)]}' { "log": [ { "commit": "a107e5a3a473a2ea62bd5af24e11b84adf1486ff", "tree": "d36c2cb38d8be88d4d75cdebc354aa140aa0e470", "parents": [ "e3e1288e86a07cdeb0aee5860a2dff111c6eff79", "a269029d0e2192046be4c07ed78a45022469ee4c" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Oct 27 23:44:47 2010 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Oct 27 23:44:47 2010 -0400" }, "message": "Merge branch \u0027next\u0027 into upstream-merge\n\nConflicts:\n\tfs/ext4/inode.c\n\tfs/ext4/mballoc.c\n\tinclude/trace/events/ext4.h\n" }, { "commit": "39e3ac2599a5f9aba499b5f8af809108e70a6163", "tree": "16c9d790029c36f217b1689b42869c8739e8ac5e", "parents": [ "58590b06d79f7ce5ab64ff3b6d537180fa50dc84" ], "author": { "name": "Brian King", "email": "brking@linux.vnet.ibm.com", "time": "Wed Oct 27 21:25:12 2010 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Oct 27 21:25:12 2010 -0400" }, "message": "jbd2: Fix I/O hang in jbd2_journal_release_jbd_inode\n\nThis fixes a hang seen in jbd2_journal_release_jbd_inode\non a lot of Power 6 systems running with ext4. When we get\nin the hung state, all I/O to the disk in question gets blocked\nwhere we stay indefinitely. Looking at the task list, I can see\nwe are stuck in jbd2_journal_release_jbd_inode waiting on a\nwake up. I added some debug code to detect this scenario and\ndump additional data if we were stuck in jbd2_journal_release_jbd_inode\nfor longer than 30 minutes. When it hit, I was able to see that\ni_flags was 0, suggesting we missed the wake up.\n\nThis patch changes i_flags to be an unsigned long, uses bit operators\nto access it, and adds barriers around the accesses. Prior to applying\nthis patch, we were regularly hitting this hang on numerous systems\nin our test environment. After applying the patch, the hangs no longer\noccur.\n\nSigned-off-by: Brian King \u003cbrking@linux.vnet.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "a2887097f25cd38cadfc11d10769e2b349fb5eca", "tree": "cd4adcb305365d6ba9acd2c02d4eb9d0125c6f8d", "parents": [ "8abfc6e7a45eb74e51904bbae676fae008b11366", "005a1d15f5a6b2bb4ada80349513effbf22b4588" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Oct 22 17:07:18 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Oct 22 17:07:18 2010 -0700" }, "message": "Merge branch \u0027for-2.6.37/barrier\u0027 of git://git.kernel.dk/linux-2.6-block\n\n* \u0027for-2.6.37/barrier\u0027 of git://git.kernel.dk/linux-2.6-block: (46 commits)\n xen-blkfront: disable barrier/flush write support\n Added blk-lib.c and blk-barrier.c was renamed to blk-flush.c\n block: remove BLKDEV_IFL_WAIT\n aic7xxx_old: removed unused \u0027req\u0027 variable\n block: remove the BH_Eopnotsupp flag\n block: remove the BLKDEV_IFL_BARRIER flag\n block: remove the WRITE_BARRIER flag\n swap: do not send discards as barriers\n fat: do not send discards as barriers\n ext4: do not send discards as barriers\n jbd2: replace barriers with explicit flush / FUA usage\n jbd2: Modify ASYNC_COMMIT code to not rely on queue draining on barrier\n jbd: replace barriers with explicit flush / FUA usage\n nilfs2: replace barriers with explicit flush / FUA usage\n reiserfs: replace barriers with explicit flush / FUA usage\n gfs2: replace barriers with explicit flush / FUA usage\n btrfs: replace barriers with explicit flush / FUA usage\n xfs: replace barriers with explicit flush / FUA usage\n block: pass gfp_mask and flags to sb_issue_discard\n dm: convey that all flushes are processed as empty\n ...\n" }, { "commit": "749ef9f8423054e326f3a246327ed2db4b6d395f", "tree": "388df763e4e731f5d9b4d8dcaedca641521571e3", "parents": [ "6d0aed7a38d06284db2a0e46c0a072b0c1c3299b" ], "author": { "name": "Corrado Zoccolo", "email": "czoccolo@gmail.com", "time": "Mon Sep 20 15:24:50 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Mon Sep 20 15:24:50 2010 +0200" }, "message": "cfq: improve fsync performance for small files\n\nFsync performance for small files achieved by cfq on high-end disks is\nlower than what deadline can achieve, due to idling introduced between\nthe sync write happening in process context and the journal commit.\n\nMoreover, when competing with a sequential reader, a process writing\nsmall files and fsync-ing them is starved.\n\nThis patch fixes the two problems by:\n- marking journal commits as WRITE_SYNC, so that they get the REQ_NOIDLE\n flag set,\n- force all queues that have REQ_NOIDLE requests to be put in the noidle\n tree.\n\nHaving the queue associated to the fsync-ing process and the one associated\n to journal commits in the noidle tree allows:\n- switching between them without idling,\n- fairness vs. competing idling queues, since they will be serviced only\n after the noidle tree expires its slice.\n\nAcked-by: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nReviewed-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\nTested-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\nSigned-off-by: Corrado Zoccolo \u003cczoccolo@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "dd3932eddf428571762596e17b65f5dc92ca361b", "tree": "57cec5ae2f862037f78b7e993323d77955bb6463", "parents": [ "8786fb70ccb36c7cff64680bb80c46d3a09d44db" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Thu Sep 16 20:51:46 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Sep 16 20:52:58 2010 +0200" }, "message": "block: remove BLKDEV_IFL_WAIT\n\nAll the blkdev_issue_* helpers can only sanely be used for synchronous\ncaller. To issue cache flushes or barriers asynchronously the caller needs\nto set up a bio by itself with a completion callback to move the asynchronous\nstate machine ahead. So drop the BLKDEV_IFL_WAIT flag that is always\nspecified when calling blkdev_issue_* and also remove the now unused flags\nargument to blkdev_issue_flush and blkdev_issue_zeroout. For\nblkdev_issue_discard we need to keep it for the secure discard flag, which\ngains a more descriptive name and loses the bitops vs flag confusion.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "9c35575bbe6b1dd4914a5323c8df8b3159edcc75", "tree": "537c521af9d3cfcd1d9031cad99ac349b586fc03", "parents": [ "f73bee49855fe968e87af18e44f2d17e1e675e74" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Wed Aug 18 05:29:17 2010 -0400" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Fri Sep 10 12:35:39 2010 +0200" }, "message": "jbd2: replace barriers with explicit flush / FUA usage\n\nSwitch to the WRITE_FLUSH_FUA flag for journal commits and remove the\nEOPNOTSUPP detection for barriers.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "f73bee49855fe968e87af18e44f2d17e1e675e74", "tree": "b8c4bc3ac8fcc6c2ef066bf641e70b7522392c12", "parents": [ "4524451ef7e88c64a868a8f5a0b49bda73beb2a3" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Aug 18 15:56:56 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Fri Sep 10 12:35:39 2010 +0200" }, "message": "jbd2: Modify ASYNC_COMMIT code to not rely on queue draining on barrier\n\nCurrently JBD2 relies blkdev_issue_flush() draining the queue when ASYNC_COMMIT\nfeature is set. This property is going away so make JBD2 wait for buffers it\nneeds on its own before submitting the cache flush.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "87e99511ea54510ffb60b98001d108794d5037f8", "tree": "359228828515e02302d9b259a066a1f1118eba0a", "parents": [ "dad5eb6daa7eeb63d4fc9d982892c59faa07e797" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Aug 11 17:05:45 2010 +0200" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Aug 18 01:09:00 2010 -0400" }, "message": "kill BH_Ordered flag\n\nInstead of abusing a buffer_head flag just add a variant of\nsync_dirty_buffer which allows passing the exact type of write\nflag required.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "8dd420466c7bfc459fa04680bd5690bfc41a4553", "tree": "6004eaaa7266979397a9d87e30066fc91d3650c3", "parents": [ "a931da6ac9331a6c80dd91c199105806f2336188" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Aug 03 21:38:29 2010 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Aug 03 21:38:29 2010 -0400" }, "message": "jbd2: Remove t_handle_lock from start_this_handle()\n\nThis should remove the last exclusive lock from start_this_handle(),\nso that we should now be able to start multiple transactions at the\nsame time on large SMP systems.