)]}' { "log": [ { "commit": "7bb46a6734a7e1ad4beaecc11cae7ed3ff81d30f", "tree": "e575d9c55e2a6ccc645dcb3ae2564de458b428f2", "parents": [ "7000d3c424e5bb350e502a477fb0e1ed42f8b10e" ], "author": { "name": "npiggin@suse.de", "email": "npiggin@suse.de", "time": "Thu May 27 01:05:33 2010 +1000" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu May 27 22:15:33 2010 -0400" }, "message": "fs: introduce new truncate sequence\n\nIntroduce a new truncate calling sequence into fs/mm subsystems. Rather than\nsetattr \u003e vmtruncate \u003e truncate, have filesystems call their truncate sequence\nfrom -\u003esetattr if filesystem specific operations are required. vmtruncate is\ndeprecated, and truncate_pagecache and inode_newsize_ok helpers introduced\npreviously should be used.\n\nsimple_setattr is introduced for simple in-ram filesystems to implement\nthe new truncate sequence. Eventually all filesystems should be converted\nto implement a setattr, and the default code in notify_change should go\naway.\n\nsimple_setsize is also introduced to perform just the ATTR_SIZE portion\nof simple_setattr (ie. changing i_size and trimming pagecache).\n\nTo implement the new truncate sequence:\n- filesystem specific manipulations (eg freeing blocks) must be done in\n the setattr method rather than -\u003etruncate.\n- vmtruncate can not be used by core code to trim blocks past i_size in\n the event of write failure after allocation, so this must be performed\n in the fs code.\n- convert usage of helpers block_write_begin, nobh_write_begin,\n cont_write_begin, and *blockdev_direct_IO* to use _newtrunc postfixed\n variants. These avoid calling vmtruncate to trim blocks (see previous).\n- inode_setattr should not be used. generic_setattr is a new function\n to be used to copy simple attributes into the generic inode.\n- make use of the better opportunity to handle errors with the new sequence.\n\nBig problem with the previous calling sequence: the filesystem is not called\nuntil i_size has already changed. This means it is not allowed to fail the\ncall, and also it does not know what the previous i_size was. Also, generic\ncode calling vmtruncate to truncate allocated blocks in case of error had\nno good way to return a meaningful error (or, for example, atomically handle\nblock deallocation).\n\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "e8bebe2f71d26871b0970ae1d9cf0ed3cdd9569d", "tree": "c0d82cbd11daaf579b74121c6641d58947091094", "parents": [ "6109e2ce2600e2db26cd0424bb9c6ed019723288", "82f3952c02add60b15eea9151d4d99b6b82066c6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 21 19:37:45 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 21 19:37:45 2010 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (69 commits)\n fix handling of offsets in cris eeprom.c, get rid of fake on-stack files\n get rid of home-grown mutex in cris eeprom.c\n switch ecryptfs_write() to struct inode *, kill on-stack fake files\n switch ecryptfs_get_locked_page() to struct inode *\n simplify access to ecryptfs inodes in -\u003ereadpage() and friends\n AFS: Don\u0027t put struct file on the stack\n Ban ecryptfs over ecryptfs\n logfs: replace inode uid,gid,mode initialization with helper function\n ufs: replace inode uid,gid,mode initialization with helper function\n udf: replace inode uid,gid,mode init with helper\n ubifs: replace inode uid,gid,mode initialization with helper function\n sysv: replace inode uid,gid,mode initialization with helper function\n reiserfs: replace inode uid,gid,mode initialization with helper function\n ramfs: replace inode uid,gid,mode initialization with helper function\n omfs: replace inode uid,gid,mode initialization with helper function\n bfs: replace inode uid,gid,mode initialization with helper function\n ocfs2: replace inode uid,gid,mode initialization with helper function\n nilfs2: replace inode uid,gid,mode initialization with helper function\n minix: replace inode uid,gid,mode init with helper\n ext4: replace inode uid,gid,mode init with helper\n ...\n\nTrivial conflict in fs/fs-writeback.c (mark bitfields unsigned)\n" }, { "commit": "01a05b337a5b647909e1d6670f57e7202318a5fb", "tree": "7877f08db14877d06346c8e1ef52aa17f2483e93", "parents": [ "35cf7ba0b46dc3582a01c3860b14bff122662aa3" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Mar 23 06:06:58 2010 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri May 21 18:31:16 2010 -0400" }, "message": "new helper: iterate_supers()\n\n... and switch the simple \"loop over superblocks and do something\"\nloops to it.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "6754af64641e8224c281ee5714e012e3ed41f701", "tree": "248a8c01aae46f98505d8c196393fae629f9691a", "parents": [ "8edd64bd6089e21f47dcdebb14b598b713213ddc" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 22 20:09:33 2010 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri May 21 18:31:15 2010 -0400" }, "message": "Convert simple loops over superblocks to list_for_each_entry_safe\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "551de6f34dfeefbeeadb32909c387d393114ecc8", "tree": "822af803e7e75ed476b0a176639c162e0395910d", "parents": [ "1712ac8fda7d8bc4dc921f5777b7423aacad7263" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Mar 22 19:36:35 2010 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri May 21 18:31:14 2010 -0400" }, "message": "Leave superblocks on s_list until the end\n\nWe used to remove from s_list and s_instances at the same\ntime. So let\u0027s *not* do the former and skip superblocks\nthat have empty s_instances in the loops over s_list.\n\nThe next step, of course, will be to get rid of rescan logics\nin those loops.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "fa4b9074cd8428958c2adf9dc0c831f46e27c193", "tree": "752103db616696fd7ba3374948fac9ebdf462d9e", "parents": [ "f1ac2502e19c59e996242d406fcc60e4c563e8ce" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Sat May 15 20:09:27 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri May 21 20:01:02 2010 +0200" }, "message": "buffer: make invalidate_bdev() drain all percpu LRU add caches\n\ninvalidate_bdev() should release all page cache pages which are clean\nand not being used; however, if some pages are still in the percpu LRU\nadd caches on other cpus, those pages are considered in used and don\u0027t\nget released. Fix it by calling lru_add_drain_all() before trying to\ninvalidate pages.\n\nThis problem was discovered while testing block automatic native\ncapacity unlocking. Null pages which were read before automatic\nunlocking didn\u0027t get released by invalidate_bdev() and ended up\ninterfering with partition scan after unlocking.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nAcked-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "c32da02342b7521df25fefc2ef20aee0e61cf887", "tree": "7e38f664fa3e13602c357d37f77d8adcf82fccc2", "parents": [ "dca1d9f6d7ae428c193f32bd3e9a4ca13176648b", "318ae2edc3b29216abd8a2510f3f80b764f06858" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 12 16:04:50 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 12 16:04:50 2010 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (56 commits)\n doc: fix typo in comment explaining rb_tree usage\n Remove fs/ntfs/ChangeLog\n doc: fix console doc typo\n doc: cpuset: Update the cpuset flag file\n Fix of spelling in arch/sparc/kernel/leon_kernel.c no longer needed\n Remove drivers/parport/ChangeLog\n Remove drivers/char/ChangeLog\n doc: typo - Table 1-2 should refer to \"status\", not \"statm\"\n tree-wide: fix typos \"ass?o[sc]iac?te\" -\u003e \"associate\" in comments\n No need to patch AMD-provided drivers/gpu/drm/radeon/atombios.h\n devres/irq: Fix devm_irq_match comment\n Remove reference to kthread_create_on_cpu\n tree-wide: Assorted spelling fixes\n tree-wide: fix \u0027lenght\u0027 typo in comments and code\n drm/kms: fix spelling in error message\n doc: capitalization and other minor fixes in pnp doc\n devres: typo fix s/dev/devm/\n Remove redundant trailing semicolons from macros\n fix typo \"definetly\" -\u003e \"definitely\" in comment\n tree-wide: s/widht/width/g typo in comments\n ...