)]}' { "log": [ { "commit": "5fc7b14177b1a1c2f2511aed62a4ca870d0332e7", "tree": "f64f2b0f424e7506dd6bb07c133348a9796b2212", "parents": [ "aa405b1f4238401616e7d98620170b424b2dbefc" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Thu May 05 12:56:51 2011 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Tue May 10 22:21:57 2011 +0900" }, "message": "nilfs2: use mark_buffer_dirty to mark btnode or meta data dirty\n\nThis replaces nilfs_mdt_mark_buffer_dirty and nilfs_btnode_mark_dirty\nmacros with mark_buffer_dirty and gets rid of nilfs_mark_buffer_dirty,\nan own mark buffer dirty function.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "aa405b1f4238401616e7d98620170b424b2dbefc", "tree": "c42986a31c3383490d010ae55a97c37f5095220c", "parents": [ "0ef28f9aec4dccfba33cef74412f601c1b48b658" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Thu May 05 12:56:51 2011 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Tue May 10 22:21:57 2011 +0900" }, "message": "nilfs2: always set back pointer to host inode in mapping-\u003ehost\n\nIn the current nilfs, page cache for btree nodes and meta data files\ndo not set a valid back pointer to the host inode in mapping-\u003ehost.\n\nThis will change it so that every address space in nilfs uses\nmapping-\u003ehost to hold its host inode.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "1cb2d38cb3e59d58e8321a0592e84b5761afb063", "tree": "151e7ffccd72f786a3da511b0f23906961835a22", "parents": [ "eaae0f37d83bed7ccd0c6d0f52de1de44f92aecc" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Mon Apr 04 12:53:28 2011 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Tue May 10 22:21:44 2011 +0900" }, "message": "nilfs2: get rid of private page allocator\n\nPreviously, nilfs was cloning pages for mmapped region to freeze their\ndata and ensure consistency of checksum during writeback cycles. A\nprivate page allocator was used for this page cloning. But, we no\nlonger need to do that since clear_page_dirty_for_io function sets up\npte so that vm_ops-\u003epage_mkwrite function is called right before the\nmmapped pages are modified and nilfs_page_mkwrite function can safely\nwait for the pages to be written back to disk.\n\nSo, this stops making a copy of mmapped pages during writeback, and\neliminates the private page allocation and deallocation functions from\nnilfs.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "4c63f5646e405b5010cc9499419060bf2e838f5b", "tree": "df91ba315032c8ec4aafeb3ab96fdfa7c6c656e1", "parents": [ "cafb0bfca1a73efd6d8a4a6a6a716e6134b96c24", "69d60eb96ae8a73cf9b79cf28051caf973006011" ], "author": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:58:35 2011 +0100" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:58:35 2011 +0100" }, "message": "Merge branch \u0027for-2.6.39/stack-plug\u0027 into for-2.6.39/core\n\nConflicts:\n\tblock/blk-core.c\n\tblock/blk-flush.c\n\tdrivers/md/raid1.c\n\tdrivers/md/raid10.c\n\tdrivers/md/raid5.c\n\tfs/nilfs2/btnode.c\n\tfs/nilfs2/mdt.c\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "7eaceaccab5f40bbfda044629a6298616aeaed50", "tree": "33954d12f63e25a47eb6d86ef3d3d0a5e62bf752", "parents": [ "73c101011926c5832e6e141682180c4debe2cf45" ], "author": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:07 2011 +0100" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:07 2011 +0100" }, "message": "block: remove per-queue plugging\n\nCode has been converted over to the new explicit on-stack plugging,\nand delay users have been converted to use the new API for that.\nSo lets kill off the old plugging along with aops-\u003esync_page().\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "2aa15890f3c191326678f1bd68af61ec6b8753ec", "tree": "347f5fdcd0678b12be92f266cd2a5e7a74749403", "parents": [ "78794b2cdeac37ac1fd950fc9c4454b56d88ac03" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Wed Feb 23 13:49:47 2011 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 23 19:52:52 2011 -0800" }, "message": "mm: prevent concurrent unmap_mapping_range() on the same inode\n\nMichael Leun reported that running parallel opens on a fuse filesystem\ncan trigger a \"kernel BUG at mm/truncate.c:475\"\n\nGurudas Pai reported the same bug on NFS.