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "a931da6ac9331a6c80dd91c199105806f2336188", "tree": "2d4ea766def9a98d21c2379f41e796b0e57ccd6b", "parents": [ "a51dca9cd3bb4ec5a05bfb6feabf024a5c808a37" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Aug 03 21:35:12 2010 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Aug 03 21:35:12 2010 -0400" }, "message": "jbd2: Change j_state_lock to be a rwlock_t\n\nLockstat reports have shown that j_state_lock is a major source of\nlock contention, especially on systems with more than 4 CPU cores. So\nchange it to be a read/write spinlock.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "a51dca9cd3bb4ec5a05bfb6feabf024a5c808a37", "tree": "31256b4fa3ec1d0e1e27196607499877c662ee86", "parents": [ "8b67f04ab9de5d8f3a71aef72bf02c995a506db5" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Aug 02 08:43:25 2010 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Aug 02 08:43:25 2010 -0400" }, "message": "jbd2: Use atomic variables to avoid taking t_handle_lock in jbd2_journal_stop\n\nBy using an atomic_t for t_updates and t_outstanding credits, this\nshould allow us to not need to take transaction t_handle_lock in\njbd2_journal_stop().\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "cc937db74bd5fe75f4cfebbfc4a04c1da10f0695", "tree": "53c314f54e5904fdfe7768adc6f852bdc60040e3", "parents": [ "e3570639c8b5f2c6a5018a2649c2b7c276af76d7" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Jul 27 11:56:08 2010 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Jul 27 11:56:08 2010 -0400" }, "message": "jbd2: Make barrier messages less scary\n\nSaying things like \"sync failed\" when a device does\nnot support barriers makes users slightly more worried than\nthey need to be; rather than talking about sync failures,\nlet\u0027s just state the barrier-based facts.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "fbd9b09a177a481eda256447c881f014f29034fe", "tree": "ef7e213045382f82a1e3e3cf134d196a1045dd7a", "parents": [ "6b4517a7913a09d3259bb1d21c9cb300f12294bd" ], "author": { "name": "Dmitry Monakhov", "email": "dmonakhov@openvz.org", "time": "Wed Apr 28 17:55:06 2010 +0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Apr 28 19:47:36 2010 +0200" }, "message": "blkdev: generalize flags for blkdev_issue_fn functions\n\nThe patch just convert all blkdev_issue_xxx function to common\nset of flags. Wait/allocation semantics preserved.\n\nSigned-off-by: Dmitry Monakhov \u003cdmonakhov@openvz.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "23e2af3518facab6838d7aac1f46fbd7a5a290ce", "tree": "97a45b7d17bd0f4f5c8a65a9583b66886351cf7a", "parents": [ "56c50f11f4d11cb14d78fe52330efb69d219c62f" ], "author": { "name": "dingdinghua", "email": "dingdinghua@nrchpc.ac.cn", "time": "Wed Feb 24 12:11:20 2010 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Feb 24 12:11:20 2010 -0500" }, "message": "jbd2: clean up an assertion in jbd2_journal_commit_transaction()\n\ncommit_transaction has the same value as journal-\u003ej_running_transaction,\nso we can simplify the assert statement.\n\nSigned-off-by: dingdinghua \u003cdingdinghua@nrchpc.ac.cn\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "ba869023eac8354b17acdcff82b851ea8e7b1809", "tree": "c461ae44ed5f9fc45573c14c74548f3f50109547", "parents": [ "aca92ff6f57c000d1b4523e383c8bd6b8269b8b1" ], "author": { "name": "dingdinghua", "email": "dingdinghua@nrchpc.ac.cn", "time": "Mon Feb 15 16:35:42 2010 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Feb 15 16:35:42 2010 -0500" }, "message": "jbd2: delay discarding buffers in journal_unmap_buffer\n\nDelay discarding buffers in journal_unmap_buffer until\nwe know that \"add to orphan\" operation has definitely been\ncommitted, otherwise the log space of committing transation\nmay be freed and reused before truncate get committed, updates\nmay get lost if crash happens.\n\nSigned-off-by: dingdinghua \u003cdingdinghua@nrchpc.ac.cn\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "cc3e1bea5d87635c519da657303690f5538bb4eb", "tree": "727b348d0389a2fe6618fb224fe1d81d207668c4", "parents": [ "034fb4c95fc0fed4ec4a50778127b92c6f2aec01" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Dec 23 06:52:08 2009 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Dec 23 06:52:08 2009 -0500" }, "message": "ext4, jbd2: Add barriers for file systems with exernal journals\n\nThis is a bit complicated because we are trying to optimize when we\nsend barriers to the fs data disk. We could just throw in an extra\nbarrier to the data disk whenever we send a barrier to the journal\ndisk, but that\u0027s not always strictly necessary.\n\nWe only need to send a barrier during a commit when there are data\nblocks which are must be written out due to an inode written in\nordered mode, or if fsync() depends on the commit to force data blocks\nto disk. Finally, before we drop transactions from the beginning of\nthe journal during a checkpoint operation, we need to guarantee that\nany blocks that were flushed out to the data disk are firmly on the\nrust platter before we drop the transaction from the journal.\n\nThanks to Oleg Drokin for pointing out this flaw in ext3/ext4.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "3126c136bc30225d7a43af741778aa50e95e467a", "tree": "71e6f0de6e1f4cde200dd632da4d2f61180289bf", "parents": [ "f4d544ee5720d336a8c64f9fd33efb888c302309", "8e0eb4011bd73d5f91b215b532f74eef478ef795" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Dec 11 15:31:13 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Dec 11 15:31:13 2009 -0800" }, "message": "Merge branch \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6\n\n* \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6: (21 commits)\n ext3: PTR_ERR return of wrong pointer in setup_new_group_blocks()\n ext3: Fix data / filesystem corruption when write fails to copy data\n ext4: Support for 64-bit quota format\n ext3: Support for vfsv1 quota format\n quota: Implement quota format with 64-bit space and inode limits\n quota: Move definition of QFMT_OCFS2 to linux/quota.h\n ext2: fix comment in ext2_find_entry about return values\n ext3: Unify log messages in ext3\n ext2: clear uptodate flag on super block I/O error\n ext2: Unify log messages in ext2\n ext3: make \"norecovery\" an alias for \"noload\"\n ext3: Don\u0027t update the superblock in ext3_statfs()\n ext3: journal all modifications in ext3_xattr_set_handle\n ext2: Explicitly assign values to on-disk enum of filetypes\n quota: Fix WARN_ON in lookup_one_len\n const: struct quota_format_ops\n ubifs: remove manual O_SYNC handling\n afs: remove manual O_SYNC handling\n kill wait_on_page_writeback_range\n vfs: Implement proper O_SYNC semantics\n ...\n" }, { "commit": "94004ed726f38a841cc51f97c4a3f9eda9fbd0d9", "tree": "786689e9fc9e686a5b54c1a1b1efcbb7d043b5ab", "parents": [ "6b2f3d1f769be5779b479c37800229d9a4809fc3" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Sep 30 22:16:33 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Thu Dec 10 15:02:50 2009 +0100" }, "message": "kill wait_on_page_writeback_range\n\nAll callers really want the more logical filemap_fdatawait_range interface,\nso convert them to use it and merge wait_on_page_writeback_range into\nfilemap_fdatawait_range.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "e6ec116b67f46e0e7808276476554727b2e6240b", "tree": "ef80d636a8d45ff7779532e493916e2873adf390", "parents": [ "3f0ca309858ee186435c608ee9eaafd1c8dcb53a" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Dec 01 09:04:42 2009 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Dec 01 09:04:42 2009 -0500" }, "message": "jbd2: Add ENOMEM checking in and for jbd2_journal_write_metadata_buffer()\n\nOOM happens.