\n\nFix trivial conflict in Documentation/laptops/00-INDEX\n" }, { "commit": "019b4d123aa7b9fc135b532e021cfde85db7665d", "tree": "91edbae535136417bc1277dd2c40d98c7ea7ec8f", "parents": [ "03affdef4fc85e416e3862abb0aa549c6034cdd0" ], "author": { "name": "Richard Kennedy", "email": "richard@rsk.demon.co.uk", "time": "Wed Mar 10 15:20:33 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 12 15:52:27 2010 -0800" }, "message": "fs: buffer_head: remove kmem_cache constructor to reduce memory usage under slub\n\nWhen using slub, having a kmem_cache constructor forces slub to add a free\npointer to the size of the cached object, which can have a significant\nimpact to the number of small objects that can fit into a slab.\n\nAs buffer_head is relatively small and we can have large numbers of them,\nremoving the constructor is a definite win.\n\nOn x86_64 removing the constructor gives me 39 objects/slab, 3 more than\nwithout the patch. And on x86_32 73 objects/slab, which is 9 more.\n\nAs alloc_buffer_head() already initializes each new object there is very\nlittle difference in actual code run.\n\nSigned-off-by: Richard Kennedy \u003crichard@rsk.demon.co.uk\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nAcked-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "2a61aa401638529cd4231f6106980d307fba98fa", "tree": "a3d7565570c5996d0b3ae5fdf0126e065e750431", "parents": [ "c41b20e721ea4f6f20f66a66e7f0c3c97a2ca9c2" ], "author": { "name": "Adam Buchbinder", "email": "adam.buchbinder@gmail.com", "time": "Fri Dec 11 16:35:40 2009 -0500" }, "committer": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Thu Feb 04 11:55:45 2010 +0100" }, "message": "Fix misspellings of \"invocation\" in comments.\n\nSome comments misspell \"invocation\"; this fixes them. No code\nchanges.\n\nSigned-off-by: Adam Buchbinder \u003cadam.buchbinder@gmail.com\u003e\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "6d7f18f6ea3a13af95bdf507fc54d42b165e1712", "tree": "8f6f3a6d46835aa767823fa7049609408a87afc2", "parents": [ "53cddfcc0e760d2b364878b6dadbd0c6d087cfae", "56a131dcf7ed36c3c6e36bea448b674ea85ed5bb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 25 09:27:30 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Sep 25 09:27:30 2009 -0700" }, "message": "Merge branch \u0027writeback\u0027 of git://git.kernel.dk/linux-2.6-block\n\n* \u0027writeback\u0027 of git://git.kernel.dk/linux-2.6-block:\n writeback: writeback_inodes_sb() should use bdi_start_writeback()\n writeback: don\u0027t delay inodes redirtied by a fast dirtier\n writeback: make the super_block pinning more efficient\n writeback: don\u0027t resort for a single super_block in move_expired_inodes()\n writeback: move inodes from one super_block together\n writeback: get rid to incorrect references to pdflush in comments\n writeback: improve readability of the wb_writeback() continue/break logic\n writeback: cleanup writeback_single_inode()\n writeback: kupdate writeback shall not stop when more io is possible\n writeback: stop background writeback when below background threshold\n writeback: balance_dirty_pages() shall write more than dirtied pages\n fs: Fix busyloop in wb_writeback()\n" }, { "commit": "5b0830cb9085f4b69f9d57d7f3aaff322ffbec26", "tree": "10040eb359269d4cd05487790b758144a69e8e39", "parents": [ "71fd05a887e0f3f6bfff76ff81b33776177d0606" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 23 19:37:09 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Sep 25 18:08:25 2009 +0200" }, "message": "writeback: get rid to incorrect references to pdflush in comments\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "c08d3b0e33edce28e9cfa7b64f7fe5bdeeb29248", "tree": "6c161cc57a2873e6f91335113192a99531901cd8", "parents": [ "25d9e2d15286281ec834b829a4aaf8969011f1cd" ], "author": { "name": "npiggin@suse.de", "email": "npiggin@suse.de", "time": "Fri Aug 21 02:35:06 2009 +1000" }, "committer": { "name": "al", "email": "al@dizzy.pdmi.ras.ru", "time": "Thu Sep 24 08:41:47 2009 -0400" }, "message": "truncate: use new helpers\n\nUpdate some fs code to make use of new helper functions introduced\nin the previous patch. Should be no significant change in behaviour\n(except CIFS now calls send_sig under i_lock, via inode_newsize_ok).\n\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nAcked-by: Miklos Szeredi \u003cmiklos@szeredi.hu\u003e\nCc: linux-nfs@vger.kernel.org\nCc: Trond.Myklebust@netapp.com\nCc: linux-cifs-client@lists.samba.org\nCc: sfrench@samba.org\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "1fe72eaa0f46a0fa4cdcd8f3f7853b6d39469784", "tree": "1bcaa65c0011cf329d5dd2a06be519460d06e434", "parents": [ "88e0fbc452ed94393bf89585c2b90edb94749b45" ], "author": { "name": "H Hartley Sweeten", "email": "hartleys@visionengravers.com", "time": "Tue Sep 22 16:43:51 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 23 07:39:29 2009 -0700" }, "message": "fs/buffer.c: clean up EXPORT* macros\n\nAccording to Documentation/CodingStyle the EXPORT* macro should follow\nimmediately after the closing function brace line.\n\nAlso, mark_buffer_async_write_endio() and do_thaw_all() are not used\nelsewhere so they should be marked as static.\n\nIn addition, file_fsync() is actually in fs/sync.c so move the EXPORT* to\nthat file.\n\nSigned-off-by: H Hartley Sweeten \u003chsweeten@visionengravers.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "03ba3782e8dcc5b0e1efe440d33084f066e38cae", "tree": "e5a6513b411de16a46199530ec98ef9b7f1efc50", "parents": [ "66f3b8e2e103a0b93b945764d98e9ba46cb926dd" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 09 09:08:54 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Sep 11 09:20:25 2009 +0200" }, "message": "writeback: switch to per-bdi threads for flushing data\n\nThis gets rid of pdflush for bdi writeout and kupdated style cleaning.\npdflush writeout suffers from lack of locality and also requires more\nthreads to handle the same workload, since it has to work in a\nnon-blocking fashion against each queue. This also introduces lumpy\nbehaviour and potential request starvation, since pdflush can be starved\nfor queue access if others are accessing it. A sample ffsb workload that\ndoes random writes to files is about 8% faster here on a simple SATA drive\nduring the benchmark phase. File layout also seems a LOT more smooth in\nvmstat:\n\n r b swpd free buff cache si so bi bo in cs us sy id wa\n 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42\n 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44\n 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37\n 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58\n 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34\n 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37\n 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44\n 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38\n 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41\n 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45\n\nwhere vanilla tends to fluctuate a lot in the creation phase:\n\n r b swpd free buff cache si so bi bo in cs us sy id wa\n 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36\n 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51\n 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40\n 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37\n 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41\n 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49\n 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36\n 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43\n 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39\n 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45\n 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34\n 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54\n\nA 10 disk test with btrfs performs 26% faster with per-bdi flushing. A\nSSD based writeback test on XFS performs over 20% better as well, with\nthe throughput being very stable around 1GB/sec, where pdflush only\nmanages 750MB/sec and fluctuates wildly while doing so. Random buffered\nwrites to many files behave a lot better as well, as does random mmap\u0027ed\nwrites.