\n\nSigned-off-by: Sachin S. Prabhu \u003csprabhu@redhat.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "f038f4d893e9ef35dd38f1ca01001dfe21c5f8f5", "tree": "c6f963140058fd3631021d69c748c53c7486bd51", "parents": [ "21fb89793df408cc116aa52399a5e6f7af459596" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Mon Apr 25 18:46:31 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 09 15:06:39 2011 -0700" }, "message": "UBIFS: seek journal heads to the latest bud in replay\n\ncommit 52c6e6f990669deac3f370f1603815adb55a1dbd upstream.\n\nThis is the second fix of the following symptom:\n\nUBIFS error (pid 34456): could not find an empty LEB\n\nwhich sometimes happens after power cuts when we mount the file-system - UBIFS\nrefuses it with the above error message which comes from the\n\u0027ubifs_rcvry_gc_commit()\u0027 function. I can reproduce this using the integck test\nwith the UBIFS power cut emulation enabled.\n\nAnalysis of the problem.\n\nCurrently UBIFS replay seeks the journal heads to the last _replayed_ bud.\nBut the buds are replayed out-of-order, so the replay basically seeks journal\nheads to the \"random\" bud belonging to this head, and not to the _last_ one.\n\nThe result of this is that the GC head may be seeked to a full LEB with no free\nspace, or very little free space. And \u0027ubifs_rcvry_gc_commit()\u0027 tries to find a\nfully or mostly dirty LEB to match the current GC head (because we need to\ngarbage-collect that dirty LEB at one go, because we do not have @c-\u003egc_lnum).\nSo \u0027ubifs_find_dirty_leb()\u0027 fails and we fall back to finding an empty LEB and\nalso fail. As a result - recovery fails and mounting fails.\n\nThis patch teaches the replay to initialize the GC heads exactly to the latest\nbuds, i.e. the buds which have the largest sequence number in corresponding\nlog reference nodes.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "21fb89793df408cc116aa52399a5e6f7af459596", "tree": "257b9e64479d83ef9c5d19a643ece07f1922f53e", "parents": [ "b000ec13075615ace30250ab6e355a3f0c956735" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Mon Apr 25 18:17:09 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 09 15:06:39 2011 -0700" }, "message": "UBIFS: do not free write-buffers when in R/O mode\n\ncommit b50b9f408502a2ea90459ae36ba8cdc9cc005cfe upstream.\n\nCurrently UBIFS has a small optimization - it frees write-buffers when it is\nre-mounted from R/W mode to R/O mode. Of course, when it is mounted R/O, it\ndoes not allocate write-buffers as well.\n\nThis optimization is nice but it leads to subtle problems and complications\nin recovery, which I can reproduce using the integck test. The symptoms are\nthat after a power cut the file-system cannot be mounted if we first mount\nit R/O, and then re-mount R/W - \u0027ubifs_rcvry_gc_commit()\u0027 prints:\n\nUBIFS error (pid 34456): could not find an empty LEB\n\nAnalysis of the problem.\n\nWhen mounting R/W, the reply process sets journal heads to buds [1], but\nwhen mounting R/O - it does not do this, because the write-buffers are not\nallocated. So \u0027ubifs_rcvry_gc_commit()\u0027 works completely differently for the\nsame file-system but for the following 2 cases:\n\n1. mounting R/W after a power cut and recover\n2. mounting R/O after a power cut, re-mounting R/W and run deferred recovery\n\nIn the former case, we have journal heads seeked to the a bud, in the latter\ncase, they are non-seeked (wbuf-\u003elnum \u003d\u003d -1). So in the latter case we do not\ntry to recover the GC LEB by garbage-collecting to the GC head, but we just\ntry to find an empty LEB, and there may be no empty LEBs, so we just fail.\nOn the other hand, in the former case (mount R/W), we are able to make a GC LEB\n(@c-\u003egc_lnum) by garbage-collecting.\n\nThus, let\u0027s remove this small nice optimization and always allocate\nwrite-buffers. This should not make too big difference - we have only 3\nof them, each of max. write unit size, which is usually 2KiB. So this is\nabout 6KiB of RAM for the typical case, and only when mounted R/O.\n\n[1]: Note, currently the replay process is setting (seeking) the journal heads\nto _some_ buds, not necessarily to the buds which had been the journal heads\nbefore the power cut happened. This will be fixed separately.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "61b5fea3a4fe2185455c019133678f987c0aac48", "tree": "701fe3ddd08b1faed229568b5e62b1169cb308e9", "parents": [ "d8a540b4ae9ed6fe089affb0389eef0593ab4a4a" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 15 17:34:18 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:04:52 2011 -0400" }, "message": "NFSv4.1: Ensure state manager thread dies on last umount\n\ncommit 47c2199b6eb5fbe38ddb844db7cdbd914d304f9c upstream.\n\nCurrently, the state manager may continue to try recovering state forever\neven after the last filesystem to reference that nfs_client has umounted.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "d8a540b4ae9ed6fe089affb0389eef0593ab4a4a", "tree": "20877cb0fe9f85a0f7d40124f24509c54bf4db77", "parents": [ "a59e1f33d21d869205f2f0449db77f7b67028d39" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed Apr 27 11:49:09 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:04:48 2011 -0400" }, "message": "nfs: don\u0027t lose MS_SYNCHRONOUS on remount of noac mount\n\ncommit 26c4c170731f00008f4317a2888a0a07ac99d90d upstream.\n\nOn a remount, the VFS layer will clear the MS_SYNCHRONOUS bit on the\nassumption that the flags on the mount syscall will have it set if the\nremounted fs is supposed to keep it.\n\nIn the case of \"noac\" though, MS_SYNCHRONOUS is implied. A remount of\nsuch a mount will lose the MS_SYNCHRONOUS flag since \"sync\" isn\u0027t part\nof the mount options.\n\nReported-by: Max Matveev \u003cmakc@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "a59e1f33d21d869205f2f0449db77f7b67028d39", "tree": "0d19763b024618d78bca99eaa8d032f7c211837e", "parents": [ "f00c65086bfe6f69bb99fc62c2b41df9c3bee613" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Apr 27 15:26:41 2011 -0700" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:04:43 2011 -0400" }, "message": "vfs: avoid large kmalloc()s for the fdtable\n\ncommit 6d4831c283530a5f2c6bd8172c13efa236eb149d upstream.\n\nAzurit reports large increases in system time after 2.6.36 when running\nApache. It was bisected down to a892e2d7dcdfa6c76e6 (\"vfs: use kmalloc()\nto allocate fdmem if possible\").\n\nThat patch caused the vfs to use kmalloc() for very large allocations and\nthis is causing excessive work (and presumably excessive reclaim) within\nthe page allocator.\n\nFix it by falling back to vmalloc() earlier - when the allocation attempt\nwould have been considered \"costly\" by reclaim.\n\nReported-by: azurIt \u003cazurit@pobox.sk\u003e\nTested-by: azurIt \u003cazurit@pobox.sk\u003e\nAcked-by: Changli Gao \u003cxiaosuo@gmail.com\u003e\nCc: Americo Wang \u003cxiyou.wangcong@gmail.com\u003e\nCc: Jiri Slaby \u003cjslaby@suse.cz\u003e\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\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@suse.de\u003e\n" }, { "commit": "301f317a24ba923485c25d6fa27515f353ac0e1d", "tree": "0e36a41318f4793fa07dcbc1ec110a6ac4a6c35c", "parents": [ "ce5fe399e508b41787ea04761fdf9a29c1caa7aa" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Wed Apr 20 18:02:45 2011 +0300" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:03:44 2011 -0400" }, "message": "UBIFS: fix false space checking failure\n\ncommit 8c230d9a5b5ec7970139acb7e2d165d7a3fe9f9e upstream.\n\nThis patch fixes UBIFS mount failure when the debugging support is enabled,\nwe are recovering from a power cut, we were first mounter R/O and we are\nre-mounting R/W. In this case we should not assume that the amount of free\nspace before we have re-mounted R/W and after are equivalent, because\nwhen we have mounted R/O the file-system is in a non-committed state so\nthe amount of free space is slightly smaller, due to the fact that we cannot\npredict the amount of free space precisely before we commit.