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "bf6993276f74d46776f35c45ddef29b981b1d1c6", "tree": "5c9cb128fed29a83add1932b12443edaa6fd06cc", "parents": [ "296c355cd6443d89fa251885a8d78778fe111dc4" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Sep 30 00:32:06 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Sep 30 00:32:06 2009 -0400" }, "message": "jbd2: Use tracepoints for history file\n\nThe /proc/fs/jbd2/\u003cdev\u003e/history was maintained manually; by using\ntracepoints, we can get all of the existing functionality of the /proc\nfile plus extra capabilities thanks to the ftrace infrastructure. We\nsave memory as a bonus.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "0e3d2a6313d03413d93327202a60256d1d726fdc", "tree": "3b4165751568e4332edad89a82e76bc6708c8ccd", "parents": [ "71290b368ad5e1e0b0b300c9d5638490a9fd1a2d" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Sep 11 09:30:12 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Sep 11 09:30:12 2009 -0400" }, "message": "ext4: Fix async commit mode to be safe by using a barrier\n\nPreviously the journal_async_commit mount option was equivalent to\nusing barrier\u003d0 (and just as unsafe). This patch fixes it so that we\neliminate the barrier before the commit block (by not using ordered\nmode), and explicitly issuing an empty barrier bio after writing the\ncommit block. Because of the journal checksum, it is safe to do this;\nif the journal blocks are not all written before a power failure, the\nchecksum in the commit block will prevent the last transaction from\nbeing replayed.\n\nUsing the fs_mark benchmark, using journal_async_commit shows a 50%\nimprovement:\n\nFSUse% Count Size Files/sec App Overhead\n 8 1000 10240 30.5 28242\n\nvs.\n\nFSUse% Count Size Files/sec App Overhead\n 8 1000 10240 45.8 28620\n\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "1fe06ad89255c211fe100d7f690d10b161398df8", "tree": "826d32bf081a6729aeaa08fe42422e49a9b86f6f", "parents": [ "2c96ce9f2084c1e04d02883e622f74a537a63aea" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Sep 15 15:10:20 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 16 15:16:18 2009 +0200" }, "message": "writeback: get rid of wbc-\u003efor_writepages\n\nIt\u0027s only set, it\u0027s never checked. Kill it.\n\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "879c5e6b7cb4c689d08ca9b2e353d8ab3dc425d5", "tree": "8b58103d6773a47d6fc29b47024e8b4092487c88", "parents": [ "65795efbd380a832ae508b04dba8f8e53f0b84d9" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Jun 17 11:47:48 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Jun 17 11:47:48 2009 -0400" }, "message": "jbd2: convert instrumentation from markers to tracepoints\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "67c457a8c378a006a34d92f9bd3078a80a92f250", "tree": "63028e792b62fa42724507b86acded4da356c185", "parents": [ "6b82f3cb2d480b7714eb0ff61aee99c22160389e" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Apr 14 07:50:56 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Apr 14 07:50:56 2009 -0400" }, "message": "jbd2: use SWRITE_SYNC_PLUG when writing synchronous revoke records\n\nThe revoke records must be written using the same way as the rest of\nthe blocks during the commit process; that is, either marked as\nsynchronous writes or as asynchornous writes.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "4194b1eaf18ddc3eee9597e8800b6d2ffdfd3614", "tree": "34083152f17c6dbf61729a2194658ebcc041a62c", "parents": [ "6c4bac6b3351fd278dc3537dae42f88f733ff12e" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Apr 06 14:48:05 2009 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 06 08:04:54 2009 -0700" }, "message": "jbd2: use WRITE_SYNC_PLUG instead of WRITE_SYNC\n\nWhen you are going to be submitting several sync writes, we want to\ngive the IO scheduler a chance to merge some of them. Instead of\nusing the implicitly unplugging WRITE_SYNC variant, use WRITE_SYNC_PLUG\nand rely on sync_buffer() doing the unplug when someone does a\nwait_on_buffer()/lock_buffer().\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7058548cd50e5bda8db086bb2e5c1d82f746d047", "tree": "acd095e9f13fcb10deab00e5da5c22d35fda646d", "parents": [ "afd4672dc7610b7feef5190168aa917cc2e417e4" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Mar 25 23:35:46 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Mar 25 23:35:46 2009 -0400" }, "message": "ext4: Use WRITE_SYNC for commits which are caused by fsync()\n\nIf a commit is triggered by fsync(), set a flag indicating the journal\nblocks associated with the transaction should be flushed out using\nWRITE_SYNC.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "2150edc6c5cf00f7adb54538b9ea2a3e9cedca3f", "tree": "f72a0d85e66f500b4cead348a231e3d3b9f357bc", "parents": [ "cd764695b67386a81964f68e9c66efd9f13f4d29", "4b905671d2ea09fd48fed72c581df17e40823f39" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 08 17:14:59 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 08 17:14:59 2009 -0800" }, "message": "Merge branch \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4\n\n* \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (57 commits)\n jbd2: Fix oops in jbd2_journal_init_inode() on corrupted fs\n ext4: Remove \"extents\" mount option\n block: Add Kconfig help which notes that ext4 needs CONFIG_LBD\n ext4: Make printk\u0027s consistently prefixed with \"EXT4-fs: \"\n ext4: Add sanity checks for the superblock before mounting the filesystem\n ext4: Add mount option to set kjournald\u0027s I/O priority\n jbd2: Submit writes to the journal using WRITE_SYNC\n jbd2: Add pid and journal device name to the \"kjournald2 starting\" message\n ext4: Add markers for better debuggability\n ext4: Remove code to create the journal inode\n ext4: provide function to release metadata pages under memory pressure\n ext3: provide function to release metadata pages under memory pressure\n add releasepage hooks to block devices which can be used by file systems\n ext4: Fix s_dirty_blocks_counter if block allocation failed with nodelalloc\n ext4: Init the complete page while building buddy cache\n ext4: Don\u0027t allow new groups to be added during block allocation\n ext4: mark the blocks/inode bitmap beyond end of group as used\n ext4: Use new buffer_head flag to check uninit group bitmaps initialization\n ext4: Fix the race between read_inode_bitmap() and ext4_new_inode()\n ext4: code cleanup\n ...\n" }, { "commit": "e06c8227fd94ec181849ba206bf032be31c4295c", "tree": "67261cd94aa86dc6112d7de74304c2a1af5b64fb", "parents": [ "754938c142ae0c28360426c43f965ddc5164b21e" ], "author": { "name": "Joel Becker", "email": "joel.becker@oracle.com", "time": "Thu Sep 11 15:35:47 2008 -0700" }, "committer": { "name": "Mark Fasheh", "email": "mfasheh@suse.com", "time": "Mon Jan 05 08:40:30 2009 -0800" }, "message": "jbd2: Add buffer triggers\n\nFilesystems often to do compute intensive operation on some\nmetadata. If this operation is repeated many times, it can be very\nexpensive. It would be much nicer if the operation could be performed\nonce before a buffer goes to disk.\n\nThis adds triggers to jbd2 buffer heads. Just before writing a metadata\nbuffer to the journal, jbd2 will optionally call a commit trigger associated\nwith the buffer. If the journal is aborted, an abort trigger will be\ncalled on any dirty buffers as they are dropped from pending\ntransactions.\n\nocfs2 will use this feature.\n\nInitially I tried to come up with a more generic trigger that could be\nused for non-buffer-related events like transaction completion. It\ndoesn\u0027t tie nicely, because the information a buffer trigger needs\n(specific to a journal_head) isn\u0027t the same as what a transaction\ntrigger needs (specific to a tranaction_t or perhaps journal_t). So I\nimplemented a buffer set, with the understanding that\njournal/transaction wide triggers should be implemented separately.\n\nThere is only one trigger set allowed per buffer. I can\u0027t think of any\nreason to attach more than one set. Contrast this with a journal or\ntransaction in which multiple places may want to watch the entire\ntransaction separately.\n\nThe trigger sets are considered static allocation from the jbd2\nperspective. ocfs2 will just have one trigger set per block type,\nsetting the same set on every bh of the same type.