\n\nA separate thread is added to sync the super blocks. In the long term,\nadding sync_supers_bdi() functionality could get rid of this thread again.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "8e9d78edea3ce5c0036f85b93091483f2f15443a", "tree": "898c98daca2602f0df70558211f30ff1bd2bcf6c", "parents": [ "4dfd79e7b42bff334128907e28c3b41f1ef1cec8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 21 17:40:08 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 21 17:40:08 2009 -0700" }, "message": "Re-introduce page mapping check in mark_buffer_dirty()\n\nIn commit a8e7d49aa7be728c4ae241a75a2a124cdcabc0c5 (\"Fix race in\ncreate_empty_buffers() vs __set_page_dirty_buffers()\"), I removed a test\nfor a NULL page mapping unintentionally when some of the code inside\n__set_page_dirty() was moved to the callers.\n\nThat removal generally didn\u0027t matter, since a filesystem would serialize\ntruncation (which clears the page mapping) against writing (which marks\nthe buffer dirty), so locking at a higher level (either per-page or an\ninode at a time) should mean that the buffer page would be stable. And\nindeed, nothing bad seemed to happen.\n\nExcept it turns out that apparently reiserfs does something odd when\nunder load and writing out the journal, and we have a number of bugzilla\nentries that look similar:\n\n\thttp://bugzilla.kernel.org/show_bug.cgi?id\u003d13556\n\thttp://bugzilla.kernel.org/show_bug.cgi?id\u003d13756\n\thttp://bugzilla.kernel.org/show_bug.cgi?id\u003d13876\n\nand it looks like reiserfs depended on that check (the common theme\nseems to be \"data\u003djournal\", and a journal writeback during a truncate).\n\nI suspect reiserfs should have some additional locking, but in the\nmeantime this should get us back to the pre-2.6.29 behavior.\n\nPattern-pointed-out-by: Roland Kletzing \u003cdevzero@web.de\u003e\nCc: stable@kernel.org (2.6.29 and 2.6.30)\nCc: Jeff Mahoney \u003cjeffm@suse.com\u003e\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c9059598ea8981d02356eead3188bf7fa4d717b8", "tree": "03e73b20a30e988da7c6a3e0ad93b2dc5843274d", "parents": [ "0a33f80a8373eca7f4bea3961d1346c3815fa5ed", "b0fd271d5fba0b2d00888363f3869e3f9b26caa9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 10:52:27 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 11:10:35 2009 -0700" }, "message": "Merge branch \u0027for-2.6.31\u0027 of git://git.kernel.dk/linux-2.6-block\n\n* \u0027for-2.6.31\u0027 of git://git.kernel.dk/linux-2.6-block: (153 commits)\n block: add request clone interface (v2)\n floppy: fix hibernation\n ramdisk: remove long-deprecated \"ramdisk\u003d\" boot-time parameter\n fs/bio.c: add missing __user annotation\n block: prevent possible io_context-\u003erefcount overflow\n Add serial number support for virtio_blk, V4a\n block: Add missing bounce_pfn stacking and fix comments\n Revert \"block: Fix bounce limit setting in DM\"\n cciss: decode unit attention in SCSI error handling code\n cciss: Remove no longer needed sendcmd reject processing code\n cciss: change SCSI error handling routines to work with interrupts enabled.\n cciss: separate error processing and command retrying code in sendcmd_withirq_core()\n cciss: factor out fix target status processing code from sendcmd functions\n cciss: simplify interface of sendcmd() and sendcmd_withirq()\n cciss: factor out core of sendcmd_withirq() for use by SCSI error handling code\n cciss: Use schedule_timeout_uninterruptible in SCSI error handling code\n block: needs to set the residual length of a bidi request\n Revert \"block: implement blkdev_readpages\"\n block: Fix bounce limit setting in DM\n Removed reference to non-existing file Documentation/PCI/PCI-DMA-mapping.txt\n ...\n\nManually fix conflicts with tracing updates in:\n\tblock/blk-sysfs.c\n\tdrivers/ide/ide-atapi.c\n\tdrivers/ide/ide-cd.c\n\tdrivers/ide/ide-floppy.c\n\tdrivers/ide/ide-tape.c\n\tinclude/trace/events/block.h\n\tkernel/trace/blktrace.c\n" }, { "commit": "e893123c7378192c094747dadec326b7c000c190", "tree": "3fac44dceaa87339819d078e2bd69ea9a01056c3", "parents": [ "27951daa71f1c91300ae4de9441916d1ffe2b078", "a41f20716975910d9beb90b7efc61107901492b8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 10:00:50 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 10:00:50 2009 -0700" }, "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: (49 commits)\n ext4: Avoid corrupting the uninitialized bit in the extent during truncate\n ext4: Don\u0027t treat a truncation of a zero-length file as replace-via-truncate\n ext4: fix dx_map_entry to support 256k directory blocks\n ext4: truncate the file properly if we fail to copy data from userspace\n ext4: Avoid leaking blocks after a block allocation failure\n ext4: Change all super.c messages to print the device\n ext4: Get rid of EXTEND_DISKSIZE flag of ext4_get_blocks_handle()\n ext4: super.c whitespace cleanup\n jbd2: Fix minor typos in comments in fs/jbd2/journal.c\n ext4: Clean up calls to ext4_get_group_desc()\n ext4: remove unused function __ext4_write_dirty_metadata\n ext2: Fix memory leak in ext2_fill_super() in case of a failed mount\n ext3: Fix memory leak in ext3_fill_super() in case of a failed mount\n ext4: Fix memory leak in ext4_fill_super() in case of a failed mount\n ext4: down i_data_sem only for read when walking tree for fiemap\n ext4: Add a comprehensive block validity check to ext4_get_blocks()\n ext4: Clean up ext4_get_blocks() so it does not depend on bh_result-\u003eb_state\n ext4: Merge ext4_da_get_block_write() into mpage_da_map_blocks()\n ext4: Add BUG_ON debugging checks to noalloc_get_block_write()\n ext4: Add documentation to the ext4_*get_block* functions\n ...\n" }, { "commit": "460bcf57b128ce1c0dd553d905fedc097f9955c6", "tree": "5e5f45d6c3b5584a086a306a67d4fb984fef5653", "parents": [ "064e38aaded5269e573ac1c765284fd65c8ebd13" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue May 12 07:37:56 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat Jun 06 06:17:25 2009 -0400" }, "message": "Fix nobh_truncate_page() to not pass stack garbage to get_block()\n\nThe nobh_truncate_page() function is used by ext2, exofs, and jfs. Of\nthese three, only ext2 and jfs\u0027s get_block() function pays attention\nto bh-\u003eb_size --- which is normally always the filesystem blocksize\nexcept when the get_block() function is called by either\nmpage_readpage(), mpage_readpages(), or the direct I/O routines in\nfs/direct_io.c.