\n\nThis patch fixes the issue by skipping the debugging check in case of\nrecovery. This issue was reported by Caizhiyong \u003ccaizhiyong@huawei.com\u003e\nhere: http://thread.gmane.org/gmane.linux.drivers.mtd/34350/focus\u003d34387\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReported-by: Caizhiyong \u003ccaizhiyong@huawei.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "87b9000d551994b2e4faada850c798f4c01cd504", "tree": "c62d77c8566f74be977605955883997f6347542f", "parents": [ "4fb705f7150fec7c5bd27d9b5bb38a9fbeb11bc6" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Thu Apr 21 14:49:55 2011 +0300" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:03:25 2011 -0400" }, "message": "UBIFS: fix master node recovery\n\ncommit 6e0d9fd38b750d678bf9fd07db23582f52fafa55 upstream.\n\nThis patch fixes the following symptoms:\n1. Unmount UBIFS cleanly.\n2. Start mounting UBIFS R/W and have a power cut immediately\n3. Start mounting UBIFS R/O, this succeeds\n4. Try to re-mount UBIFS R/W - this fails immediately or later on,\n because UBIFS will write the master node to the flash area\n which has been written before.\n\nThe analysis of the problem:\n\n1. UBIFS is unmounted cleanly, both copies of the master node are clean.\n2. UBIFS is being mounter R/W, starts changing master node copy 1, and\n a power cut happens. The copy N1 becomes corrupted.\n3. UBIFS is being mounted R/O. It notices the copy N1 is corrupted and\n reads copy N2. Copy N2 is clean.\n4. Because of R/O mode, UBIFS cannot recover copy 1.\n5. The mount code (ubifs_mount()) sees that the master node is clean,\n so it decides that no recovery is needed.\n6. We are re-mounting R/W. UBIFS believes no recovery is needed and\n starts updating the master node, but copy N1 is still corrupted\n and was not recovered!\n\nFix this problem by marking the master node as dirty every time we\nrecover it and we are in R/O mode. This forces further recovery and\nthe UBIFS cleans-up the corruptions and recovers the copy N1 when\nre-mounting R/W later.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "c1cc22677d272a7a192200b19bfc2d9e03e641ab", "tree": "b2051540aedc468fd119a7dbe9bc33e69f175d35", "parents": [ "575eae895c6c52834ba728cf564d6cff439c09a5" ], "author": { "name": "OGAWA Hirofumi", "email": "hirofumi@mail.parknet.co.jp", "time": "Mon Apr 18 11:48:55 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:01:48 2011 -0400" }, "message": "nfsd4: Fix filp leak\n\ncommit a96e5b90804be8b540d30f4a1453fc87f95b3149 upstream.\n\n23fcf2ec93fb8573a653408316af599939ff9a8e (nfsd4: fix oops on lock failure)\n\nThe above patch breaks free path for stp-\u003est_file. If stp was inserted\ninto sop-\u003eso_stateids, we have to free stp-\u003est_file refcount. Because\nstp-\u003est_file refcount itself is taken whether or not any refcounts are\ntaken on the stp-\u003est_file-\u003efi_fds[].\n\nSigned-off-by: OGAWA Hirofumi \u003chirofumi@mail.parknet.co.jp\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "575eae895c6c52834ba728cf564d6cff439c09a5", "tree": "a7541819caf4a2f16dae5db45c1f20f0d4713d60", "parents": [ "75f014b78c4d45d565c0ce106d7d62cd58a17bb7" ], "author": { "name": "J. Bruce Fields", "email": "bfields@redhat.com", "time": "Fri Apr 15 18:08:26 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Mon May 02 19:01:43 2011 -0400" }, "message": "nfsd4: fix struct file leak on delegation\n\ncommit 4ee63624fd927376b97ead3a8d00728d437bc8e8 upstream.\n\nIntroduced by acfdf5c383b38f7f4dddae41b97c97f1ae058f49.\n\nReported-by: Gerhard Heift \u003cml-nfs-linux-20110412-ef47@gheift.de\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "20857f111ec4c3e5aea039c5c06d5254aeeb620c", "tree": "866780cec4cd0d1e2e52e8d8f3c71f62bb1f77dd", "parents": [ "558cc1035db8ac0a510b4582f5354de1510224d6" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Apr 15 17:34:18 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:46 2011 -0700" }, "message": "NFSv4.1: Ensure state manager thread dies on last umount\n\ncommit 47c2199b6eb5fbe38ddb844db7cdbd914d304f9c upstream.\n\nCurrently, the state manager may continue to try recovering state forever\neven after the last filesystem to reference that nfs_client has umounted.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "558cc1035db8ac0a510b4582f5354de1510224d6", "tree": "d39abc5836a1bd041f761542d991e8df177733ca", "parents": [ "146129d6d999532951565a0c25bc98e724b1b07e" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Wed Apr 27 11:49:09 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:46 2011 -0700" }, "message": "nfs: don\u0027t lose MS_SYNCHRONOUS on remount of noac mount\n\ncommit 26c4c170731f00008f4317a2888a0a07ac99d90d upstream.\n\nOn a remount, the VFS layer will clear the MS_SYNCHRONOUS bit on the\nassumption that the flags on the mount syscall will have it set if the\nremounted fs is supposed to keep it.\n\nIn the case of \"noac\" though, MS_SYNCHRONOUS is implied. A remount of\nsuch a mount will lose the MS_SYNCHRONOUS flag since \"sync\" isn\u0027t part\nof the mount options.\n\nReported-by: Max Matveev \u003cmakc@redhat.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "146129d6d999532951565a0c25bc98e724b1b07e", "tree": "d37402222b46dfeacd134fb0a90dd808f57a2363", "parents": [ "8858587af25efc06d5cce42676786b3d7a9160f2" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Apr 27 15:26:41 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:45 2011 -0700" }, "message": "vfs: avoid large kmalloc()s for the fdtable\n\ncommit 6d4831c283530a5f2c6bd8172c13efa236eb149d upstream.\n\nAzurit reports large increases in system time after 2.6.36 when running\nApache. It was bisected down to a892e2d7dcdfa6c76e6 (\"vfs: use kmalloc()\nto allocate fdmem if possible\").\n\nThat patch caused the vfs to use kmalloc() for very large allocations and\nthis is causing excessive work (and presumably excessive reclaim) within\nthe page allocator.\n\nFix it by falling back to vmalloc() earlier - when the allocation attempt\nwould have been considered \"costly\" by reclaim.\n\nReported-by: azurIt \u003cazurit@pobox.sk\u003e\nTested-by: azurIt \u003cazurit@pobox.sk\u003e\nAcked-by: Changli Gao \u003cxiaosuo@gmail.com\u003e\nCc: Americo Wang \u003cxiyou.wangcong@gmail.com\u003e\nCc: Jiri Slaby \u003cjslaby@suse.cz\u003e\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\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@suse.de\u003e\n\n" }, { "commit": "718731e7c909827e15dd3a43d169eeb4bf6c6f8d", "tree": "32635586b1e687475d9a67b9ccd4fea383b0243a", "parents": [ "5002397e6c6f9c2f5fb92d8990eb2663e43a3012" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Wed Apr 20 18:02:45 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:42 2011 -0700" }, "message": "UBIFS: fix false space checking failure\n\ncommit 8c230d9a5b5ec7970139acb7e2d165d7a3fe9f9e upstream.\n\nThis patch fixes UBIFS mount failure when the debugging support is enabled,\nwe are recovering from a power cut, we were first mounter R/O and we are\nre-mounting R/W. In this case we should not assume that the amount of free\nspace before we have re-mounted R/W and after are equivalent, because\nwhen we have mounted R/O the file-system is in a non-committed state so\nthe amount of free space is slightly smaller, due to the fact that we cannot\npredict the amount of free space precisely before we commit.\n\nThis patch fixes the issue by skipping the debugging check in case of\nrecovery. This issue was reported by Caizhiyong \u003ccaizhiyong@huawei.com\u003e\nhere: http://thread.gmane.org/gmane.linux.drivers.mtd/34350/focus\u003d34387\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReported-by: Caizhiyong \u003ccaizhiyong@huawei.