\n\nSigned-off-by: Joel Becker \u003cjoel.becker@oracle.com\u003e\nCc: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nSigned-off-by: Mark Fasheh \u003cmfasheh@suse.com\u003e\n" }, { "commit": "40a1984d22294ab202f616e432bb8d3481897675", "tree": "42d9970637c0c26b723b51115a6a290d8761eb8e", "parents": [ "4a9bf99b205448ec1f0cbdee1776a29f9c503ce4" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sun Jan 04 19:55:57 2009 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sun Jan 04 19:55:57 2009 -0500" }, "message": "jbd2: Submit writes to the journal using WRITE_SYNC\n\nSince we will be waiting the write of the commit record to the journal\nto complete in journal_submit_commit_record(), submit it using\nWRITE_SYNC.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "fb68407b0d9efba962c03f55009c797e22f024bc", "tree": "b3f33f513969d6f77c5b6a6f71b662057d3440a5", "parents": [ "c3a326a657562dab81acf05aee106dc1fe345eb4" ], "author": { "name": "Aneesh Kumar K.V", "email": "aneesh.kumar@linux.vnet.ibm.com", "time": "Thu Nov 06 17:50:21 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Nov 06 17:50:21 2008 -0500" }, "message": "jbd2: Call journal commit callback without holding j_list_lock\n\nAvoid freeing the transaction in __jbd2_journal_drop_transaction() so\nthe journal commit callback can run without holding j_list_lock, to\navoid lock contention on this spinlock.\n\nSigned-off-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.vnet.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "e07f7183a486cf9783d1f8c9d2997b5b39eeb2d4", "tree": "74ed3a563add5fa57e80af03f3f712f2910ac39f", "parents": [ "032115fcef837a00336ddf7bda584e89789ea498" ], "author": { "name": "Josef Bacik", "email": "jbacik@redhat.com", "time": "Wed Nov 26 01:14:26 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Nov 26 01:14:26 2008 -0500" }, "message": "jbd2: improve jbd2 fsync batching\n\nThis patch removes the static sleep time in favor of a more self\noptimizing approach where we measure the average amount of time it\ntakes to commit a transaction to disk and the ammount of time a\ntransaction has been running. If somebody does a sync write or an\nfsync() traditionally we would sleep for 1 jiffies, which depending on\nthe value of HZ could be a significant amount of time compared to how\nlong it takes to commit a transaction to the underlying storage. With\nthis patch instead of sleeping for a jiffie, we check to see if the\namount of time this transaction has been running is less than the\naverage commit time, and if it is we sleep for the delta using\nschedule_hrtimeout to give us a higher precision sleep time. This\ngreatly benefits high end storage where you could end up sleeping for\nlonger than it takes to commit the transaction and therefore sitting\nidle instead of allowing the transaction to be committed by keeping\nthe sleep time to a minimum so you are sure to always be doing\nsomething.\n\nSigned-off-by: Josef Bacik \u003cjbacik@redhat.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "fd98496f467b3d26d05ab1498f41718b5ef13de5", "tree": "27e29934f5c2ee5ccc060de27dee261d31d6842c", "parents": [ "0390131ba84fd3f726f9e24fc4553828125700bb" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Jan 05 21:34:13 2009 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Jan 05 21:34:13 2009 -0500" }, "message": "jbd2: Add barrier not supported test to journal_wait_on_commit_record\n\nXen doesn\u0027t report that barriers are not supported until buffer I/O is\nreported as completed, instead of when the buffer I/O is submitted.\nAdd a check and a fallback codepath to journal_wait_on_commit_record()\nto detect this case, so that attempts to mount ext4 filesystems on\nLVM/devicemapper devices on Xen guests don\u0027t blow up with an \"Aborting\njournal on device XXX\"; \"Remounting filesystem read-only\" error.\n\nThanks to Andreas Sundstrom for reporting this issue.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: stable@kernel.org\n" }, { "commit": "6c20ec850360bc6e5c66a787f0523a80450d65ab", "tree": "567f496cf53dbfbc235809edac134e0247a062c8", "parents": [ "ef2cabf7c6d838eb0ee2b4fb8ef84f7c06ce16d9" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Oct 28 21:08:20 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Oct 28 21:08:20 2008 -0400" }, "message": "jbd2: Call the commit callback before the transaction could get dropped\n\nThe transaction can potentially get dropped if there are no buffers\nthat need to be written. Make sure we call the commit callback before\npotentially deciding to drop the transaction. Also avoid\ndereferencing the commit_transaction pointer in the marker for the\nsame reason.\n\nThis patch fixes the bug reported by Eric Paris at:\nhttp://bugzilla.kernel.org/show_bug.cgi?id\u003d11838\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nAcked-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nTested-by: Eric Paris \u003ceparis@redhat.com\u003e\n" }, { "commit": "3e624fc72fba09b6f999a9fbb87b64efccd38036", "tree": "f0a1feac377788b27ea704519a33f367a6d415f9", "parents": [ "22359f5745eb26bd3205a1ede7968c8944398220" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Oct 16 20:00:24 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Oct 16 20:00:24 2008 -0400" }, "message": "ext4: Replace hackish ext4_mb_poll_new_transaction with commit callback\n\nThe multiblock allocator needs to be able to release blocks (and issue\na blkdev discard request) when the transaction which freed those\nblocks is committed. Previously this was done via a polling mechanism\nwhen blocks are allocated or freed. A much better way of doing things\nis to create a jbd2 callback function and attaching the list of blocks\nto be freed directly to the transaction structure.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "5bf5683a33f3584da6eced480967c4f7e11515a8", "tree": "4986ba3932d9f259fa1bc674deff3e35f689b243", "parents": [ "7ad7445f60fe4d46c4c9d2a9463db180d2a3b270" ], "author": { "name": "Hidehiro Kawai", "email": "hidehiro.kawai.ez@hitachi.com", "time": "Fri Oct 10 22:12:43 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Oct 10 22:12:43 2008 -0400" }, "message": "ext4: add an option to control error handling on file data\n\nIf the journal doesn\u0027t abort when it gets an IO error in file data\nblocks, the file data corruption will spread silently. Because\nmost of applications and commands do buffered writes without fsync(),\nthey don\u0027t notice the IO error. It\u0027s scary for mission critical\nsystems. On the other hand, if the journal aborts whenever it gets\nan IO error in file data blocks, the system will easily become\ninoperable. So this patch introduces a filesystem option to\ndetermine whether it aborts the journal or just call printk() when\nit gets an IO error in file data.\n\nIf you mount an ext4 fs with data_err\u003dabort option, it aborts on file\ndata write error. If you mount it with data_err\u003dignore, it doesn\u0027t\nabort, just call printk(). data_err\u003dignore is the default.\n\nHere is the corresponding patch of the ext3 version:\nhttp://kerneltrap.org/mailarchive/linux-kernel/2008/9/9/3239374\n\nSigned-off-by: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "7ad7445f60fe4d46c4c9d2a9463db180d2a3b270", "tree": "cd4e0f69f90664e5ca4185271fbacbb7b4f59491", "parents": [ "7ffe1ea8949c75ecffb7a4d988bb881a9fa62fbe" ], "author": { "name": "Hidehiro Kawai", "email": "hidehiro.kawai.ez@hitachi.com", "time": "Fri Oct 10 20:29:31 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Oct 10 20:29:31 2008 -0400" }, "message": "jbd2: don\u0027t dirty original metadata buffer on abort\n\nCurrently, original metadata buffers are dirtied when they are\nunfiled whether the journal has aborted or not. Eventually these\nbuffers will be written-back to the filesystem by pdflush. This\nmeans some metadata buffers are written to the filesystem without\njournaling if the journal aborts. So if both journal abort and\nsystem crash happen at the same time, the filesystem would become\ninconsistent state. Additionally, replaying journaled metadata\ncan overwrite the latest metadata on the filesystem partly.