\n\nUnfortunately, nobh_truncate_page() does not initialize map_bh before\ncalling the filesystem-supplied get_block() function. So ext2 and jfs\nwill try to calculate the number of blocks to map by taking stack\ngarbage and shifting it left by inode-\u003ei_blkbits. This should be\n*mostly* harmless (except the filesystem will do some unnneeded work)\nunless the stack garbage is less than filesystem\u0027s blocksize, in which\ncase maxblocks will be zero, and the attempt to find out whether or\nnot the filesystem has a hole at a given logical block will fail, and\nthe page cache entry might not get zero\u0027ed out.\n\nAlso if the stack garbage in in map_bh-\u003estate happens to have the\nBH_Mapped bit set, there could be an attempt to call readpage() on a\nnon-existent page, which could cause nobh_truncate_page() to return an\nerror when it should not.\n\nFix this by initializing map_bh-\u003estate and map_bh-\u003esize.\n\nFortunately, it\u0027s probably fairly unlikely that ext2 and jfs users\nmount with nobh these days.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nCc: linux-fsdevel@vger.kernel.org\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "e1defc4ff0cf57aca6c5e3ff99fa503f5943c1f1", "tree": "d60d15a082171c58ac811d547d51a9c3119f23e3", "parents": [ "9bd7de51ee8537094656149eaf45338cadb7d7d4" ], "author": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Fri May 22 17:17:49 2009 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri May 22 23:22:54 2009 +0200" }, "message": "block: Do away with the notion of hardsect_size\n\nUntil now we have had a 1:1 mapping between storage device physical\nblock size and the logical block sized used when addressing the device.\nWith SATA 4KB drives coming out that will no longer be the case. The\nsector size will be 4KB but the logical block size will remain\n512-bytes. Hence we need to distinguish between the physical block size\nand the logical ditto.\n\nThis patch renames hardsect_size to logical_block_size.\n\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "b827e496c893de0c0f142abfaeb8730a2fd6b37f", "tree": "a86aecd5d811f9306b9662ceb5a5a45091b62b97", "parents": [ "a5fc1abe438b87a9d128beebc377f78e2681a76d" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Thu Apr 30 15:08:16 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 02 15:36:09 2009 -0700" }, "message": "mm: close page_mkwrite races\n\nChange page_mkwrite to allow implementations to return with the page\nlocked, and also change it\u0027s callers (in page fault paths) to hold the\nlock until the page is marked dirty. This allows the filesystem to have\nfull control of page dirtying events coming from the VM.\n\nRather than simply hold the page locked over the page_mkwrite call, we\ncall page_mkwrite with the page unlocked and allow callers to return with\nit locked, so filesystems can avoid LOR conditions with page lock.\n\nThe problem with the current scheme is this: a filesystem that wants to\nassociate some metadata with a page as long as the page is dirty, will\nperform this manipulation in its -\u003epage_mkwrite. It currently then must\nreturn with the page unlocked and may not hold any other locks (according\nto existing page_mkwrite convention).\n\nIn this window, the VM could write out the page, clearing page-dirty. The\nfilesystem has no good way to detect that a dirty pte is about to be\nattached, so it will happily write out the page, at which point, the\nfilesystem may manipulate the metadata to reflect that the page is no\nlonger dirty.\n\nIt is not always possible to perform the required metadata manipulation in\n-\u003eset_page_dirty, because that function cannot block or fail. The\nfilesystem may need to allocate some data structure, for example.\n\nAnd the VM cannot mark the pte dirty before page_mkwrite, because\npage_mkwrite is allowed to fail, so we must not allow any window where the\npage could be written to if page_mkwrite does fail.\n\nThis solution of holding the page locked over the 3 critical operations\n(page_mkwrite, setting the pte dirty, and finally setting the page dirty)\ncloses out races nicely, preventing page cleaning for writeout being\ninitiated in that window. This provides the filesystem with a strong\nsynchronisation against the VM here.\n\n- Sage needs this race closed for ceph filesystem.\n- Trond for NFS (http://bugzilla.kernel.org/show_bug.cgi?id\u003d12913).\n- I need it for fsblock.\n- I suspect other filesystems may need it too (eg. btrfs).\n- I have converted buffer.c to the new locking. Even simple block allocation\n under dirty pages might be susceptible to i_size changing under partial page\n at the end of file (we also have a buffer.c-side problem here, but it cannot\n be fixed properly without this patch).\n- Other filesystems (eg. NFS, maybe btrfs) will need to change their\n page_mkwrite functions themselves.\n\n[ This also moves page_mkwrite another step closer to fault, which should\n eventually allow page_mkwrite to be moved into -\u003efault, and thus avoiding a\n filesystem calldown and page lock/unlock cycle in __do_fault. ]\n\n[akpm@linux-foundation.org: fix derefs of NULL -\u003emapping]\nCc: Sage Weil \u003csage@newdream.net\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Valdis Kletnieks \u003cValdis.Kletnieks@vt.edu\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": "8fb0e342481c4d80040670fec915f0b9c7c6499a", "tree": "6574f483e45cb24ebf9eb2121b67b20a75a9f180", "parents": [ "79ffab34391933ee3b95dac7f25c0478fa2f8f1e" ], "author": { "name": "Aneesh Kumar K.V", "email": "aneesh.kumar@linux.vnet.ibm.com", "time": "Tue May 12 16:22:37 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue May 12 16:22:37 2009 -0400" }, "message": "vfs: Add BUG_ON for delayed and unwritten flags in submit_bh()\n\nThe BH_Delay and BH_Unwritten flags should never leak out to\nsubmit_bh(). So add some BUG_ON() checks to submit_bh so we can get a\nstack trace and determine how and why this might have happened.\n\n(Note that only XFS and ext4 use these buffer head flags, and XFS does\nnot use submit_bh(). So this patch should only modify behavior for\next4.)\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\nCc: linux-fsdevel@vger.kernel.org\n" }, { "commit": "35c80d5f400f68f2eccf3069d1c068e154bde9c9", "tree": "43dd3ebb3e654b1a1ed4210089b7dc27b1f516e9", "parents": [ "f69955855eac55a048d26a1618f50dfaa160a006" ], "author": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Wed Apr 15 13:22:38 2009 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 16 07:47:49 2009 -0700" }, "message": "Add block_write_full_page_endio for passing endio handler\n\nblock_write_full_page doesn\u0027t allow the caller to control what happens\nwhen the IO is over. This adds a new call named block_write_full_page_endio\nso the buffer head end_io handler can be provided by the caller.\n\nThis will be used by the ext3 data\u003dguarded mode to do i_size updates in\na workqueue based end_io handler. end_buffer_async_write is also\nexported so it can be called to do the dirty work of managing page\nwriteback for the higher level end_io handler.\n\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\nAcked-by: Theodore Tso \u003ctytso@mit.edu\u003e\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "053c525fcf976810f023d96472f414c0d5e6339b", "tree": "2e1954da38bd4c59edb9ff3daf0111b02c181b0a", "parents": [ "329007ce25d56fc7113df7b4828d607806d8bc21" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Apr 08 13:44:08 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Apr 15 08:28:12 2009 +0200" }, "message": "buffer: switch do_emergency_thaw() away from pdflush_operation()\n\nThis is (again) a preparatory patch similar to commit\na2a9537ac0b37a5da6fbe7e1e9cb06c524d2a9c4. It open codes a simple\nasync way of executing do_thaw_all() out of context, so we can get\nrid of pdflush.