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "75db8ad812878495309d3d0b40467e9b9b61b29a", "tree": "07e0f21bf8506398d140a1c4fcf4bc0435a3042a", "parents": [ "fb57290759588bbcb60d6d9104f8947fdba5aa19" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Thu Apr 21 14:49:55 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:40 2011 -0700" }, "message": "UBIFS: fix master node recovery\n\ncommit 6e0d9fd38b750d678bf9fd07db23582f52fafa55 upstream.\n\nThis patch fixes the following symptoms:\n1. Unmount UBIFS cleanly.\n2. Start mounting UBIFS R/W and have a power cut immediately\n3. Start mounting UBIFS R/O, this succeeds\n4. Try to re-mount UBIFS R/W - this fails immediately or later on,\n because UBIFS will write the master node to the flash area\n which has been written before.\n\nThe analysis of the problem:\n\n1. UBIFS is unmounted cleanly, both copies of the master node are clean.\n2. UBIFS is being mounter R/W, starts changing master node copy 1, and\n a power cut happens. The copy N1 becomes corrupted.\n3. UBIFS is being mounted R/O. It notices the copy N1 is corrupted and\n reads copy N2. Copy N2 is clean.\n4. Because of R/O mode, UBIFS cannot recover copy 1.\n5. The mount code (ubifs_mount()) sees that the master node is clean,\n so it decides that no recovery is needed.\n6. We are re-mounting R/W. UBIFS believes no recovery is needed and\n starts updating the master node, but copy N1 is still corrupted\n and was not recovered!\n\nFix this problem by marking the master node as dirty every time we\nrecover it and we are in R/O mode. This forces further recovery and\nthe UBIFS cleans-up the corruptions and recovers the copy N1 when\nre-mounting R/W later.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "eceb743cdc65d1b78d91f786cc2a2acb9ab2d116", "tree": "cd56fd72c206c9c9e4e8f8c68f499c33a4005f21", "parents": [ "9acb84c8847ed2008ea86d1e0635f30b3db17026" ], "author": { "name": "OGAWA Hirofumi", "email": "hirofumi@mail.parknet.co.jp", "time": "Mon Apr 18 11:48:55 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:34 2011 -0700" }, "message": "nfsd4: Fix filp leak\n\ncommit a96e5b90804be8b540d30f4a1453fc87f95b3149 upstream.\n\n23fcf2ec93fb8573a653408316af599939ff9a8e (nfsd4: fix oops on lock failure)\n\nThe above patch breaks free path for stp-\u003est_file. If stp was inserted\ninto sop-\u003eso_stateids, we have to free stp-\u003est_file refcount. Because\nstp-\u003est_file refcount itself is taken whether or not any refcounts are\ntaken on the stp-\u003est_file-\u003efi_fds[].\n\nSigned-off-by: OGAWA Hirofumi \u003chirofumi@mail.parknet.co.jp\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "9acb84c8847ed2008ea86d1e0635f30b3db17026", "tree": "ca3d36fc58c04092f57845f31bbf2c71c457729f", "parents": [ "da0309a4a778c46eaba98c58a2b821349d658520" ], "author": { "name": "J. Bruce Fields", "email": "bfields@redhat.com", "time": "Fri Apr 15 18:08:26 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon May 02 09:19:34 2011 -0700" }, "message": "nfsd4: fix struct file leak on delegation\n\ncommit 4ee63624fd927376b97ead3a8d00728d437bc8e8 upstream.\n\nIntroduced by acfdf5c383b38f7f4dddae41b97c97f1ae058f49.\n\nReported-by: Gerhard Heift \u003cml-nfs-linux-20110412-ef47@gheift.de\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "d7112d551c57d5b5f3b7aa342c9c16f79d6c806b", "tree": "3cd0ca8aca5941fcf3436adc593656920ac0bdb1", "parents": [ "de17cd0ce567c5de100ab99a5b03f784c54e64a2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 18 10:36:54 2011 -0700" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:41:09 2011 -0400" }, "message": "proc: do proper range check on readdir offset\n\ncommit d8bdc59f215e62098bc5b4256fd9928bf27053a1 upstream.\n\nRather than pass in some random truncated offset to the pid-related\nfunctions, check that the offset is in range up-front.\n\nThis is just cleanup, the previous commit fixed the real problem.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "3608428decef537a1d9d7f5f6df1c6314f480fef", "tree": "f94aa544ab027c36dce79211b4f6131158c75243", "parents": [ "0f00f08f776fdb920da1b1b9e03dfeb7314e8d46" ], "author": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Thu Apr 14 15:22:09 2011 -0700" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:40:35 2011 -0400" }, "message": "brk: COMPAT_BRK: fix detection of randomized brk\n\ncommit 4471a675dfc7ca676c165079e91c712b09dc9ce4 upstream.\n\n5520e89 (\"brk: fix min_brk lower bound computation for COMPAT_BRK\")\ntried to get the whole logic of brk randomization for legacy\n(libc5-based) applications finally right.\n\nIt turns out that the way to detect whether brk has actually been\nrandomized in the end or not introduced by that patch still doesn\u0027t work\nfor those binaries, as reported by Geert:\n\n: /sbin/init from my old m68k ramdisk exists prematurely.\n:\n: Before the patch:\n:\n: | brk(0x80005c8e) \u003d 0x80006000\n:\n: After the patch:\n:\n: | brk(0x80005c8e) \u003d 0x80005c8e\n:\n: Old libc5 considers brk() to have failed if the return value is not\n: identical to the requested value.\n\nI don\u0027t like it, but currently see no better option than a bit flag in\ntask_struct to catch the CONFIG_COMPAT_BRK \u0026\u0026 randomize_va_space \u003d\u003d 2\ncase.\n\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\nTested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nReported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.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@suse.de\u003e\n" }, { "commit": "9da573a886cef8095b49baae25671d8336e2e961", "tree": "a9b77d90be7c50fe1416130b8fdcd1b1a8a03d07", "parents": [ "cd1e974b08622d7a328d1effb22713704193c522" ], "author": { "name": "Tim Chen", "email": "tim.c.chen@linux.intel.com", "time": "Fri Apr 15 11:39:29 2011 -0700" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:40:06 2011 -0400" }, "message": "vfs: Fix absolute RCU path walk failures due to uninitialized seq number\n\ncommit c1530019e311c91d14b24d8e74d233152d806e45 upstream.\n\nDuring RCU walk in path_lookupat and path_openat, the rcu lookup\nfrequently failed if looking up an absolute path, because when root\ndirectory was looked up, seq number was not properly set in nameidata.\n\nWe dropped out of RCU walk in nameidata_drop_rcu due to mismatch in\ndirectory entry\u0027s seq number. We reverted to slow path walk that need\nto take references.\n\nWith the following patch, I saw a 50% increase in an exim mail server\nbenchmark throughput on a 4-socket Nehalem-EX system.\n\nSigned-off-by: Tim Chen \u003ctim.c.chen@linux.intel.com\u003e\nReviewed-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "ee6d9264c13bb4c88a52c39f5ea3f37ec7be6f8e", "tree": "a5f32fb0def02644755fbe5ca42b793baaff92c6", "parents": [ "cda24233fa38d46fc7d32dbd1d8e782e03768bb6" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Wed Apr 13 10:31:52 2011 +0300" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:39:52 2011 -0400" }, "message": "UBIFS: fix oops when R/O file-system is fsync\u0027ed\n\ncommit 78530bf7f2559b317c04991b52217c1608d5a58d upstream.\n\nThis patch fixes severe UBIFS bug: UBIFS oopses when we \u0027fsync()\u0027 an\nfile on R/O-mounter file-system. We (the UBIFS authors) incorrectly\nthought that VFS would not propagate \u0027fsync()\u0027 down to the file-system\nif it is read-only, but this is not the case.\n\nIt is easy to exploit this bug using the following simple perl script:\n\nuse strict;\nuse File::Sync qw(fsync sync);\n\ndie \"File path is not specified\" if not defined $ARGV[0];\nmy $path \u003d $ARGV[0];\n\nopen FILE, \"\u003c\", \"$path\" or die \"Cannot open $path: $!\";\nfsync(\\*FILE) or die \"cannot fsync $path: $!\";\nclose FILE or die \"Cannot close $path: $!\";\n\nThanks to Reuben Dowle \u003cReuben.Dowle@navico.com\u003e for reporting about this\nissue.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReported-by: Reuben Dowle \u003cReuben.Dowle@navico.