\nBecause, if the journal gets aborted, journaled metadata are\npreserved and replayed during the next mount not to lose\nuncheckpointed metadata. This would also break the consistency\nof the filesystem.\n\nThis patch prevents original metadata buffers from being dirtied\non abort by clearing BH_JBDDirty flag from those buffers. Thus,\nno metadata buffers are written to the filesystem without journaling.\n\nSigned-off-by: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "77e841de8abac4755cc83ca224fdf71418d65380", "tree": "937694b6bfb13b2ef6e154e8ab25f5f7200159fb", "parents": [ "fd048088306656824958e7783ffcee27e241b361" ], "author": { "name": "Hidehiro Kawai", "email": "hidehiro.kawai.ez@hitachi.com", "time": "Sun Oct 12 16:39:16 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sun Oct 12 16:39:16 2008 -0400" }, "message": "jbd2: abort when failed to log metadata buffers\n\nIf we failed to write metadata buffers to the journal space and\nsucceeded to write the commit record, stale data can be written\nback to the filesystem as metadata in the recovery phase.\n\nTo avoid this, when we failed to write out metadata buffers,\nabort the journal before writing the commit record.\n\nWe can also avoid this kind of corruption by using the journal\nchecksum feature because it can detect invalid metadata blocks in the\njournal and avoid them from being replayed. So we don\u0027t need to care\nabout asynchronous commit record writeout with a checksum.\n\nSigned-off-by: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "45a90bfd90c1215bf824c0f705b409723f52361b", "tree": "e314c36e14c380b87d19e3054494768f4c14e458", "parents": [ "ede86cc473defab74d778aeac14b19f43129d4d1" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Oct 06 12:04:02 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Oct 06 12:04:02 2008 -0400" }, "message": "jbd2: Fix buffer head leak when writing the commit block\n\nAlso make sure the buffer heads are marked clean before submitting bh\nfor writing. The previous code was marking the buffer head dirty,\nwhich would have forced an unneeded write (and seek) to the journal\nfor no good reason.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "ede86cc473defab74d778aeac14b19f43129d4d1", "tree": "19da8009accf1737171f5949413e46a9086d5fdc", "parents": [ "23f8b79eae8a74e42a006ffa7c456e295c7e1c0d" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sun Oct 05 20:50:06 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sun Oct 05 20:50:06 2008 -0400" }, "message": "ext4: Add debugging markers that can be used by systemtap\n\nThis debugging markers are designed to debug problems such as the\nrandom filesystem latency problems reported by Arjan.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "05496769e5da83ce22ed97345afd9c7b71d6bd24", "tree": "545c4d66296fe69ab24d34fd2f0298fb8ee66e4f", "parents": [ "899fc1a4cf404747de2666534d508804597ee22f" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Sep 16 14:36:17 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Sep 16 14:36:17 2008 -0400" }, "message": "jbd2: clean up how the journal device name is printed\n\nCalculate the journal device name once and stash it away in the\njournal_s structure. This avoids needing to call bdevname()\neverywhere and reduces stack usage by not needing to allocate an\non-stack buffer. In addition, we eliminate the \u0027/\u0027 that can appear in\ndevice names (e.g. \"cciss/c0d0p9\" --- see kernel bugzilla #11321) that\ncan cause problems when creating proc directory names, and include the\ninode number to support ocfs2 which creates multiple journals with\ndifferent inode numbers.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "529ae9aaa08378cfe2a4350bded76f32cc8ff0ce", "tree": "d3ae998f9876c72a83a022805103a92111852b21", "parents": [ "e9ba9698187ddbc0c5bfcf41de0349a662d23d02" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Sat Aug 02 12:01:03 2008 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Aug 04 21:31:34 2008 -0700" }, "message": "mm: rename page trylock\n\nConverting page lock to new locking bitops requires a change of page flag\noperation naming, so we might as well convert it to something nicer\n(!TestSetPageLocked_Lock \u003d\u003e trylock_page, SetPageLocked \u003d\u003e set_page_locked).\n\nThis also facilitates lockdeping of page lock.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nAcked-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nAcked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nAcked-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nAcked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e9e34f4e8f42177c66754fec1edfd35e70c18f99", "tree": "659e21aa3ea54c7643c0e3112aa3aba54ca7b4cf", "parents": [ "9c83a923c67df311c467ec956009f0eb4019195d" ], "author": { "name": "Hidehiro Kawai", "email": "hidehiro.kawai.ez@hitachi.com", "time": "Thu Jul 31 22:26:04 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Jul 31 22:26:04 2008 -0400" }, "message": "jbd2: don\u0027t abort if flushing file data failed\n\nIn ordered mode, the current jbd2 aborts the journal if a file data buffer\nhas an error. But this behavior is unintended, and we found that it has\nbeen adopted accidentally.\n\nThis patch undoes it and just calls printk() instead of aborting the\njournal. Unlike a similar patch for ext3/jbd, file data buffers are\nwritten via generic_writepages(). But we also need to set AS_EIO\ninto their mappings because wait_on_page_writeback_range() clears\nAS_EIO before a user process sees it.\n\nSigned-off-by: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "cd1aac32923a9c8adcc0ae85e33c1ca0c5855838", "tree": "3d55d9249ef960a7e345969404d537e36dbd9609", "parents": [ "61628a3f3a37af2bf25daf8e26fd6b76a78c4f76" ], "author": { "name": "Aneesh Kumar K.V", "email": "aneesh.kumar@linux.vnet.ibm.com", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "message": "ext4: Add ordered mode support for delalloc\n\nThis provides a new ordered mode implementation which gets rid of using\nbuffer heads to enforce the ordering between metadata change with the\nrelated data chage. Instead, in the new ordering mode, it keeps track\nof all of the inodes touched by each transaction on a list, and when\nthat transaction is committed, it flushes all of the dirty pages for\nthose inodes. In addition, the new ordered mode reverses the lock\nordering of the page lock and transaction lock, which provides easier\nsupport for delayed allocation.\n\nSigned-off-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.vnet.ibm.com\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "87c89c232c8f7b3820c33c3b9bc803e9358027da", "tree": "9a714242513ec3e5e1c28fad1bfff852efd033d5", "parents": [ "678aaf481496b01473b778685eca231d6784098b" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "message": "jbd2: Remove data\u003dordered mode support using jbd buffer heads\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "c851ed540173736e60d48b53b91a16ea5c903896", "tree": "828fe0d71b7f18dc170090dbb2fb5ac9deae4ee0", "parents": [ "f4c0a0fdfae708f7aa438c27a380ed4071294e11" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "message": "jbd2: Implement data\u003dordered mode handling via inodes\n\nThis patch adds necessary framework into JBD2 to be able to track inodes\nwith each transaction and write-out their dirty data during transaction\ncommit time.\n\nThis new ordered mode brings all sorts of advantages such as possibility \nto get rid of journal heads and buffer heads for data buffers in ordered \nmode, better ordering of writes on transaction commit, simplification of\n some JBD code, no more anonymous pages when truncate of data being \ncommitted happens. Also with this new ordered mode, delayed allocation \non ordered mode is much simpler.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "736603ab297506f4396cb5af592004499950fcfd", "tree": "80d1f84702e1bfaca5d60482540b8294acd30bd3", "parents": [ "4db9c54a53135b7c1c1f403f1aeaf9fc0d7738b8" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Jul 11 19:27:31 2008 -0400" }, "message": "jbd2: Add commit time into the commit block\n\nCarlo Wood has demonstrated that it\u0027s possible to recover deleted\nfiles from the journal. Something that will make this easier is if we\ncan put the time of the commit into commit block.