\n\nThe reason is, unmap_mapping_range() is not prepared for more than\none concurrent invocation per inode. For example:\n\n thread1: going through a big range, stops in the middle of a vma and\n stores the restart address in vm_truncate_count.\n\n thread2: comes in with a small (e.g. single page) unmap request on\n the same vma, somewhere before restart_address, finds that the\n vma was already unmapped up to the restart address and happily\n returns without doing anything.\n\nAnother scenario would be two big unmap requests, both having to\nrestart the unmapping and each one setting vm_truncate_count to its\nown value. This could go on forever without any of them being able to\nfinish.\n\nTruncate and hole punching already serialize with i_mutex. Other\ncallers of unmap_mapping_range() do not, and it\u0027s difficult to get\ni_mutex protection for all callers. In particular -\u003ed_revalidate(),\nwhich calls invalidate_inode_pages2_range() in fuse, may be called\nwith or without i_mutex.\n\nThis patch adds a new mutex to \u0027struct address_space\u0027 to prevent\nrunning multiple concurrent unmap_mapping_range() on the same mapping.\n\n[ We\u0027ll hopefully get rid of all this with the upcoming mm\n preemptibility series by Peter Zijlstra, the \"mm: Remove i_mmap_mutex\n lockbreak\" patch in particular. But that is for 2.6.39 ]\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nReported-by: Michael Leun \u003clkml20101129@newton.leun.net\u003e\nReported-by: Gurudas Pai \u003cgurudas.pai@oracle.com\u003e\nTested-by: Gurudas Pai \u003cgurudas.pai@oracle.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "622daaff0a8975fb5c5b95f24f3234550ba32e92", "tree": "5ae5de9044f962471fa2694654134f5b58f73ace", "parents": [ "27e6c7a3ce29ae5fa5bec4ed5917f8508bfac120" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Sun Dec 26 16:38:43 2010 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Mon Jan 10 14:05:46 2011 +0900" }, "message": "nilfs2: fiemap support\n\nThis adds fiemap to nilfs. Two new functions, nilfs_fiemap and\nnilfs_find_uncommitted_extent are added.\n\nnilfs_fiemap() implements the fiemap inode operation, and\nnilfs_find_uncommitted_extent() helps to get a range of data blocks\nwhose physical location has not been determined.\n\nnilfs_fiemap() collects extent information by looping through\nnilfs_bmap_lookup_contig and nilfs_find_uncommitted_extent routines.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "b1f6a4f294088b3fcf9ae67915ca550a1ded2819", "tree": "e4eab590fddc20111138e97ee4d55f10af519c8c", "parents": [ "ebdfed4dc59d177cf26013a0c9b8ee9652e9a140" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Tue Aug 31 11:40:34 2010 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Sat Oct 23 09:24:37 2010 +0900" }, "message": "nilfs2: add routines to redirect access to buffers of DAT file\n\nDuring garbage collection (GC), DAT file, which converts virtual block\nnumber to real block number, may return disk block number that is not\nyet written to the device.\n\nTo avoid access to unwritten blocks, the current implementation stores\nchanges to the caches of GCDAT during GC and atomically commit the\nchanges into the DAT file after they are written to the device.\n\nThis patch, instead, adds a function that makes a copy of specified\nbuffer and stores it in nilfs_shadow_map, and a function to get the\nbackup copy as needed (nilfs_mdt_freeze_buffer and\nnilfs_mdt_get_frozen_buffer respectively).\n\nBefore DAT changes block number in an entry block, it makes a copy and\nredirect access to the buffer so that address conversion function\n(i.e. nilfs_dat_translate) refers to the old address saved in the\ncopy.\n\nThis patch gives requisites for such redirection.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "ebdfed4dc59d177cf26013a0c9b8ee9652e9a140", "tree": "6ef90f068ae41c55234181c93d8e30a303126c43", "parents": [ "a8070dd365dd995f6139a2fc3aeee10159bdcc45" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Mon Sep 06 12:05:43 2010 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Sat Oct 23 09:24:37 2010 +0900" }, "message": "nilfs2: add routines to roll back state of DAT file\n\nThis adds optional function to metadata files which makes a copy of\nbmap, page caches, and b-tree node cache, and rolls back to the copy\nas needed.