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "6e34eeddf7deec1444bbddab533f03f520d8458c", "tree": "a0e189c329a15363073eab257a3f704cf778107e", "parents": [ "577c9c456f0e1371cbade38eaf91ae8e8a308555" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Apr 07 18:12:43 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Apr 08 13:15:09 2009 -0400" }, "message": "block_write_full_page: switch synchronous writes to use WRITE_SYNC_PLUG\n\nNow that we have a distinction between WRITE_SYNC and WRITE_SYNC_PLUG,\nuse WRITE_SYNC_PLUG in __block_write_full_page() to avoid unplugging\nthe block device I/O queue between each page that gets flushed out.\n\nOtherwise, when we run sync() or fsync() and we need to write out a\nlarge number of pages, the block device queue will get unplugged\nbetween for every page that is flushed out, which will be a pretty\nserious performance regression caused by commit a64c8610.\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n" }, { "commit": "1aa2a7cc6fd7b5c86681a6ae9dfd1072c261a435", "tree": "4910dbeec6f365b133249091af74e1a0db829f1f", "parents": [ "aeb6fafb8fa53266d70ca7474fcda2bdaf96524a" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Apr 06 14:48:08 2009 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 06 08:04:54 2009 -0700" }, "message": "block: switch sync_dirty_buffer() over to WRITE_SYNC\n\nWe should now have the logic in place to handle this properly\nwithout regressing on the write performance, so re-enable\nthe sync writes.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9cf6b720f84d6999ff9a514d0a939dd183846aaf", "tree": "13f58881927769c792b8b1f685282aa67264b9ed", "parents": [ "a1f242524c3c1f5d40f1c9c343427e34d1aadd6e" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Apr 06 14:48:03 2009 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 06 08:04:53 2009 -0700" }, "message": "block: fsync_buffers_list() should use SWRITE_SYNC_PLUG\n\nThen it can submit all the buffers without unplugging for each one.\nWe will kick off the pending IO if we come across a new address space.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "20bec8ab1458c24bed0d5492ee15d87807fc415a", "tree": "e5f910947dbe314b96a591e41e2cfb2d3322caad", "parents": [ "18b34b9546dc192d978dda940673f40928d2e36e", "e7c8f5079ed9ec9e6eb1abe3defc5fb4ebfdf1cb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 03 11:10:33 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 03 11:10:33 2009 -0700" }, "message": "Merge branch \u0027ext3-latency-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4\n\n* \u0027ext3-latency-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:\n ext3: Add replace-on-rename hueristics for data\u003dwriteback mode\n ext3: Add replace-on-truncate hueristics for data\u003dwriteback mode\n ext3: Use WRITE_SYNC for commits which are caused by fsync()\n block_write_full_page: Use synchronous writes for WBC_SYNC_ALL writebacks\n" }, { "commit": "8fe74cf053de7ad2124a894996f84fa890a81093", "tree": "77dcd8fbf33ce53a3821942233962fb28c6f2848", "parents": [ "c2eb2fa6d2b6fe122d3479ec5b28d978418b2698", "ced117c73edc917e96dea7cca98c91383f0792f7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 21:09:10 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 21:09:10 2009 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:\n Remove two unneeded exports and make two symbols static in fs/mpage.c\n Cleanup after commit 585d3bc06f4ca57f975a5a1f698f65a45ea66225\n Trim includes of fdtable.h\n Don\u0027t crap into descriptor table in binfmt_som\n Trim includes in binfmt_elf\n Don\u0027t mess with descriptor table in load_elf_binary()\n Get rid of indirect include of fs_struct.h\n New helper - current_umask()\n check_unsafe_exec() doesn\u0027t care about signal handlers sharing\n New locking/refcounting for fs_struct\n Take fs_struct handling to new file (fs/fs_struct.c)\n Get rid of bumping fs_struct refcount in pivot_root(2)\n Kill unsharing fs_struct in __set_personality()\n" }, { "commit": "97f76d3d197f201ac8a8a3ced5b8fef81568e50e", "tree": "8264f920a1f1a83e9cb0247ffe9390a7b6332852", "parents": [ "9a896c9a48ac6704c0ce8ee081b836644d0afe40" ], "author": { "name": "Nikanth Karthikesan", "email": "knikanth@suse.de", "time": "Thu Apr 02 16:56:46 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 02 19:04:49 2009 -0700" }, "message": "vfs: check bh-\u003eb_blocknr only if BH_Mapped is set\n\nCheck bh-\u003eb_blocknr only if BH_Mapped is set.\n\nakpm: I doubt if b_blocknr is ever uninitialised here, but it could\nconceivably cause a problem if we\u0027re doing a lookup for block zero.\n\nSigned-off-by: Nikanth Karthikesan \u003cknikanth@suse.de\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c2d7543851849a6923680cdd7e1047ed1a84a1c5", "tree": "bf3038819d4be83a1d1e64d7b95bbb3d9d908544", "parents": [ "55a63998b8967615a15e2211ba0ff3a84a565824" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 31 15:23:46 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 01 08:59:17 2009 -0700" }, "message": "filesystem freeze: allow SysRq emergency thaw to thaw frozen filesystems\n\nNow that the filesystem freeze operation has been elevated to the VFS, and\nis just an ioctl away, some sort of safety net for unintentionally frozen\nroot filesystems may be in order.\n\nThe timeout thaw originally proposed did not get merged, but perhaps\nsomething like this would be useful in emergencies.\n\nFor example, freeze /path/to/mountpoint may freeze your root filesystem if\nyou forgot that you had that unmounted.\n\nI chose \u0027j\u0027 as the last remaining character other than \u0027h\u0027 which is sort\nof reserved for help (because help is generated on any unknown character).\n\nI\u0027ve tested this on a non-root fs with multiple (nested) freezers, as well\nas on a system rendered unresponsive due to a frozen root fs.\n\n[randy.dunlap@oracle.com: emergency thaw only if CONFIG_BLOCK enabled]\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nCc: Takashi Sato \u003ct-sato@yk.jp.nec.com\u003e\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": "327c0e968645f2601a43f5ea7c19c7b3a5fa0a34", "tree": "acc6789c120a6d5000ceebf51e690d14c6cfcacb", "parents": [ "2678958e1225f350806d90f211a3b475f64aee80" ], "author": { "name": "KAMEZAWA Hiroyuki", "email": "kamezawa.hiroyu@jp.fujitsu.com", "time": "Tue Mar 31 15:23:31 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 01 08:59:15 2009 -0700" }, "message": "vmscan: fix it to take care of nodemask\n\ntry_to_free_pages() is used for the direct reclaim of up to\nSWAP_CLUSTER_MAX pages when watermarks are low. The caller to\nalloc_pages_nodemask() can specify a nodemask of nodes that are allowed to\nbe used but this is not passed to try_to_free_pages(). This can lead to\nunnecessary reclaim of pages that are unusable by the caller and int the\nworst case lead to allocation failure as progress was not been make where\nit is needed.\n\nThis patch passes the nodemask used for alloc_pages_nodemask() to\ntry_to_free_pages().\n\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nSigned-off-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "56a76f8275c379ed73c8a43cfa1dfa2f5e9cfa19", "tree": "cbeaa82516c4818f72535b6ebe48a607cef88af4", "parents": [ "c2ec175c39f62949438354f603f4aa170846aabb" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Mar 31 15:23:23 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 01 08:59:14 2009 -0700" }, "message": "fs: fix page_mkwrite error cases in core code and btrfs\n\npage_mkwrite is called with neither the page lock nor the ptl held. This\nmeans a page can be concurrently truncated or invalidated out from\nunderneath it. Callers are supposed to prevent truncate races themselves,\nhowever previously the only thing they can do in case they hit one is to\nraise a SIGBUS. A sigbus is wrong for the case that the page has been\ninvalidated or truncated within i_size (eg. hole punched). Callers may\nalso have to perform memory allocations in this path, where again, SIGBUS\nwould be wrong.