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "4d4adc6ef5f0a4457d91d1e039ba617bfc8779d0", "tree": "8f3b425e6d14947ae58bbd491a9eb5ba168fa9f1", "parents": [ "bb6bb11280ba6a424de78ba6c9388cbc95bfc572" ], "author": { "name": "Bob Liu", "email": "lliubbo@gmail.com", "time": "Thu Apr 14 15:22:20 2011 -0700" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:39:38 2011 -0400" }, "message": "ramfs: fix memleak on no-mmu arch\n\ncommit b836aec53e2bce71de1d5415313380688c851477 upstream.\n\nOn no-mmu arch, there is a memleak during shmem test. The cause of this\nmemleak is ramfs_nommu_expand_for_mapping() added page refcount to 2\nwhich makes iput() can\u0027t free that pages.\n\nThe simple test file is like this:\n\n int main(void)\n {\n\tint i;\n\tkey_t k \u003d ftok(\"/etc\", 42);\n\n\tfor ( i\u003d0; i\u003c100; ++i) {\n\t\tint id \u003d shmget(k, 10000, 0644|IPC_CREAT);\n\t\tif (id \u003d\u003d -1) {\n\t\t\tprintf(\"shmget error\\n\");\n\t\t}\n\t\tif(shmctl(id, IPC_RMID, NULL ) \u003d\u003d -1) {\n\t\t\tprintf(\"shm rm error\\n\");\n\t\t\treturn -1;\n\t\t}\n\t}\n\tprintf(\"run ok...\\n\");\n\treturn 0;\n }\n\nAnd the result:\n\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 17912 42408 0 0\n -/+ buffers: 17912 42408\n root:/\u003e shmem\n run ok...\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 19096 41224 0 0\n -/+ buffers: 19096 41224\n root:/\u003e shmem\n run ok...\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 20296 40024 0 0\n -/+ buffers: 20296 40024\n ...\n\nAfter this patch the test result is:(no memleak anymore)\n\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 16668 43652 0 0\n -/+ buffers: 16668 43652\n root:/\u003e shmem\n run ok...\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 16668 43652 0 0\n -/+ buffers: 16668 43652\n\nSigned-off-by: Bob Liu \u003clliubbo@gmail.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.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@suse.de\u003e\n" }, { "commit": "0b74830b7153a982b83c2bcb1bff85b951c54833", "tree": "4ebf969dcea3f61ebe3da3ae87c46cd2b1d4266e", "parents": [ "72cc40c6b67760e48425561b7d6195be83f152cc" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Tue Mar 29 09:45:21 2011 +0300" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:37:55 2011 -0400" }, "message": "UBIFS: fix assertion warnings\n\ncommit c88ac00c5af70c2a0741da14b22cdcf8507ddd92 upstream.\n\nThis patch fixes UBIFS assertion warnings like:\n\nUBIFS assert failed in ubifs_leb_unmap at 135 (pid 29365)\nPid: 29365, comm: integck Tainted: G I 2.6.37-ubi-2.6+ #34\nCall Trace:\n [\u003cffffffffa047c663\u003e] ubifs_lpt_init+0x95e/0x9ee [ubifs]\n [\u003cffffffffa04623a7\u003e] ubifs_remount_fs+0x2c7/0x762 [ubifs]\n [\u003cffffffff810f066e\u003e] do_remount_sb+0xb6/0x101\n [\u003cffffffff81106ff4\u003e] ? do_mount+0x191/0x78e\n [\u003cffffffff811070bb\u003e] do_mount+0x258/0x78e\n [\u003cffffffff810da1e8\u003e] ? alloc_pages_current+0xa2/0xc5\n [\u003cffffffff81107674\u003e] sys_mount+0x83/0xbd\n [\u003cffffffff81009a12\u003e] system_call_fastpath+0x16/0x1b\n\nThey happen when we re-mount from R/O mode to R/W mode. While\nre-mounting, we write to the media, but we still have the c-\u003ero_mount\nflag set. The fix is very simple - just clear the flag before\nstarting re-mounting R/W.\n\nThese warnings are caused by the following commit:\n2ef13294d29bcfb306e0d360f1b97f37b647b0c0\n\nFor -stable guys: this bug was introduced in 2.6.38, this is materieal\nfor 2.6.38-stable.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "04fca00b6f5bea42c47f7ad495c2a849aedf4dac", "tree": "57b2ff4c110a9f2bd10349743316a48edd67695f", "parents": [ "e3016b2acd0c790d5a4ebbc7742741057ec5b206" ], "author": { "name": "Vasiliy Kulikov", "email": "segoon@openwall.com", "time": "Fri Feb 04 15:24:19 2011 +0300" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:37:21 2011 -0400" }, "message": "UBIFS: restrict world-writable debugfs files\n\ncommit 8c559d30b4e59cf6994215ada1fe744928f494bf upstream.\n\nDon\u0027t allow everybody to dump sensitive information about filesystems.\n\nSigned-off-by: Vasiliy Kulikov \u003csegoon@openwall.com\u003e\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "c45b1f155fe706ed626db1987f139f9915a495d3", "tree": "0e9f2055c56ecad28bdbdefb45ba5f1a98e6adcc", "parents": [ "a07f1bfabdef56710dd9c9bdc101beec18eaa04a" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Sat Apr 02 07:34:30 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:37:11 2011 -0400" }, "message": "cifs: wrap received signature check in srv_mutex\n\ncommit 157c249114508aa71daa308a426e15d81a4eed00 upstream.\n\nWhile testing my patchset to fix asynchronous writes, I hit a bunch\nof signature problems when testing with signing on. The problem seems\nto be that signature checks on receive can be running at the same\ntime as a process that is sending, or even that multiple receives can\nbe checking signatures at the same time, clobbering the same data\nstructures.\n\nWhile we\u0027re at it, clean up the comments over cifs_calculate_signature\nand add a note that the srv_mutex should be held when calling this\nfunction.\n\nThis patch seems to fix the problems for me, but I\u0027m not clear on\nwhether it\u0027s the best approach. If it is, then this should probably\ngo to stable too.\n\nCc: Shirish Pargaonkar \u003cshirishpargaonkar@gmail.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "a07f1bfabdef56710dd9c9bdc101beec18eaa04a", "tree": "894ce375a19e283eb5f93dab5ed5977ad2079ce5", "parents": [ "5a3589697222ffa1ffedb3f5f1568cd8037e4973" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Fri Mar 25 16:25:57 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:37:06 2011 -0400" }, "message": "cifs: set ra_pages in backing_dev_info\n\ncommit 2b6c26a0a62cc0bab0ad487533d5581d7c293fef upstream.\n\nCommit 522440ed made cifs set backing_dev_info on the mapping attached\nto new inodes. This change caused a fairly significant read performance\nregression, as cifs started doing page-sized reads exclusively.\n\nBy virtue of the fact that they\u0027re allocated as part of cifs_sb_info by\nkzalloc, the ra_pages on cifs BDIs get set to 0, which prevents any\nreadahead. This forces the normal read codepaths to use readpage instead\nof readpages causing a four-fold increase in the number of read calls\nwith the default rsize.\n\nFix it by setting ra_pages in the BDI to the same value as that in the\ndefault_backing_dev_info.\n\nFixes https://bugzilla.kernel.org/show_bug.cgi?id\u003d31662\n\nReported-and-Tested-by: Till \u003ctill2.schaefer@uni-dortmund.de\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "5a3589697222ffa1ffedb3f5f1568cd8037e4973", "tree": "46ee2692c2394774370cb8b9992ea83f7f033b3c", "parents": [ "5e4ee2297b57d38880a7a7d0f6574dd2d7581c7a" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue Apr 05 16:23:47 2011 -0700" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:37:01 2011 -0400" }, "message": "cifs: check for private_data before trying to put it\n\ncommit 7797069305d13252fd66cf722aa8f2cbeb3c95cd upstream.\n\ncifs_close doesn\u0027t check that the filp-\u003eprivate_data is non-NULL before\ntrying to put it. That can cause an oops in certain error conditions\nthat can occur on open or lookup before the private_data is set.\n\nReported-by: Ben Greear \u003cgreearb@candelatech.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "5e4ee2297b57d38880a7a7d0f6574dd2d7581c7a", "tree": "5950497d9788177c7a8d79ef545f37d77a3a0533", "parents": [ "2a3467164465a4d4d49866341a30f5dde9bb74a8" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Mon Mar 14 13:48:08 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 21 19:36:56 2011 -0400" }, "message": "cifs: always do is_path_accessible check in cifs_mount\n\ncommit 70945643722ffeac779d2529a348f99567fa5c33 upstream.\n\nCurrently, we skip doing the is_path_accessible check in cifs_mount if\nthere is no prefixpath. I have a report of at least one server however\nthat allows a TREE_CONNECT to a share that has a DFS referral at its\nroot. The reporter in this case was using a UNC that had no prefixpath,\nso the is_path_accessible check was not triggered and the box later hit\na BUG() because we were chasing a DFS referral on the root dentry for\nthe mount.