\n\nThis enhancement is intended to displace gcdat inode that provides a\nsimilar function in a different way.\n\nIn this patch, nilfs_shadow_map structure is added to store a copy of\nthe foregoing states. nilfs_mdt_setup_shadow_map relates this\nstructure to a metadata file. And, nilfs_mdt_save_to_shadow_map() and\nnilfs_mdt_restore_from_shadow_map() provides save and restore\nfunctions respectively. Finally, nilfs_mdt_clear_shadow_map() clears\nstates of nilfs_shadow_map.\n\nThe copy of b-tree node cache and page cache is made by duplicating\nonly dirty pages into corresponding caches in nilfs_shadow_map. Their\nrestoration is done by clearing dirty pages from original caches and\nby copying dirty pages back from nilfs_shadow_map.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "4e13e66bee2d792c1aae21797f16c181024834eb", "tree": "1fa51dade28bc025d9bda0e09dec1f202281a498", "parents": [ "464ece88630d0fb715ca942eabb1da825046a534" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Sun Jul 18 10:42:25 2010 +0900" }, "committer": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Fri Jul 23 10:02:15 2010 +0900" }, "message": "nilfs2: introduce check flag to btree node buffer\n\nnilfs_btree_get_block() now may return untested buffer due to\nread-ahead. This adds a new flag for buffer heads so that the btree\ncode can check whether the buffer is already verified or not.\n\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\n" }, { "commit": "0bd49f9446130a6a3914eb07b54db489e3222b34", "tree": "5122a227c9cb1cfe92a74c53baf1bb7ca7053435", "parents": [ "a60be987d45dd510aeb54389526f9957cfab106c" ], "author": { "name": "Ryusuke Konishi", "email": "konishi.ryusuke@lab.ntt.co.jp", "time": "Mon Apr 06 19:01:27 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 07 08:31:13 2009 -0700" }, "message": "nilfs2: buffer and page operations\n\nThis adds common routines for buffer/page operations used in B-tree\nnode caches, meta data files, or segment constructor (log writer).\n\nNILFS uses copy functions for buffers and pages due to the following\nreasons:\n\n 1) Relocation required for COW\n Since NILFS changes address of on-disk blocks, moving buffers\n in page cache is needed for the buffers which are not addressed\n by a file offset. If buffer size is smaller than page size,\n this involves partial copy of pages.\n\n 2) Freezing mmapped pages\n NILFS calculates checksums for each log to ensure its validity.\n If page data changes after the checksum calculation, this validity\n check will not work correctly. To avoid this failure for mmaped\n pages, NILFS freezes their data by copying.\n\n 3) Copy-on-write for DAT pages\n NILFS makes clones of DAT page caches in a copy-on-write manner\n during GC processes, and this ensures atomicity and consistency\n of the DAT in the transient state.\n\nIn addition, NILFS uses two obsolete functions, nilfs_mark_buffer_dirty()\nand nilfs_clear_page_dirty() respectively.\n\n* nilfs_mark_buffer_dirty() was required to avoid NULL pointer\n dereference faults:\n\n Since the page cache of B-tree node pages or data page cache of pseudo\n inodes does not have a valid mapping-\u003ehost, calling mark_buffer_dirty()\n for their buffers causes the fault; it calls __mark_inode_dirty(NULL)\n through __set_page_dirty().\n\n* nilfs_clear_page_dirty() was needed in the two cases:\n\n 1) For B-tree node pages and data pages of the dat/gcdat, NILFS2 clears\n page dirty flags when it copies back pages from the cloned cache\n (gcdat-\u003e{i_mapping,i_btnode_cache}) to its original cache\n (dat-\u003e{i_mapping,i_btnode_cache}).\n\n 2) Some B-tree operations like insertion or deletion may dispose buffers\n in dirty state, and this needs to cancel the dirty state of their\n pages. clear_page_dirty_for_io() caused faults because it does not\n clear the dirty tag on the page cache.\n\nSigned-off-by: Seiji Kihara \u003ckihara.seiji@lab.ntt.co.jp\u003e\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@lab.ntt.co.jp\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ] }