\n\nThe previous patch (\"mm: page_mkwrite change prototype to match fault\")\nmade it possible to properly specify errors. Convert the generic buffer.c\ncode and btrfs to return sane error values (in the case of page removed\nfrom pagecache, VM_FAULT_NOPAGE will cause the fault handler to exit\nwithout doing anything, and the fault will be retried properly).\n\nThis fixes core code, and converts btrfs as a template/example. All other\nfilesystems defining their own page_mkwrite should be fixed in a similar\nmanner.\n\nAcked-by: Chris Mason \u003cchris.mason@oracle.com\u003e\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c2ec175c39f62949438354f603f4aa170846aabb", "tree": "f2c9bf1bec2deabe2d3a5092405b027637b6ead3", "parents": [ "c2fdf3a9b2d52842808a8e551b53b55dd9b45030" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Mar 31 15:23:21 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 01 08:59:14 2009 -0700" }, "message": "mm: page_mkwrite change prototype to match fault\n\nChange the page_mkwrite prototype to take a struct vm_fault, and return\nVM_FAULT_xxx flags. There should be no functional change.\n\nThis makes it possible to return much more detailed error information to\nthe VM (and also can provide more information eg. virtual_address to the\ndriver, which might be important in some special cases).\n\nThis is required for a subsequent fix. And will also make it easier to\nmerge page_mkwrite() with fault() in future.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Chris Mason \u003cchris.mason@oracle.com\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nCc: Miklos Szeredi \u003cmiklos@szeredi.hu\u003e\nCc: Steven Whitehouse \u003cswhiteho@redhat.com\u003e\nCc: Mark Fasheh \u003cmfasheh@suse.com\u003e\nCc: Joel Becker \u003cjoel.becker@oracle.com\u003e\nCc: Artem Bityutskiy \u003cdedekind@infradead.org\u003e\nCc: Felix Blyakher \u003cfelixb@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e3a7cca1ef4c1af9b0acef9bd66eff6582a737b5", "tree": "717097ec82ce0de55e44b2fe3e35fa041e5169b2", "parents": [ "bd2f6199cf9af472aeefa1b642c9f504f19e6008" ], "author": { "name": "Edward Shishkin", "email": "edward.shishkin@gmail.com", "time": "Tue Mar 31 15:19:39 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 01 08:59:12 2009 -0700" }, "message": "vfs: add/use account_page_dirtied()\n\nAdd a helper function account_page_dirtied(). Use that from two\ncallsites. reiser4 adds a function which adds a third callsite.\n\nSigned-off-by: Edward Shishkin\u003cedward.shishkin@gmail.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "47e4491b40df73c3b117e3d80b31b5b512a4b19f", "tree": "a751af5e30db8f935b97bd671d0055d670b75fe5", "parents": [ "e5824c97a993ac5bd22cec34e6cb6f85d39b73ba" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Apr 01 07:07:16 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Apr 01 07:07:16 2009 -0400" }, "message": "Cleanup after commit 585d3bc06f4ca57f975a5a1f698f65a45ea66225\n\nfsync_bdev() export and a bunch of stubs for !CONFIG_BLOCK case had\nbeen left behind\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "a64c8610bd3b753c6aff58f51c04cdf0ae478c18", "tree": "ca9ec4d9ca3d7195521a4004c6a599a35faafe5e", "parents": [ "8e0ee43bc2c3e19db56a4adaa9a9b04ce885cd84" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Mar 27 22:14:10 2009 -0400" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Mar 27 22:14:10 2009 -0400" }, "message": "block_write_full_page: Use synchronous writes for WBC_SYNC_ALL writebacks\n\nWhen doing synchronous writes because wbc-\u003esync_mode is set to\nWBC_SYNC_ALL, send the write request using WRITE_SYNC, so that we\ndon\u0027t unduly block system calls such as fsync().\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "585d3bc06f4ca57f975a5a1f698f65a45ea66225", "tree": "393cbd213e714f00c0cae85851e7837ed88b9a99", "parents": [ "3ba13d179e8c24c68eac32b93593a6b10fcd1572" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Wed Feb 25 10:44:19 2009 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Mar 27 14:44:03 2009 -0400" }, "message": "fs: move bdev code out of buffer.c\n\nMove some block device related code out from buffer.c and put it in\nblock_dev.c. I\u0027m trying to move non-buffer_head code out of buffer.c\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "a8e7d49aa7be728c4ae241a75a2a124cdcabc0c5", "tree": "fec49351496b886d8aa0e7b55e58c2f1394b051a", "parents": [ "68df3755e383e6fecf2354a67b08f92f18536594" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 19 11:32:05 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 19 11:32:05 2009 -0700" }, "message": "Fix race in create_empty_buffers() vs __set_page_dirty_buffers()\n\nNick Piggin noticed this (very unlikely) race between setting a page\ndirty and creating the buffers for it - we need to hold the mapping\nprivate_lock until we\u0027ve set the page dirty bit in order to make sure\nthat create_empty_buffers() might not build up a set of buffers without\nthe dirty bits set when the page is dirty.\n\nI doubt anybody has ever hit this race (and it didn\u0027t solve the issue\nNick was looking at), but as Nick says: \"Still, it does appear to solve\na real race, which we should close.\"\n\nAcked-by: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ba95fd47d177d46743ad94055908d22840370e06", "tree": "f29e6921fefba2728c3b7f6854ac7f7729f602b2", "parents": [ "59af0a0b5848caf38f1bf7013905c3e9cdba4d1d", "be987fdb55a4726e2fcbab7501f89276bdb57288" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 18 18:33:04 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 18 18:33:04 2009 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.dk/linux-2.6-block\n\n* \u0027for-linus\u0027 of git://git.kernel.dk/linux-2.6-block:\n block: fix deadlock in blk_abort_queue() for drivers that readd to timeout list\n block: fix booting from partitioned md array\n block: revert part of 18ce3751ccd488c78d3827e9f6bf54e6322676fb\n cciss: PCI power management reset for kexec\n paride/pg.c: xs(): \u0026\u0026/|| confusion\n fs/bio: bio_alloc_bioset: pass right object ptr to mempool_free\n block: fix bad definition of BIO_RW_SYNC\n bsg: Fix sense buffer bug in SG_IO\n" }, { "commit": "1cf6e7d83bf334cc5916137862c920a97aabc018", "tree": "6bdbc4a98e9b332919beaeba0649797fff544795", "parents": [ "610d18f4128ebbd88845d0fc60cce67b49af881e" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Wed Feb 18 14:48:18 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 18 15:37:54 2009 -0800" }, "message": "mm: task dirty accounting fix\n\nYAMAMOTO-san noticed that task_dirty_inc doesn\u0027t seem to be called properly for\ncases where set_page_dirty is not used to dirty a page (eg. mark_buffer_dirty).\n\nAdditionally, there is some inconsistency about when task_dirty_inc is\ncalled. It is used for dirty balancing, however it even gets called for\n__set_page_dirty_no_writeback.\n\nSo rather than increment it in a set_page_dirty wrapper, move it down to\nexactly where the dirty page accounting stats are incremented.