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "034772b068be62a79470d6c1b81b01fbe27793ac", "tree": "408b55ae9241d1dca1145e16ff781b5f0112dd35", "parents": [ "cd0b6a39a1d68b61b1073662f40f747c8b728f98" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Jun 03 22:31:11 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Jun 03 22:31:11 2008 -0400" }, "message": "jbd2: Fix barrier fallback code to re-lock the buffer head\n\nIf the device doesn\u0027t support write barriers, the write is retried\nwithout ordered mode. But the buffer head needs to be re-locked or\nsubmit_bh will fail with on BUG(!buffer_locked(bh)).\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "02c471cb17203c748e9bc87003052c1f46e5df69", "tree": "a9d389f614d8d9d230920f6cf258f4d640a44662", "parents": [ "519deca0496a4df07d15acf3181ca5d573bffdec" ], "author": { "name": "Mingming Cao", "email": "cmm@us.ibm.com", "time": "Thu May 15 14:46:17 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu May 15 14:46:17 2008 -0400" }, "message": "jbd2: update transaction t_state to T_COMMIT fix\n\nUpdating the current transaction\u0027s t_state is protected by j_state_lock. We\nneed to do the same when updating the t_state to T_COMMIT.\n\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "1dfc3220d963385a317264b11154c462a83596ed", "tree": "8c35f3233c3d10110b3b48d8f476ddf9404d7192", "parents": [ "9fc7c63a1d6e9920038ced782390a54888ed70a6" ], "author": { "name": "Josef Bacik", "email": "jbacik@redhat.com", "time": "Thu Apr 17 10:38:59 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Apr 17 10:38:59 2008 -0400" }, "message": "jbd2: fix possible journal overflow issues\n\nThere are several cases where the running transaction can get buffers\nadded to its BJ_Metadata list which it never dirtied, which makes its\nt_nr_buffers counter end up larger than its t_outstanding_credits\ncounter.\n\nThis will cause issues when starting new transactions as while we are\nlogging buffers we decrement t_outstanding_buffers, so when\nt_outstanding_buffers goes negative, we will report that we need less\nspace in the journal than we actually need, so transactions will be\nstarted even though there may not be enough room for them. In the worst\ncase scenario (which admittedly is almost impossible to reproduce) this\nwill result in the journal running out of space.\n\nThe fix is to only refile buffers from the committing transaction to the\nrunning transactions BJ_Modified list when b_modified is set on that\njournal, which is the only way to be sure if the running transaction has\nmodified that buffer.\n\nThis patch also fixes an accounting error in journal_forget, it is\npossible that we can call journal_forget on a buffer without having\nmodified it, only gotten write access to it, so instead of freeing a\ncredit, we only do so if the buffer was modified. The assert will help\ncatch if this problem occurs. Without these two patches I could hit\nthis assert within minutes of running postmark, with them this issue no\nlonger arises.\n\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nCc: Jan Kara \u003cjack@ucw.cz\u003e\nSigned-off-by: Josef Bacik \u003cjbacik@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "9fc7c63a1d6e9920038ced782390a54888ed70a6", "tree": "d378eb9089fe1618e180a4bfe38f3fc6d4d1a6c5", "parents": [ "33ae0cdd3eaba219e7c2f0647b6db4be540e2130" ], "author": { "name": "Josef Bacik", "email": "jbacik@redhat.com", "time": "Thu Apr 17 10:38:59 2008 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Apr 17 10:38:59 2008 -0400" }, "message": "jbd2: fix the way the b_modified flag is cleared\n\nCurrently at the start of a journal commit we loop through all of the buffers\non the committing transaction and clear the b_modified flag (the flag that is\nset when a transaction modifies the buffer) under the j_list_lock.\n\nThe problem is that everywhere else this flag is modified only under the jbd2\nlock buffer flag, so it will race with a running transaction who could\npotentially set it, and have it unset by the committing transaction.\n\nThis is also a big waste, you can have several thousands of buffers that you\nare clearing the modified flag on when you may not need to. This patch\nremoves this code and instead clears the b_modified flag upon entering\ndo_get_write_access/journal_get_create_access, so if that transaction does\nindeed use the buffer then it will be accounted for properly, and if it does\nnot then we know we didn\u0027t use it.\n\nThat will be important for the next patch in this series. Tested thoroughly\nby myself using postmark/iozone/bonnie++.\n\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nCc: Jan Kara \u003cjack@ucw.cz\u003e\nSigned-off-by: Josef Bacik \u003cjbacik@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "c4e35e07af162ea4d642b1c6ffacbb63c3ed1804", "tree": "c1f74ee04e54f705d0b18c947a2efa9eea8344a2", "parents": [ "7fb5409df092589b86cc9412d926879cb572b7f0" ], "author": { "name": "Dave Kleikamp", "email": "shaggy@linux.vnet.ibm.com", "time": "Sun Feb 10 01:09:32 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sun Feb 10 01:09:32 2008 -0500" }, "message": "JBD2: Clear buffer_ordered flag for barried IO request on success\n\nIn JBD2 jbd2_journal_write_commit_record(), clear the buffer_ordered\nflag for the bh after barried IO has succeed. This prevents later, if\nthe same buffer head were submitted to the underlying device, which has\nbeen reconfigured to not support barrier request, the JBD2 commit code\ncould treat it as a normal IO (without barrier).\n\nThis is a port from JBD/ext3 fix from Neil Brown.\n\nMore details from Neil:\n\nSome devices - notably dm and md - can change their behaviour in\nresponse to BIO_RW_BARRIER requests. They might start out accepting\nsuch requests but on reconfiguration, they find out that they cannot\nany more. JBD2 deal with this by always testing if BIO_RW_BARRIER\nrequests fail with EOPNOTSUPP, and retrying the write\nrequests without the barrier (probably after waiting for any pending\nwrites to complete).\n\nHowever there is a bug in the handling this in JBD2 for ext4 .\n\nWhen ext4/JBD2 to submit a BIO_RW_BARRIER request,\nit sets the buffer_ordered flag on the buffer head.\nIf the request completes successfully, the flag STAYS SET.\n\nOther code might then write the same buffer_head after the device has\nbeen reconfigured to not accept barriers. This write will then fail,\nbut the \"other code\" is not ready to handle EOPNOTSUPP errors and the\nerror will be treated as fatal.\n\nCc: Neil Brown \u003cneilb@suse.de\u003e\nSigned-off-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "4d605179723a3fb8ba594d9516897426e6629a5b", "tree": "42f7b1a77dcb9a7c348a3ebde0c49f9e8878a707", "parents": [ "c4b8e635f525441b9cb0bab428b527858d977e8f" ], "author": { "name": "Aneesh Kumar K.V", "email": "aneesh.kumar@linux.vnet.ibm.com", "time": "Tue Feb 05 10:56:15 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Feb 05 10:56:15 2008 -0500" }, "message": "JBD2: Use the incompat macro for testing the incompat feature.\n\nJBD2_FEATURE_INCOMPAT_ASYNC_COMMIT needs to be checked with\nJBD2_HAS_INCOMPAT_FEATURE\n\nSigned-off-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.vnet.ibm.com\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "c4b8e635f525441b9cb0bab428b527858d977e8f", "tree": "3ebf433583c1a456d251fdcffd524fdc2f5518d3", "parents": [ "5315217efea54a07950758005686adedb8e8e680" ], "author": { "name": "Aneesh Kumar K.V", "email": "aneesh.kumar@linux.vnet.ibm.com", "time": "Tue Feb 05 10:55:26 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Feb 05 10:55:26 2008 -0500" }, "message": "jbd2: Fix reference counting on the journal commit block\u0027s buffer head\n\nWith journal checksum patch we added asynchronous commits of journal\ncommit headers, and accidentally dropped taking a reference on the\nbuffer head.\n\n(Before the change, sync_dirty_buffer did the get_bh(). The associative\nput_bh is done by journal_wait_on_commit_record().)