\n\nThis patch fixes this by removing the check for a zero-length\nprefixpath. That should make the is_path_accessible check be done in\nthis situation and should allow the client to chase the DFS referral at\nmount time instead.\n\nReported-and-Tested-by: Yogesh Sharma \u003cysharma@cymer.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "a506a5b9282e4b944ed314d2b30ce387c44ff9fe", "tree": "3f132b8344f0fbefb1739d4e049fd166426702e3", "parents": [ "c8b265db1e63fdb296ec2721f5c1473c44bc4dc5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 18 10:36:54 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:53 2011 -0700" }, "message": "proc: do proper range check on readdir offset\n\ncommit d8bdc59f215e62098bc5b4256fd9928bf27053a1 upstream.\n\nRather than pass in some random truncated offset to the pid-related\nfunctions, check that the offset is in range up-front.\n\nThis is just cleanup, the previous commit fixed the real problem.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "510edc2527fd3d25b6119d980dddff7213868b19", "tree": "aa2760509c081bcf7a52914f4ef3638f08af7b86", "parents": [ "411999731208babec42c9d17f931a8d87586b05f" ], "author": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Thu Apr 14 15:22:09 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:49 2011 -0700" }, "message": "brk: COMPAT_BRK: fix detection of randomized brk\n\ncommit 4471a675dfc7ca676c165079e91c712b09dc9ce4 upstream.\n\n5520e89 (\"brk: fix min_brk lower bound computation for COMPAT_BRK\")\ntried to get the whole logic of brk randomization for legacy\n(libc5-based) applications finally right.\n\nIt turns out that the way to detect whether brk has actually been\nrandomized in the end or not introduced by that patch still doesn\u0027t work\nfor those binaries, as reported by Geert:\n\n: /sbin/init from my old m68k ramdisk exists prematurely.\n:\n: Before the patch:\n:\n: | brk(0x80005c8e) \u003d 0x80006000\n:\n: After the patch:\n:\n: | brk(0x80005c8e) \u003d 0x80005c8e\n:\n: Old libc5 considers brk() to have failed if the return value is not\n: identical to the requested value.\n\nI don\u0027t like it, but currently see no better option than a bit flag in\ntask_struct to catch the CONFIG_COMPAT_BRK \u0026\u0026 randomize_va_space \u003d\u003d 2\ncase.\n\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\nTested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nReported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.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@suse.de\u003e\n\n" }, { "commit": "0b76f8fcb3f302aaf18c65337a472b32d8a74b67", "tree": "902469d6ad9fa8d3d2e0eaadcdd874ca3df3b2f0", "parents": [ "669479155adaf2c0e21934ef1911c9e583d10f5e" ], "author": { "name": "Tim Chen", "email": "tim.c.chen@linux.intel.com", "time": "Fri Apr 15 11:39:29 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:47 2011 -0700" }, "message": "vfs: Fix absolute RCU path walk failures due to uninitialized seq number\n\ncommit c1530019e311c91d14b24d8e74d233152d806e45 upstream.\n\nDuring RCU walk in path_lookupat and path_openat, the rcu lookup\nfrequently failed if looking up an absolute path, because when root\ndirectory was looked up, seq number was not properly set in nameidata.\n\nWe dropped out of RCU walk in nameidata_drop_rcu due to mismatch in\ndirectory entry\u0027s seq number. We reverted to slow path walk that need\nto take references.\n\nWith the following patch, I saw a 50% increase in an exim mail server\nbenchmark throughput on a 4-socket Nehalem-EX system.\n\nSigned-off-by: Tim Chen \u003ctim.c.chen@linux.intel.com\u003e\nReviewed-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "78caa50ce2c5d8ccf9547978aa74e1ba5de08b00", "tree": "f4e4f83fbd98201fbeb877f7d54e66e3a0aae619", "parents": [ "800f0cbf5b425df719f1905f9eff71b44dc244ab" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Wed Apr 13 10:31:52 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:45 2011 -0700" }, "message": "UBIFS: fix oops when R/O file-system is fsync\u0027ed\n\ncommit 78530bf7f2559b317c04991b52217c1608d5a58d upstream.\n\nThis patch fixes severe UBIFS bug: UBIFS oopses when we \u0027fsync()\u0027 an\nfile on R/O-mounter file-system. We (the UBIFS authors) incorrectly\nthought that VFS would not propagate \u0027fsync()\u0027 down to the file-system\nif it is read-only, but this is not the case.\n\nIt is easy to exploit this bug using the following simple perl script:\n\nuse strict;\nuse File::Sync qw(fsync sync);\n\ndie \"File path is not specified\" if not defined $ARGV[0];\nmy $path \u003d $ARGV[0];\n\nopen FILE, \"\u003c\", \"$path\" or die \"Cannot open $path: $!\";\nfsync(\\*FILE) or die \"cannot fsync $path: $!\";\nclose FILE or die \"Cannot close $path: $!\";\n\nThanks to Reuben Dowle \u003cReuben.Dowle@navico.com\u003e for reporting about this\nissue.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReported-by: Reuben Dowle \u003cReuben.Dowle@navico.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "2fa669669e7646c45037b3d9a2bb9cc9a0c7f928", "tree": "b01ac37ad64af8575d8172e42e2d02fdb895a8a0", "parents": [ "6f517d4325c1c1da7ec91b6cfd58c67a72c3b421" ], "author": { "name": "Bob Liu", "email": "lliubbo@gmail.com", "time": "Thu Apr 14 15:22:20 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:42 2011 -0700" }, "message": "ramfs: fix memleak on no-mmu arch\n\ncommit b836aec53e2bce71de1d5415313380688c851477 upstream.\n\nOn no-mmu arch, there is a memleak during shmem test. The cause of this\nmemleak is ramfs_nommu_expand_for_mapping() added page refcount to 2\nwhich makes iput() can\u0027t free that pages.\n\nThe simple test file is like this:\n\n int main(void)\n {\n\tint i;\n\tkey_t k \u003d ftok(\"/etc\", 42);\n\n\tfor ( i\u003d0; i\u003c100; ++i) {\n\t\tint id \u003d shmget(k, 10000, 0644|IPC_CREAT);\n\t\tif (id \u003d\u003d -1) {\n\t\t\tprintf(\"shmget error\\n\");\n\t\t}\n\t\tif(shmctl(id, IPC_RMID, NULL ) \u003d\u003d -1) {\n\t\t\tprintf(\"shm rm error\\n\");\n\t\t\treturn -1;\n\t\t}\n\t}\n\tprintf(\"run ok...\\n\");\n\treturn 0;\n }\n\nAnd the result:\n\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 17912 42408 0 0\n -/+ buffers: 17912 42408\n root:/\u003e shmem\n run ok...\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 19096 41224 0 0\n -/+ buffers: 19096 41224\n root:/\u003e shmem\n run ok...\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 20296 40024 0 0\n -/+ buffers: 20296 40024\n ...\n\nAfter this patch the test result is:(no memleak anymore)\n\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 16668 43652 0 0\n -/+ buffers: 16668 43652\n root:/\u003e shmem\n run ok...\n root:/\u003e free\n total used free shared buffers\n Mem: 60320 16668 43652 0 0\n -/+ buffers: 16668 43652\n\nSigned-off-by: Bob Liu \u003clliubbo@gmail.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.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@suse.de\u003e\n\n" }, { "commit": "8eb83708d2595c06de19d5a304eaf984ab7d31f0", "tree": "969af2d044ad612386dd7b7837e8e8586bfe37fc", "parents": [ "c2961b693a892ddc79c1548af50e92124f48cd1c" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Tue Mar 29 09:45:21 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:25 2011 -0700" }, "message": "UBIFS: fix assertion warnings\n\ncommit c88ac00c5af70c2a0741da14b22cdcf8507ddd92 upstream.\n\nThis patch fixes UBIFS assertion warnings like:\n\nUBIFS assert failed in ubifs_leb_unmap at 135 (pid 29365)\nPid: 29365, comm: integck Tainted: G I 2.6.37-ubi-2.6+ #34\nCall Trace:\n [\u003cffffffffa047c663\u003e] ubifs_lpt_init+0x95e/0x9ee [ubifs]\n [\u003cffffffffa04623a7\u003e] ubifs_remount_fs+0x2c7/0x762 [ubifs]\n [\u003cffffffff810f066e\u003e] do_remount_sb+0xb6/0x101\n [\u003cffffffff81106ff4\u003e] ? do_mount+0x191/0x78e\n [\u003cffffffff811070bb\u003e] do_mount+0x258/0x78e\n [\u003cffffffff810da1e8\u003e] ? alloc_pages_current+0xa2/0xc5\n [\u003cffffffff81107674\u003e] sys_mount+0x83/0xbd\n [\u003cffffffff81009a12\u003e] system_call_fastpath+0x16/0x1b\n\nThey happen when we re-mount from R/O mode to R/W mode. While\nre-mounting, we write to the media, but we still have the c-\u003ero_mount\nflag set. The fix is very simple - just clear the flag before\nstarting re-mounting R/W.