\n\nCc: YAMAMOTO Takashi \u003cyamamoto@valinux.co.jp\u003e\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "78f707bfc723552e8309b7c38a8d0cc51012e813", "tree": "67f47d32eac2cb8288a9469b47c1d8cefc6ce42a", "parents": [ "82eb03cfd862a65363fa2826de0dbd5474cfe5e2" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Feb 17 13:59:08 2009 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Feb 18 10:32:01 2009 +0100" }, "message": "block: revert part of 18ce3751ccd488c78d3827e9f6bf54e6322676fb\n\nThe above commit added WRITE_SYNC and switched various places to using\nthat for committing writes that will be waited upon immediately after\nsubmission. However, this causes a performance regression with AS and CFQ\nfor ext3 at least, since sync_dirty_buffer() will submit some writes with\nWRITE_SYNC while ext3 has sumitted others dependent writes without the sync\nflag set. This causes excessive anticipation/idling in the IO scheduler\nbecause sync and async writes get interleaved, causing a big performance\nregression for the below test case (which is meant to simulate sqlite\nlike behaviour).\n\n---- test case ----\n\nint main(int argc, char **argv)\n{\n\n\tint fdes, i;\n\tFILE *fp;\n\tstruct timeval start;\n\tstruct timeval end;\n\tstruct timeval res;\n\n\tgettimeofday(\u0026start, NULL);\n\tfor (i\u003d0; i\u003cROWS; i++) {\n\t\tfp \u003d fopen(\"test_file\", \"a\");\n\t\tfprintf(fp, \"Some Text Data\\n\");\n\t\tfdes \u003d fileno(fp);\n\t\tfsync(fdes);\n\t\tfclose(fp);\n\t}\n\tgettimeofday(\u0026end, NULL);\n\n\ttimersub(\u0026end, \u0026start, \u0026res);\n\tfprintf(stdout, \"time to write %d lines is %ld(msec)\\n\", ROWS,\n\t\t\t(res.tv_sec*1000000 + res.tv_usec)/1000);\n\n\treturn 0;\n}\n\n-------------------\n\nThanks to Sean.White@APCC.com for tracking down this performance\nregression and providing a test case.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "d4cf109f05ff04c6f5065c3e14165ef01a57dd53", "tree": "898a0d6b4d8934ec4939a8572e1b78dbb9689d93", "parents": [ "6cec50838ed04a9833fb5549f698d3756bbe7e72" ], "author": { "name": "Dave Kleikamp", "email": "shaggy@linux.vnet.ibm.com", "time": "Fri Feb 06 14:59:26 2009 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 06 13:34:22 2009 -0800" }, "message": "vfs: Don\u0027t call attach_nobh_buffers() with an empty list\n\nThis is a modification of a patch by Bill Pemberton \u003cwfp5p@virginia.edu\u003e\n\nnobh_write_end() could call attach_nobh_buffers() with head \u003d\u003d NULL.\nThis would result in a trap when attach_nobh_buffers() attempted to\naccess bh-\u003eb_this_page.\n\nThis can be illustrated by running the writev01 testcase from LTP on jfs.\n\nThis error was introduced by commit 5b41e74a \"vfs: fix data leak in\nnobh_write_end()\". That patch did not take into account that if\nPageMappedToDisk() is true upon entry to nobh_write_begin(), then no\nbuffers will be allocated for the page. In that case, we won\u0027t have to\nworry about a failed write leaving unitialized data in the page.\n\nOf course, head !\u003d NULL implies !page_has_buffers(page), so no need to\ntest both.\n\nSigned-off-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nCc: Bill Pemberton \u003cwfp5p@virginia.edu\u003e\nCc: Dmitri Monakhov \u003cdmonakhov@openvz.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "bdc480e3bef6eb0e7071770834cbdda7e30a5436", "tree": "2bfea8d2b61ccedc3edea4f78e3250eea7f770cc", "parents": [ "a5f8fa9e9ba5ef3305e147f41ad6e1e84ac1f0bd" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:14:12 2009 +0100" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Jan 14 14:15:22 2009 +0100" }, "message": "[CVE-2009-0029] System call wrappers part 10\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n" }, { "commit": "fcccf502540e3d752d33b2d8e976034dee81f9f7", "tree": "401be4ad2fc3c535a2fc77e413b1be4a736f320c", "parents": [ "c4be0c1dc4cdc37b175579be1460f15ac6495e9a" ], "author": { "name": "Takashi Sato", "email": "t-sato@yk.jp.nec.com", "time": "Fri Jan 09 16:40:59 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 09 16:54:42 2009 -0800" }, "message": "filesystem freeze: implement generic freeze feature\n\nThe ioctls for the generic freeze feature are below.\no Freeze the filesystem\n int ioctl(int fd, int FIFREEZE, arg)\n fd: The file descriptor of the mountpoint\n FIFREEZE: request code for the freeze\n arg: Ignored\n Return value: 0 if the operation succeeds. Otherwise, -1\n\no Unfreeze the filesystem\n int ioctl(int fd, int FITHAW, arg)\n fd: The file descriptor of the mountpoint\n FITHAW: request code for unfreeze\n arg: Ignored\n Return value: 0 if the operation succeeds. Otherwise, -1\n Error number: If the filesystem has already been unfrozen,\n errno is set to EINVAL.\n\n[akpm@linux-foundation.org: fix CONFIG_BLOCK\u003dn]\nSigned-off-by: Takashi Sato \u003ct-sato@yk.jp.nec.com\u003e\nSigned-off-by: Masayuki Hamaguchi \u003cm-hamaguchi@ys.jp.nec.com\u003e\nCc: \u003cxfs-masters@oss.sgi.com\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nCc: Dave Chinner \u003cdavid@fromorbit.com\u003e\nCc: Alasdair G Kergon \u003cagk@redhat.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c4be0c1dc4cdc37b175579be1460f15ac6495e9a", "tree": "716ea88318211ed27cadcebda0fd85c1f8246edb", "parents": [ "69347a236b22c3962ea812511495e502dedfd50c" ], "author": { "name": "Takashi Sato", "email": "t-sato@yk.jp.nec.com", "time": "Fri Jan 09 16:40:58 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 09 16:54:42 2009 -0800" }, "message": "filesystem freeze: add error handling of write_super_lockfs/unlockfs\n\nCurrently, ext3 in mainline Linux doesn\u0027t have the freeze feature which\nsuspends write requests. So, we cannot take a backup which keeps the\nfilesystem\u0027s consistency with the storage device\u0027s features (snapshot and\nreplication) while it is mounted.\n\nIn many case, a commercial filesystem (e.g. VxFS) has the freeze feature\nand it would be used to get the consistent backup.\n\nIf Linux\u0027s standard filesystem ext3 has the freeze feature, we can do it\nwithout a commercial filesystem.\n\nSo I have implemented the ioctls of the freeze feature.\nI think we can take the consistent backup with the following steps.\n1. Freeze the filesystem with the freeze ioctl.\n2. Separate the replication volume or create the snapshot\n with the storage device\u0027s feature.\n3. Unfreeze the filesystem with the unfreeze ioctl.\n4. Take the backup from the separated replication volume\n or the snapshot.\n\nThis patch:\n\nVFS:\nChanged the type of write_super_lockfs and unlockfs from \"void\"\nto \"int\" so that they can return an error.\nRename write_super_lockfs and unlockfs of the super block operation\nfreeze_fs and unfreeze_fs to avoid a confusion.\n\next3, ext4, xfs, gfs2, jfs:\nChanged the type of write_super_lockfs and unlockfs from \"void\"\nto \"int\" so that write_super_lockfs returns an error if needed,\nand unlockfs always returns 0.\n\nreiserfs:\nChanged the type of write_super_lockfs and unlockfs from \"void\"\nto \"int\" so that they always return 0 (success) to keep a current behavior.\n\nSigned-off-by: Takashi Sato \u003ct-sato@yk.jp.nec.com\u003e\nSigned-off-by: Masayuki Hamaguchi \u003cm-hamaguchi@ys.jp.nec.com\u003e\nCc: \u003cxfs-masters@oss.sgi.com\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nCc: Dave Chinner \u003cdavid@fromorbit.com\u003e\nCc: Alasdair G Kergon \u003cagk@redhat.