\n\nSigned-off-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.vnet.ibm.com\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "b048d8462652159c5314d19b191220b0ec384edb", "tree": "06198272d89d7263d4d13df0ed00c3d0cf4e8aa0", "parents": [ "531021f2ca681faf58f926771f85bb5c76f13eba" ], "author": { "name": "Mingming Cao", "email": "cmm@u.ibm.com", "time": "Tue Feb 05 08:52:45 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Feb 05 08:52:45 2008 -0500" }, "message": "jbd2: Add error check to journal_wait_on_commit_record to avoid oops\n\nThe buffer head pointer passed to journal_wait_on_commit_record() could\nbe NULL if the previous journal_submit_commit_record() failed or journal\nhas already aborted.\n\nLooking at the jbd2 debug messages, before the oops happened, the jbd2\nis aborted due to trying to access the next log block beyond the end\nof device. This might be caused by using a corrupted image.\n\nWe need to check the error returns from journal_submit_commit_record()\nand avoid calling journal_wait_on_commit_record() in the failure case.\n\nThis addresses Kernel Bugzilla #9849\n\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "95c354fe9f7d6decc08a92aa26eb233ecc2155bf", "tree": "ec9267032ea875e84216cfb20acb2cfc7c62149f", "parents": [ "a95d67f87e1a5f1b4429be3ba3bf7b4051657908" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Wed Jan 30 13:31:20 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Jan 30 13:31:20 2008 +0100" }, "message": "spinlock: lockbreak cleanup\n\nThe break_lock data structure and code for spinlocks is quite nasty.\nNot only does it double the size of a spinlock but it changes locking to\na potentially less optimal trylock.\n\nPut all of that under CONFIG_GENERIC_LOCKBREAK, and introduce a\n__raw_spin_is_contended that uses the lock data itself to determine whether\nthere are waiters on the lock, to be used if CONFIG_GENERIC_LOCKBREAK is\nnot set.\n\nRename need_lockbreak to spin_needbreak, make it use spin_is_contended to\ndecouple it from the spinlock implementation, and make it typesafe (rwlocks\ndo not have any need_lockbreak sites -- why do they even get bloated up\nwith that break_lock then?).\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "818d276ceb83aa9fdebb5e0a53188290312de987", "tree": "de3fb4ffadd72caea2876c5232ce76cd14b3646e", "parents": [ "8e85fb3f305b24b79c6d9cb7a56d22b062335ad3" ], "author": { "name": "Girish Shilamkar", "email": "girish@clusterfs.com", "time": "Mon Jan 28 23:58:27 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Jan 28 23:58:27 2008 -0500" }, "message": "ext4: Add the journal checksum feature\n\nThe journal checksum feature adds two new flags i.e\nJBD2_FEATURE_INCOMPAT_ASYNC_COMMIT and JBD2_FEATURE_COMPAT_CHECKSUM.\n\nJBD2_FEATURE_CHECKSUM flag indicates that the commit block contains the\nchecksum for the blocks described by the descriptor blocks.\nDue to checksums, writing of the commit record no longer needs to be\nsynchronous. Now commit record can be sent to disk without waiting for\ndescriptor blocks to be written to disk. This behavior is controlled\nusing JBD2_FEATURE_ASYNC_COMMIT flag. Older kernels/e2fsck should not be\nable to recover the journal with _ASYNC_COMMIT hence it is made\nincompat.\nThe commit header has been extended to hold the checksum along with the\ntype of the checksum.\n\nFor recovery in pass scan checksums are verified to ensure the sanity\nand completeness(in case of _ASYNC_COMMIT) of every transaction.\n\nSigned-off-by: Andreas Dilger \u003cadilger@clusterfs.com\u003e\nSigned-off-by: Girish Shilamkar \u003cgirish@clusterfs.com\u003e\nSigned-off-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\n" }, { "commit": "8e85fb3f305b24b79c6d9cb7a56d22b062335ad3", "tree": "9b037c1b1649bce1338911fd8ca3980229c1548a", "parents": [ "4df3d265bf8f3762e1d77f554ee279c39dedb020" ], "author": { "name": "Johann Lombardi", "email": "johann.lombardi@bull.net", "time": "Mon Jan 28 23:58:27 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Jan 28 23:58:27 2008 -0500" }, "message": "jbd2: jbd2 stats through procfs\n\nThe patch below updates the jbd stats patch to 2.6.20/jbd2.\nThe initial patch was posted by Alex Tomas in December 2005\n(http://marc.info/?l\u003dlinux-ext4\u0026m\u003d113538565128617\u0026w\u003d2).\nIt provides statistics via procfs such as transaction lifetime and size.\n\nSometimes, investigating performance problems, i find useful to have\nstats from jbd about transaction\u0027s lifetime, size, etc. here is a\npatch for review and inclusion probably.\n\nfor example, stats after creation of 3M files in htree directory:\n\n[root@bob ~]# cat /proc/fs/jbd/sda/history\nR/C tid wait run lock flush log hndls block inlog ctime write drop close\nR 261 8260 2720 0 0 750 9892 8170 8187\nC 259 750 0 4885 1\nR 262 20 2200 10 0 770 9836 8170 8187\nR 263 30 2200 10 0 3070 9812 8170 8187\nR 264 0 5000 10 0 1340 0 0 0\nC 261 8240 3212 4957 0\nR 265 8260 1470 0 0 4640 9854 8170 8187\nR 266 0 5000 10 0 1460 0 0 0\nC 262 8210 2989 4868 0\nR 267 8230 1490 10 0 4440 9875 8171 8188\nR 268 0 5000 10 0 1260 0 0 0\nC 263 7710 2937 4908 0\nR 269 7730 1470 10 0 3330 9841 8170 8187\nR 270 0 5000 10 0 830 0 0 0\nC 265 8140 3234 4898 0\nC 267 720 0 4849 1\nR 271 8630 2740 20 0 740 9819 8170 8187\nC 269 800 0 4214 1\nR 272 40 2170 10 0 830 9716 8170 8187\nR 273 40 2280 0 0 3530 9799 8170 8187\nR 274 0 5000 10 0 990 0 0 0\n\n\nwhere,\n\nR - line for transaction\u0027s life from T_RUNNING to T_FINISHED\nC - line for transaction\u0027s checkpointing\ntid - transaction\u0027s id\nwait - for how long we were waiting for new transaction to start\n (the longest period journal_start() took in this transaction)\nrun - real transaction\u0027s lifetime (from T_RUNNING to T_LOCKED\nlock - how long we were waiting for all handles to close\n (time the transaction was in T_LOCKED)\nflush - how long it took to flush all data (data\u003dordered)\nlog - how long it took to write the transaction to the log\nhndls - how many handles got to the transaction\nblock - how many blocks got to the transaction\ninlog - how many blocks are written to the log (block + descriptors)\nctime - how long it took to checkpoint the transaction\nwrite - how many blocks have been written during checkpointing\ndrop - how many blocks have been dropped during checkpointing\nclose - how many running transactions have been closed to checkpoint this one\n\nall times are in msec.\n\n\n[root@bob ~]# cat /proc/fs/jbd/sda/info\n280 transaction, each upto 8192 blocks\naverage:\n 1633ms waiting for transaction\n 3616ms running transaction\n 5ms transaction was being locked\n 1ms flushing data (in ordered mode)\n 1799ms logging transaction\n 11781 handles per transaction\n 5629 blocks per transaction\n 5641 logged blocks per transaction\n\nSigned-off-by: Johann Lombardi \u003cjohann.lombardi@bull.net\u003e\nSigned-off-by: Mariusz Kozlowski \u003cm.kozlowski@tuxland.pl\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\n" }, { "commit": "f5a7a6b0d9b6af7d46124ed3f6b3995225cb62d0", "tree": "cbb2bc79a488461bf3bec9b60d15e21717dc8360", "parents": [ "36df53f4a3e445175fc1e9d7f433599482ec6d7f" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Jan 28 23:58:27 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Jan 28 23:58:27 2008 -0500" }, "message": "jbd2: Fix assertion failure in fs/jbd2/checkpoint.c\n\nBefore we start committing a transaction, we call\n__journal_clean_checkpoint_list() to cleanup transaction\u0027s written-back\nbuffers.\n\nIf this call happens to remove all of them (and there were already some\nbuffers), __journal_remove_checkpoint() will decide to free the transaction\nbecause it isn\u0027t (yet) a committing transaction and soon we fail some\nassertion - the transaction really isn\u0027t ready to be freed :).\n\nWe change the check in __journal_remove_checkpoint() to free only a\ntransaction in T_FINISHED state. The locking there is subtle though (as\neverywhere in JBD ;(). We use j_list_lock to protect the check and a\nsubsequent call to __journal_drop_transaction() and do the same in the end\nof journal_commit_transaction() which is the only place where a transaction\ncan get to T_FINISHED state.