\n\nThese warnings are caused by the following commit:\n2ef13294d29bcfb306e0d360f1b97f37b647b0c0\n\nFor -stable guys: this bug was introduced in 2.6.38, this is materieal\nfor 2.6.38-stable.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "27a2ce012061fc1b12754136e2f64e060e01c904", "tree": "a4b4afbfc9fdb4c686642f51484751848fc5d245", "parents": [ "86d3c0e94b7cc32bd6ee638771fea5bf4360c039" ], "author": { "name": "Vasiliy Kulikov", "email": "segoon@openwall.com", "time": "Fri Feb 04 15:24:19 2011 +0300" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:22 2011 -0700" }, "message": "UBIFS: restrict world-writable debugfs files\n\ncommit 8c559d30b4e59cf6994215ada1fe744928f494bf upstream.\n\nDon\u0027t allow everybody to dump sensitive information about filesystems.\n\nSigned-off-by: Vasiliy Kulikov \u003csegoon@openwall.com\u003e\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "3b76dafa88d04ea91e9b16a4cd7489977d60602b", "tree": "2268306c6fa4b2b14194ebdfeeadf7eeef43b3d4", "parents": [ "591c80b5a9a9331a9b2a6d1b7d9b97d3445107c9" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Sat Apr 02 07:34:30 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:21 2011 -0700" }, "message": "cifs: wrap received signature check in srv_mutex\n\ncommit 157c249114508aa71daa308a426e15d81a4eed00 upstream.\n\nWhile testing my patchset to fix asynchronous writes, I hit a bunch\nof signature problems when testing with signing on. The problem seems\nto be that signature checks on receive can be running at the same\ntime as a process that is sending, or even that multiple receives can\nbe checking signatures at the same time, clobbering the same data\nstructures.\n\nWhile we\u0027re at it, clean up the comments over cifs_calculate_signature\nand add a note that the srv_mutex should be held when calling this\nfunction.\n\nThis patch seems to fix the problems for me, but I\u0027m not clear on\nwhether it\u0027s the best approach. If it is, then this should probably\ngo to stable too.\n\nCc: Shirish Pargaonkar \u003cshirishpargaonkar@gmail.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "591c80b5a9a9331a9b2a6d1b7d9b97d3445107c9", "tree": "ad013ed7abb0b7eadb9e456a90a95c0bccbdf41a", "parents": [ "7cb41edf395b70a6e48162382c5577967ce92bbf" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Fri Mar 25 16:25:57 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:20 2011 -0700" }, "message": "cifs: set ra_pages in backing_dev_info\n\ncommit 2b6c26a0a62cc0bab0ad487533d5581d7c293fef upstream.\n\nCommit 522440ed made cifs set backing_dev_info on the mapping attached\nto new inodes. This change caused a fairly significant read performance\nregression, as cifs started doing page-sized reads exclusively.\n\nBy virtue of the fact that they\u0027re allocated as part of cifs_sb_info by\nkzalloc, the ra_pages on cifs BDIs get set to 0, which prevents any\nreadahead. This forces the normal read codepaths to use readpage instead\nof readpages causing a four-fold increase in the number of read calls\nwith the default rsize.\n\nFix it by setting ra_pages in the BDI to the same value as that in the\ndefault_backing_dev_info.\n\nFixes https://bugzilla.kernel.org/show_bug.cgi?id\u003d31662\n\nReported-and-Tested-by: Till \u003ctill2.schaefer@uni-dortmund.de\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "7cb41edf395b70a6e48162382c5577967ce92bbf", "tree": "36603a6cf5fb361d7cbbdaa6caad4f9f8377c027", "parents": [ "1bd6bcd950ab6bd208a4748ff593d612a21b4284" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue Apr 05 16:23:47 2011 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:20 2011 -0700" }, "message": "cifs: check for private_data before trying to put it\n\ncommit 7797069305d13252fd66cf722aa8f2cbeb3c95cd upstream.\n\ncifs_close doesn\u0027t check that the filp-\u003eprivate_data is non-NULL before\ntrying to put it. That can cause an oops in certain error conditions\nthat can occur on open or lookup before the private_data is set.\n\nReported-by: Ben Greear \u003cgreearb@candelatech.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "1bd6bcd950ab6bd208a4748ff593d612a21b4284", "tree": "170080ec65c91ff75c2c37b55370555e5caa3d27", "parents": [ "9b659653fa4b6fc626ba31f82c7cb57e6bdd6089" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Mon Mar 14 13:48:08 2011 -0400" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Apr 21 14:32:20 2011 -0700" }, "message": "cifs: always do is_path_accessible check in cifs_mount\n\ncommit 70945643722ffeac779d2529a348f99567fa5c33 upstream.\n\nCurrently, we skip doing the is_path_accessible check in cifs_mount if\nthere is no prefixpath. I have a report of at least one server however\nthat allows a TREE_CONNECT to a share that has a DFS referral at its\nroot. The reporter in this case was using a UNC that had no prefixpath,\nso the is_path_accessible check was not triggered and the box later hit\na BUG() because we were chasing a DFS referral on the root dentry for\nthe mount.\n\nThis patch fixes this by removing the check for a zero-length\nprefixpath. That should make the is_path_accessible check be done in\nthis situation and should allow the client to chase the DFS referral at\nmount time instead.\n\nReported-and-Tested-by: Yogesh Sharma \u003cysharma@cymer.com\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Steve French \u003csfrench@us.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n" }, { "commit": "6b0e89f098957148561d732a14566b50281e24d4", "tree": "0f30b4cc4d95d9c5ce2a81d534d74c00e00d3fcb", "parents": [ "27829fed6c02aa689f6ef98256180f67d4df65de" ], "author": { "name": "J. Bruce Fields", "email": "bfields@redhat.com", "time": "Mon Mar 28 15:15:09 2011 +0800" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:08:09 2011 -0400" }, "message": "nfsd4: fix oops on lock failure\n\ncommit 23fcf2ec93fb8573a653408316af599939ff9a8e upstream.\n\nLock stateid\u0027s can have access_bmap 0 if they were only partially\ninitialized (due to a failed lock request); handle that case in\nfree_generic_stateid.\n\n------------[ cut here ]------------\nkernel BUG at fs/nfsd/nfs4state.c:380!\ninvalid opcode: 0000 [#1] SMP\nlast sysfs file: /sys/kernel/mm/ksm/run\nModules linked in: nfs fscache md4 nls_utf8 cifs ip6table_filter ip6_tables ebtable_nat ebtables ipt_MASQUERADE iptable_nat nf_nat bridge stp llc nfsd lockd nfs_acl auth_rpcgss sunrpc ipv6 ppdev parport_pc parport pcnet32 mii pcspkr microcode i2c_piix4 BusLogic floppy [last unloaded: mperf]\n\nPid: 1468, comm: nfsd Not tainted 2.6.38+ #120 VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform\nEIP: 0060:[\u003ce24f180d\u003e] EFLAGS: 00010297 CPU: 0\nEIP is at nfs4_access_to_omode+0x1c/0x29 [nfsd]\nEAX: ffffffff EBX: dd758120 ECX: 00000000 EDX: 00000004\nESI: dd758120 EDI: ddfe657c EBP: dd54dde0 ESP: dd54dde0\n DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068\nProcess nfsd (pid: 1468, ti\u003ddd54c000 task\u003dddc92580 task.ti\u003ddd54c000)\nStack:\n dd54ddf0 e24f19ca 00000000 ddfe6560 dd54de08 e24f1a5d dd758130 deee3a20\n ddfe6560 31270000 dd54df1c e24f52fd 0000000f dd758090 e2505dd0 0be304cf\n dbb51d68 0000000e ddfe657c ddcd8020 dd758130 dd758128 dd7580d8 dd54de68\nCall Trace:\n [\u003ce24f19ca\u003e] free_generic_stateid+0x1c/0x3e [nfsd]\n [\u003ce24f1a5d\u003e] release_lockowner+0x71/0x8a [nfsd]\n [\u003ce24f52fd\u003e] nfsd4_lock+0x617/0x66c [nfsd]\n [\u003ce24e57b6\u003e] ? nfsd_setuser+0x199/0x1bb [nfsd]\n [\u003ce24e056c\u003e] ? nfsd_setuser_and_check_port+0x65/0x81 [nfsd]\n [\u003cc07a0052\u003e] ? _cond_resched+0x8/0x1c\n [\u003cc04ca61f\u003e] ? slab_pre_alloc_hook.clone.33+0x23/0x27\n [\u003cc04cac01\u003e] ? kmem_cache_alloc+0x1a/0xd2\n [\u003cc04835a0\u003e] ? __call_rcu+0xd7/0xdd\n [\u003ce24e0dfb\u003e] ? fh_verify+0x401/0x452 [nfsd]\n [\u003ce24f0b61\u003e] ? nfsd4_encode_operation+0x52/0x117 [nfsd]\n [\u003ce24ea0d7\u003e] ? nfsd4_putfh+0x33/0x3b [nfsd]\n [\u003ce24f4ce6\u003e] ? nfsd4_delegreturn+0xd4/0xd4 [nfsd]\n [\u003ce24ea2c9\u003e] nfsd4_proc_compound+0x1ea/0x33e [nfsd]\n [\u003ce24de6ee\u003e] nfsd_dispatch+0xd1/0x1a5 [nfsd]\n [\u003ce1d6e1c7\u003e] svc_process_common+0x282/0x46f [sunrpc]\n [\u003ce1d6e578\u003e] svc_process+0xdc/0xfa [sunrpc]\n [\u003ce24de0fa\u003e] nfsd+0xd6/0x115 [nfsd]\n [\u003ce24de024\u003e] ? nfsd_shutdown+0x24/0x24 [nfsd]\n [\u003cc0454322\u003e] kthread+0x62/0x67\n [\u003cc04542c0\u003e] ? kthread_worker_fn+0x114/0x114\n [\u003cc07a6ebe\u003e] kernel_thread_helper+0x6/0x10\nCode: eb 05 b8 00 00 27 4f 8d 65 f4 5b 5e 5f 5d c3 83 e0 03 55 83 f8 02 89 e5 74 17 83 f8 03 74 05 48 75 09 eb 09 b8 02 00 00 00 eb 0b \u003c0f\u003e 0b 31 c0 eb 05 b8 01 00 00 00 5d c3 55 89 e5 57 56 89 d6 8d\nEIP: [\u003ce24f180d\u003e] nfs4_access_to_omode+0x1c/0x29 [nfsd] SS:ESP 0068:dd54dde0\n---[ end trace 2b0bf6c6557cb284 ]---\n\nThe trace route is:\n\n -\u003e nfsd4_lock()\n -\u003e if (lock-\u003elk_is_new) {\n -\u003e alloc_init_lock_stateid()\n\n 3739: stp-\u003est_access_bmap \u003d 0;\n\n -\u003eif (status \u0026\u0026 lock-\u003elk_is_new \u0026\u0026 lock_sop)\n -\u003e release_lockowner()\n -\u003e free_generic_stateid()\n -\u003e nfs4_access_bmap_to_omode()\n -\u003e nfs4_access_to_omode()\n\n 380: BUG(); *****\n\nThis problem was introduced by 0997b173609b9229ece28941c118a2a9b278796e.\n\nReported-by: Mi Jinlong \u003cmijinlong@cn.fujitsu.com\u003e\nTested-by: Mi Jinlong \u003cmijinlong@cn.fujitsu.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "27829fed6c02aa689f6ef98256180f67d4df65de", "tree": "cecf18b20987636592478291261ea3aa80fc146c", "parents": [ "31015356ff05612b3f5648d323cea9d743783f15" ], "author": { "name": "J. Bruce Fields", "email": "bfields@redhat.com", "time": "Thu Mar 24 22:51:14 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:08:04 2011 -0400" }, "message": "nfsd: fix auth_domain reference leak on nlm operations\n\ncommit 954032d2527f2fce7355ba70709b5e143d6b686f upstream.\n\nThis was noticed by users who performed more than 2^32 lock operations\nand hence made this counter overflow (eventually leading to\nuse-after-free\u0027s). Setting rq_client to NULL here means that it won\u0027t\nlater get auth_domain_put() when it should be.\n\nAppears to have been introduced in 2.5.42 by \"[PATCH] kNFSd: Move auth\ndomain lookup into svcauth\" which moved most of the rq_client handling\nto common svcauth code, but left behind this one line.\n\nCc: Neil Brown \u003cneilb@suse.de\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "31015356ff05612b3f5648d323cea9d743783f15", "tree": "907da57342dac311681f7c5c529a46982836a575", "parents": [ "5e30c16023a166237224448f67246742427a77ef" ], "author": { "name": "Yongqiang Yang", "email": "xiaoqiangnk@gmail.com", "time": "Mon Apr 04 15:40:24 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:07:59 2011 -0400" }, "message": "ext4: fix credits computing for indirect mapped files\n\ncommit 5b41395fcc0265fc9f193aef9df39ce49d64677c upstream.\n\nWhen writing a contiguous set of blocks, two indirect blocks could be\nneeded depending on how the blocks are aligned, so we need to increase\nthe number of credits needed by one.\n\n[ Also fixed a another bug which could further underestimate the\n number of journal credits needed by 1; the code was using integer\n division instead of DIV_ROUND_UP() -- tytso]\n\nSigned-off-by: Yongqiang Yang \u003cxiaoqiangnk@gmail.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "5e30c16023a166237224448f67246742427a77ef", "tree": "6762c58885b00e6e0940e185dbef7eb5223e19f7", "parents": [ "77f74c355086d3fe62f20dd117b9053eb09db139" ], "author": { "name": "Tao Ma", "email": "boyu.mt@taobao.com", "time": "Mon Apr 04 16:00:49 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:07:54 2011 -0400" }, "message": "ext4: fix a double free in ext4_register_li_request\n\ncommit 46e4690bbd9a4f8d9e7c4f34e34b48f703ad47e0 upstream.\n\nIn ext4_register_li_request, we malloc a ext4_li_request and\ninserts it into ext4_li_info-\u003eli_request_list. In case of any\nerror later, we free it in the end. But if we have some error\nin ext4_run_lazyinit_thread, the whole li_request_list will be\ndropped and freed in it. So we will double free this ext4_li_request.\n\nThis patch just sets elr to NULL after it is inserted to the list\nso that the latter kfree won\u0027t double free it.\n\nSigned-off-by: Tao Ma \u003cboyu.mt@taobao.com\u003e\nReviewed-by: Lukas Czerner \u003clczerner@redhat.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "e8ac4f67e533c9571ae06f8c6c4b543faeaf13a8", "tree": "e5fb1bb20ec9a088cac4131d85633564d2c9ee9d", "parents": [ "3170c7d866308867848b6eeca2fcf15e23c752e9" ], "author": { "name": "Phillip Lougher", "email": "phillip@lougher.demon.co.uk", "time": "Tue Mar 15 22:09:55 2011 +0000" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:07:39 2011 -0400" }, "message": "Squashfs: handle corruption of directory structure\n\ncommit 44cff8a9ee8a974f9e931df910688e7fc1f0b0f9 upstream.\n\nHandle the rare case where a directory metadata block is uncompressed and\ncorrupted, leading to a kernel oops in directory scanning (memcpy).\nNormally corruption is detected at the decompression stage and dealt with\nthen, however, this will not happen if:\n\n- metadata isn\u0027t compressed (users can optionally request no metadata\n compression), or\n- the compressed metadata block was larger than the original, in which\n case the uncompressed version was used, or\n- the data was corrupt after decompression\n\nThis patch fixes this by adding some sanity checks against known maximum\nvalues.\n\nSigned-off-by: Phillip Lougher \u003cphillip@lougher.demon.co.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "3170c7d866308867848b6eeca2fcf15e23c752e9", "tree": "07477fc25231a82824354cb576691b5dea53da69", "parents": [ "27fadaa861246727ad5919863281677a36231dcf" ], "author": { "name": "Phillip Lougher", "email": "phillip@lougher.demon.co.uk", "time": "Tue Mar 22 23:01:26 2011 +0000" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:07:34 2011 -0400" }, "message": "Squashfs: Use vmalloc rather than kmalloc for zlib workspace\n\ncommit 117a91e0f25fd7698e20ac3dfa62086be3dc82a3 upstream.\n\nBugzilla bug 31422 reports occasional \"page allocation failure. order:4\"\nat Squashfs mount time. Fix this by making zlib workspace allocation\nuse vmalloc rather than kmalloc.\n\nReported-by: Mehmet Giritli \u003cmehmet@giritli.eu\u003e\nSigned-off-by: Phillip Lougher \u003cphillip@lougher.demon.co.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "855850d3fd5b25360de8adce3f97e09000db56e0", "tree": "624ea9638e5cb9c4b03d45c17292e0a4f020087c", "parents": [ "80cfc9a5ab40d0629d5f216f3ba006860ec25c6b" ], "author": { "name": "Goldwyn Rodrigues", "email": "rgoldwyn@gmail.com", "time": "Thu Feb 17 09:44:40 2011 -0600" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:05:50 2011 -0400" }, "message": "Treat writes as new when holes span across page boundaries\n\ncommit 272b62c1f0f6f742046e45b50b6fec98860208a0 upstream.\n\nWhen a hole spans across page boundaries, the next write forces\na read of the block. This could end up reading existing garbage\ndata from the disk in ocfs2_map_page_blocks. This leads to\nnon-zero holes. In order to avoid this, mark the writes as new\nwhen the holes span across page boundaries.\n\nSigned-off-by: Goldwyn Rodrigues \u003crgoldwyn@suse.de\u003e\nSigned-off-by: jlbec \u003cjlbec@evilplan.