\n\nProbably I\u0027m too paranoid here and such locking is not really necessary -\ncheckpoint lists are processed only from log_do_checkpoint() where a\ntransaction must be already committed to be processed or from\n__journal_clean_checkpoint_list() where kjournald itself calls it and thus\ntransaction cannot change state either. Better be safe if something\nchanges in future...\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "a7fa2baf8e2a6c0eb0a21f75e919c226179e8ff4", "tree": "3e41a4b90821d95835986945f1191e0f226a56a5", "parents": [ "cd02ff0b14b7e27cbdfd7ff2814b08199a2b7168" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Tue Oct 16 18:38:25 2007 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Oct 17 18:49:58 2007 -0400" }, "message": "jbd2: fix commit code to properly abort journal\n\nWe should really call journal_abort() and not __journal_abort_hard() in\ncase of errors. The latter call does not record the error in the journal\nsuperblock and thus filesystem won\u0027t be marked as with errors later (and\nuser could happily mount it without any warning).\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "cd02ff0b14b7e27cbdfd7ff2814b08199a2b7168", "tree": "fd9158bb92d086838a1b81b549243f3cdc23a7b8", "parents": [ "d802ffa8850f2a80d141457d7221809182ed8c9f" ], "author": { "name": "Mingming Cao", "email": "cmm@us.ibm.com", "time": "Tue Oct 16 18:38:25 2007 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Oct 17 18:49:58 2007 -0400" }, "message": "jbd2: JBD_XXX to JBD2_XXX naming cleanup\n\nchange JBD_XXX macros to JBD2_XXX in JBD2/Ext4\n\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "af1e76d6b3f37cb89d9192eaf83588adaf4728eb", "tree": "8b30dd421361a61c3f2e9c96bd574986b4e78c9e", "parents": [ "c089d490dfbf53bc0893dc9ef57cf3ee6448314d" ], "author": { "name": "Mingming Cao", "email": "cmm@us.ibm.com", "time": "Tue Oct 16 18:38:25 2007 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Oct 17 18:49:56 2007 -0400" }, "message": "JBD2: jbd2 slab allocation cleanups\n\nJBD2: Replace slab allocations with page allocations\n\nJBD2 allocate memory for committed_data and frozen_data from slab. However\nJBD2 should not pass slab pages down to the block layer. Use page allocator\npages instead. This will also prepare JBD for the large blocksize patchset.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\n" }, { "commit": "f89b779508f457b2af05dd5ac82a425db4a56ee3", "tree": "868baf5eb5cdb8ddce1b8e0512238b596c83c438", "parents": [ "fe28e42b99173ba088b1d8448e53029e100bff27" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Sun Jul 15 23:37:20 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Jul 16 09:05:34 2007 -0700" }, "message": "jbd2 commit: fix transaction dropping\n\nWe have to check that also the second checkpoint list is non-empty before\ndropping the transaction.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: Chuck Ebbert \u003ccebbert@redhat.com\u003e\nCc: Kirill Korotaev \u003cdev@openvz.org\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e63340ae6b6205fef26b40a75673d1c9c0c8bb90", "tree": "8d3212705515edec73c3936bb9e23c71d34a7b41", "parents": [ "04c9167f91e309c9c4ea982992aa08e83b2eb42e" ], "author": { "name": "Randy Dunlap", "email": "randy.dunlap@oracle.com", "time": "Tue May 08 00:28:08 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue May 08 11:15:07 2007 -0700" }, "message": "header cleaning: don\u0027t include smp_lock.h when not used\n\nRemove includes of \u003clinux/smp_lock.h\u003e where it is not used/needed.\nSuggested by Al Viro.\n\nBuilds cleanly on x86_64, i386, alpha, ia64, powerpc, sparc,\nsparc64, and arm (all 59 defconfigs).\n\nSigned-off-by: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "126039256cf164f69d8cfa20f1952d00fa61f52f", "tree": "cb0467f1d0bd7a80e10946d51a0605df3e274b74", "parents": [ "702a28b1e64be3dc313f5f0ceb6dc95edfbc5e18" ], "author": { "name": "Hisashi Hifumi", "email": "hifumi.hisashi@oss.ntt.co.jp", "time": "Wed Dec 06 20:39:17 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.osdl.org", "time": "Thu Dec 07 08:39:42 2006 -0800" }, "message": "[PATCH] jbd2: wait for already submitted t_sync_datalist buffer to complete\n\nIn the current jbd code, if a buffer on BJ_SyncData list is dirty and not\nlocked, the buffer is refiled to BJ_Locked list, submitted to the IO and\nwaited for IO completion.\n\nBut the fsstress test showed the case that when a buffer was already\nsubmitted to the IO just before the buffer_dirty(bh) check, the buffer was\nnot waited for IO completion.\n\nFollowing patch solves this problem. If it is assumed that a buffer is\nsubmitted to the IO before the buffer_dirty(bh) check and still being\nwritten to disk, this buffer is refiled to BJ_Locked list.\n\nSigned-off-by: Hisashi Hifumi \u003chifumi.hisashi@oss.ntt.co.jp\u003e\nCc: Jan Kara \u003cjack@ucw.cz\u003e\nCc: \"Stephen C. Tweedie\" \u003csct@redhat.com\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "18eba7aae080d4a5c0d850ea810e83d11f0a8d77", "tree": "58277e871787fad73c588b3ba81899f9c8b98553", "parents": [ "9b8f1f0106ab39ad58765d4e7c57189835f51127" ], "author": { "name": "Mingming Cao", "email": "cmm@us.ibm.com", "time": "Wed Oct 11 01:21:13 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Oct 11 11:14:18 2006 -0700" }, "message": "[PATCH] jbd2: switch blks_type from sector_t to ull\n\nSimilar to ext4, change blocks in JBD2 from sector_t to unsigned long long.\n\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "299717696d48531d70aeb4614c3939e4a28456c1", "tree": "d63c05b0192e480917c48126b2ac7c0f74d08018", "parents": [ "b517bea1c74e4773482b3f41b3f493522a8c8e30" ], "author": { "name": "Mingming Cao", "email": "cmm@us.ibm.com", "time": "Wed Oct 11 01:21:09 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Oct 11 11:14:17 2006 -0700" }, "message": "[PATCH] jbd2: sector_t conversion\n\nJBD layer in-kernel block varibles type fixes to support \u003e32 bit block number\nand convert to sector_t type.\n\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "b517bea1c74e4773482b3f41b3f493522a8c8e30", "tree": "1e98110151f1a35cde84ce1a2cb6f0cb871e9551", "parents": [ "d0d856e8bd6e697cb44b2b4dd038f3bec576a70e" ], "author": { "name": "Zach Brown", "email": "zach.brown@oracle.com", "time": "Wed Oct 11 01:21:08 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Oct 11 11:14:16 2006 -0700" }, "message": "[PATCH] 64-bit jbd2 core\n\nHere is the patch to JBD to handle 64 bit block numbers, originally from Zach\nBrown. This patch is useful only after adding support for 64-bit block\nnumbers in the filesystem.\n\nSigned-off-by: Badari Pulavarty \u003cpbadari@us.ibm.com\u003e\nSigned-off-by: Zach Brown \u003czach.brown@oracle.com\u003e\nSigned-off-by: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "f7f4bccb729844a0fa873e224e3a6f7eeed095bb", "tree": "b5b8ce65b5c17b4a7d45e43a3530f2cc977fb894", "parents": [ "470decc613ab2048b619a01028072d932d9086ee" ], "author": { "name": "Mingming Cao", "email": "cmm@us.ibm.com", "time": "Wed Oct 11 01:20:59 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Oct 11 11:14:15 2006 -0700" }, "message": "[PATCH] jbd2: rename jbd2 symbols to avoid duplication of jbd symbols\n\nMingming Cao originally did this work, and Shaggy reproduced it using some\nscripts from her.\n\nSigned-off-by: Mingming Cao \u003ccmm@us.ibm.com\u003e\nSigned-off-by: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "470decc613ab2048b619a01028072d932d9086ee", "tree": "5268576f5099040db94b8e984983c0bb28b2a9a7", "parents": [ "02ea2104c55b625cf5b5d9ba8586a4fc17920f5c" ], "author": { "name": "Dave Kleikamp", "email": "shaggy@austin.ibm.com", "time": "Wed Oct 11 01:20:57 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Oct 11 11:14:15 2006 -0700" }, "message": "[PATCH] jbd2: initial copy of files from jbd\n\nThis is a simple copy of the files in fs/jbd to fs/jbd2 and\n/usr/incude/linux/[ext4_]jbd.h to /usr/include/[ext4_]jbd2.h\n\nSigned-off-by: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" } ] }