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "47c2bfc9cff176abaf6bace2762d83f10a8bd47e", "tree": "56d0aa49c42466e558d28dfc8107378529e80113", "parents": [ "c91a297be661578c6e8019efac247e5893784e43" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Thu Mar 31 18:36:52 2011 +0200" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:04:41 2011 -0400" }, "message": "quota: Don\u0027t write quota info in dquot_commit()\n\ncommit b03f24567ce7caf2420b8be4c6eb74c191d59a91 upstream.\n\nThere\u0027s no reason to write quota info in dquot_commit(). The writing is a\nrelict from the old days when we didn\u0027t have dquot_acquire() and\ndquot_release() and thus dquot_commit() could have created / removed quota\nstructures from the file. These days dquot_commit() only updates usage counters\n/ limits in quota structure and thus there\u0027s no need to write quota info.\n\nThis also fixes an issue with journaling filesystem which didn\u0027t reserve\nenough space in the transaction for write of quota info (it could have been\ndirty at the time of dquot_commit() because of a race with other operation\nchanging it).\n\nReported-and-tested-by: Lukas Czerner \u003clczerner@redhat.com\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "c91a297be661578c6e8019efac247e5893784e43", "tree": "3885e72c2703f8f21e9a41b556e4b77a3c386e7a", "parents": [ "5fa2bf4c2d4264da28d425d9de13955008dd253b" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Mon Apr 04 17:16:39 2011 +0300" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:04:36 2011 -0400" }, "message": "UBIFS: fix debugging failure in dbg_check_space_info\n\ncommit 7da6443aca9be29c6948dcbd636ad50154d0bc0c upstream.\n\nThis patch fixes a debugging failure with which looks like this:\nUBIFS error (pid 32313): dbg_check_space_info: free space changed from 6019344 to 6022654\n\nThe reason for this failure is described in the comment this patch adds\nto the code. But in short - \u0027c-\u003efreeable_cnt\u0027 may be different before\nand after re-mounting, and this is normal. So the debugging code should\nmake sure that free space calculations do not depend on \u0027c-\u003efreeable_cnt\u0027.\n\nA similar issue has been reported here:\nhttp://lists.infradead.org/pipermail/linux-mtd/2011-April/034647.html\n\nThis patch should fix it.\n\nFor the -stable guys: this patch is only relevant for kernels 2.6.30\nonwards.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "5fa2bf4c2d4264da28d425d9de13955008dd253b", "tree": "d3c3649a1772621d520f4ec92bb365f2b11551c4", "parents": [ "dca70d5b87a077db871cde341262042d3dd8243f" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Fri Mar 25 19:09:54 2011 +0200" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:04:31 2011 -0400" }, "message": "UBIFS: fix oops on error path in read_pnode\n\ncommit 54acbaaa523ca0bd284a18f67ad213c379679e86 upstream.\n\nThanks to coverity which spotted that UBIFS will oops if \u0027kmalloc()\u0027\nin \u0027read_pnode()\u0027 fails and we dereference a NULL \u0027pnode\u0027 pointer\nwhen we \u0027goto out\u0027.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "dca70d5b87a077db871cde341262042d3dd8243f", "tree": "28b91797e350caf32a3f383372f6b11564f58175", "parents": [ "c887414d3bbf1e5adca00f21e0c057edb3e4211f" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Fri Mar 25 18:33:57 2011 +0200" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:04:26 2011 -0400" }, "message": "UBIFS: do not read flash unnecessarily\n\ncommit 8b229c76765816796eec7ccd428f03bd8de8b525 upstream.\n\nThis fix makes the \u0027dbg_check_old_index()\u0027 function return\nimmediately if debugging is disabled, instead of executing\nincorrect \u0027goto out\u0027 which causes UBIFS to:\n\n1. Allocate memory\n2. Read the flash\n\nOn every commit. OK, we do not commit that often, but it is\nstill silly to do unneeded I/O anyway.\n\nCredits to coverity for spotting this silly issue.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "aa6b6bda1fcae89d7b678c4d401a39140871cfba", "tree": "1195c2faddaaa3dc2ba3b2f12bcbacce56d307e5", "parents": [ "6615b496b85cd0a05eb757651edf244192511bc0" ], "author": { "name": "Li Zefan", "email": "lizf@cn.fujitsu.com", "time": "Mon Mar 28 02:01:25 2011 +0000" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:03:42 2011 -0400" }, "message": "Btrfs: Fix uninitialized root flags for subvolumes\n\ncommit 08fe4db170b4193603d9d31f40ebaf652d07ac9c upstream.\n\nroot_item-\u003eflags and root_item-\u003ebyte_limit are not initialized when\na subvolume is created. This bug is not revealed until we added\nreadonly snapshot support - now you mount a btrfs filesystem and you\nmay find the subvolumes in it are readonly.\n\nTo work around this problem, we steal a bit from root_item-\u003einode_item-\u003eflags,\nand use it to indicate if those fields have been properly initialized.\nWhen we read a tree root from disk, we check if the bit is set, and if\nnot we\u0027ll set the flag and initialize the two fields of the root item.\n\nReported-by: Andreas Philipp \u003cphilipp.andreas@gmail.com\u003e\nSigned-off-by: Li Zefan \u003clizf@cn.fujitsu.com\u003e\nTested-by: Andreas Philipp \u003cphilipp.andreas@gmail.com\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "777f709a6cbad9876eb0f10ce6aa63590cc3dadb", "tree": "ba46158104df108e962d4ab5778f729cb8d92a36", "parents": [ "315b809a6e4a3635729cc2db1c1838098b411991" ], "author": { "name": "Eric Paris", "email": "eparis@redhat.com", "time": "Tue Apr 05 17:20:50 2011 -0400" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:03:17 2011 -0400" }, "message": "inotify: fix double free/corruption of stuct user\n\ncommit d0de4dc584ec6aa3b26fffea320a8457827768fc upstream.\n\nOn an error path in inotify_init1 a normal user can trigger a double\nfree of struct user. This is a regression introduced by a2ae4cc9a16e\n(\"inotify: stop kernel memory leak on file creation failure\").\n\nWe fix this by making sure that if a group exists the user reference is\ndropped when the group is cleaned up. We should not explictly drop the\nreference on error and also drop the reference when the group is cleaned\nup.\n\nThe new lifetime rules are that an inotify group lives from\ninotify_new_group to the last fsnotify_put_group. Since the struct user\nand inotify_devs are directly tied to this lifetime they are only\nchanged/updated in those two locations. We get rid of all special\ncasing of struct user or user-\u003einotify_devs.\n\nSigned-off-by: Eric Paris \u003ceparis@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "5d2a4a244e87f6ef9696f136762fa4aa549b8d9e", "tree": "7e777ebb1e148da65ed660a2b5138d53f918f403", "parents": [ "7633c02779ce0f9be5108f19fc2af25a15f08a2e" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Sun Mar 27 22:50:49 2011 +0900" }, "committer": { "name": "Prashant Somashekar", "email": "prashsomash@gmail.com", "time": "Thu Apr 14 18:02:07 2011 -0400" }, "message": "nilfs2: fix data loss in mmap page write for hole blocks\n\ncommit 34094537943113467faee98fe67c8a3d3f9a0a8b upstream.\n\nFrom the result of a function test of mmap, mmap write to shared pages\nturned out to be broken for hole blocks. It doesn\u0027t write out filled\nblocks and the data will be lost after umount. This is due to a bug\nthat the target file is not queued for log writer when filling hole\nblocks.\n\nAlso, nilfs_page_mkwrite function exits normal code path even after\nsuccessfully filled hole blocks due to a change of block_page_mkwrite\nfunction; just after nilfs was merged into the mainline,\nblock_page_mkwrite() started to return VM_FAULT_LOCKED instead of zero\nby the patch \"mm: close page_mkwrite races\" (commit:\nb827e496c893de0c). The current nilfs_page_mkwrite() is not handling\nthis value properly.\n\nThis corrects nilfs_page_mkwrite() and will resolve the data loss\nproblem in mmap write.\n\n[This should be applied to every kernel since 2.6.30 but a fix is\n needed for 2.6.37 and prior kernels]\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: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" } ], "next": "5b74dec635bd4a0838d1efcdd685e0b358a0d7df" }