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "69e9930993cfd70d82c8d9dd96fc3a88854d06fc", "tree": "2467f8e1fdb745f4ca5e94acf0aba09c74063125", "parents": [ "4779280d1ea4d361af13ae77ba55217fbcd16d4c" ], "author": { "name": "Franck Bui-Huu", "email": "fbuihuu@gmail.com", "time": "Tue Jan 06 14:40:19 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 06 15:59:08 2009 -0800" }, "message": "block_write_begin(): remove useless goto\n\nSigned-off-by: Franck Bui-Huu \u003cfbuihuu@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "54566b2c1594c2326a645a3551f9d989f7ba3c5e", "tree": "b373f3283fe5e197d0df29cd6b645c35adf1076c", "parents": [ "e687d691cb3790d25e31c74f5941fd7c565e9df5" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Sun Jan 04 12:00:53 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jan 04 13:33:20 2009 -0800" }, "message": "fs: symlink write_begin allocation context fix\n\nWith the write_begin/write_end aops, page_symlink was broken because it\ncould no longer pass a GFP_NOFS type mask into the point where the\nallocations happened. They are done in write_begin, which would always\nassume that the filesystem can be entered from reclaim. This bug could\ncause filesystem deadlocks.\n\nThe funny thing with having a gfp_t mask there is that it doesn\u0027t really\nallow the caller to arbitrarily tinker with the context in which it can be\ncalled. It couldn\u0027t ever be GFP_ATOMIC, for example, because it needs to\ntake the page lock. The only thing any callers care about is __GFP_FS\nanyway, so turn that into a single flag.\n\nAdd a new flag for write_begin, AOP_FLAG_NOFS. Filesystems can now act on\nthis flag in their write_begin function. Change __grab_cache_page to\naccept a nofs argument as well, to honour that flag (while we\u0027re there,\nchange the name to grab_cache_page_write_begin which is more instructive\nand does away with random leading underscores).\n\nThis is really a more flexible way to go in the end anyway -- if a\nfilesystem happens to want any extra allocations aside from the pagecache\nones in ints write_begin function, it may now use GFP_KERNEL (rather than\nGFP_NOFS) for common case allocations (eg. ocfs2_alloc_write_ctxt, for a\nrandom example).\n\n[kosaki.motohiro@jp.fujitsu.com: fix ubifs]\n[kosaki.motohiro@jp.fujitsu.com: fix fuse]\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: \u003cstable@kernel.org\u003e\t\t[2.6.28.x]\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n[ Cleaned up the calling convention: just pass in the AOP flags\n untouched to the grab_cache_page_write_begin() function. That\n just simplifies everybody, and may even allow future expansion of the\n logic. - Linus ]\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "08bafc0341f2f7920e9045bc32c40299cac8c21b", "tree": "ab4ad0f73f457409239a98c050bf7edd47c03cf1", "parents": [ "7c239517d9f18427fc2e7ed259fb3b866595f5af" ], "author": { "name": "Keith Mannthey", "email": "kmannth@us.ibm.com", "time": "Tue Nov 25 10:24:35 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:44 2008 +0100" }, "message": "block: Supress Buffer I/O errors when SCSI REQ_QUIET flag set\n\nAllow the scsi request REQ_QUIET flag to be propagated to the buffer\nfile system layer. The basic ideas is to pass the flag from the scsi\nrequest to the bio (block IO) and then to the buffer layer. The buffer\nlayer can then suppress needless printks.\n\nThis patch declutters the kernel log by removed the 40-50 (per lun)\nbuffer io error messages seen during a boot in my multipath setup . It\nis a good chance any real errors will be missed in the \"noise\" it the\nlogs without this patch.\n\nDuring boot I see blocks of messages like\n\"\n__ratelimit: 211 callbacks suppressed\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242847\nBuffer I/O error on device sdm, logical block 1\nBuffer I/O error on device sdm, logical block 5242878\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242872\n\"\nin my logs.\n\nMy disk environment is multipath fiber channel using the SCSI_DH_RDAC\ncode and multipathd. This topology includes an \"active\" and \"ghost\"\npath for each lun. IO\u0027s to the \"ghost\" path will never complete and the\nSCSI layer, via the scsi device handler rdac code, quick returns the IOs\nto theses paths and sets the REQ_QUIET scsi flag to suppress the scsi\nlayer messages.\n\n I am wanting to extend the QUIET behavior to include the buffer file\nsystem layer to deal with these errors as well. I have been running this\npatch for a while now on several boxes without issue. A few runs of\nbonnie++ show no noticeable difference in performance in my setup.\n\nThanks for John Stultz for the quiet_error finalization.\n\nSubmitted-by: Keith Mannthey \u003ckmannth@us.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "52b19ac993f1aeadbce15b55302be9a35346e235", "tree": "90695f8fdaa97d3a30b1e5f82016fb1cce3da16b", "parents": [ "ed313489badef16d700f5a3be50e8fd8f8294bc8" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Tue Sep 23 18:24:08 2008 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Thu Nov 27 17:38:28 2008 +0100" }, "message": "udf: Fix BUG_ON() in destroy_inode()\n\nudf_clear_inode() can leave behind buffers on mapping\u0027s i_private list (when\nwe truncated preallocation). Call invalidate_inode_buffers() so that the list\nis properly cleaned-up before we return from udf_clear_inode(). This is ugly\nand suggest that we should cleanup preallocation earlier than in clear_inode()\nbut currently there\u0027s no such call available since drop_inode() is called under\ninode lock and thus is unusable for disk operations.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "51b07fc3c5c830bb49c80fc5eac041e1f66a72e7", "tree": "7d01d39f5eb5ac601d95c1e133492bd83135af6c", "parents": [ "8413ac9d8c9a1366a4f57880723126cd24e5a5c3" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Sat Oct 18 20:27:00 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 20 08:52:32 2008 -0700" }, "message": "fs: buffer lock use lock bitops\n\ntrylock_buffer and unlock_buffer open and close a critical section.\nHence, we can use the lock bitops to get the desired memory ordering.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "48fd4f93a00eac844678629f2f00518e146ed30d", "tree": "c71f6792742886db9912078c70b3e8806e9f260f", "parents": [ "a4a778971b9cfcbedb8648ba502d801f7db04c47" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Aug 22 10:00:36 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Aug 27 09:50:19 2008 +0200" }, "message": "block: submit_bh() inadvertently discards barrier flag on a sync write\n\nReported by Milan Broz \u003cmbroz@redhat.com\u003e, commit 18ce3751 inadvertently\nmade submit_bh() discard the barrier bit for a WRITE_SYNC request. Fix\nthat up.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "ca5de404ff036a29b25e9a83f6919c9f606c5841", "tree": "474da867c4d4086313cee90cdc3560bb17fade96", "parents": [ "529ae9aaa08378cfe2a4350bded76f32cc8ff0ce" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Sat Aug 02 12:02:13 2008 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Aug 04 21:56:09 2008 -0700" }, "message": "fs: rename buffer trylock\n\nLike the page lock change, this also requires name change, so convert the\nraw test_and_set bitop to a trylock.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "dbacefc9c4f6bd365243db379473ab7041656d90", "tree": "01bcdbe23ea7008971161366dab56bfc47b2049a", "parents": [ "d406f66ddb0d7491ddd53e4600c425d76a8a245f" ], "author": { "name": "Thomas Petazzoni", "email": "thomas.petazzoni@free-electrons.com", "time": "Tue Jul 29 22:33:47 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jul 30 09:41:46 2008 -0700" }, "message": "fs/buffer.c: uninline __remove_assoc_queue()\n\nUninline the __remove_assoc_queue() function in fs/buffer.c, called at too\nmany places and too long to really be inlined. Size results:\n\n text\t data\t bss\t dec\t hex\tfilename\n1134606\t 118840\t 212992\t1466438\t 166046\tvmlinux.old\n1134303\t 118840\t 212992\t1466135\t 165f17\tvmlinux\n -303 0 0 -303 -12F +/-\n\nThis patch is part of the Linux Tiny project and has been originally\nwritten by Matt Mackall \u003cmpm@selenic.com\u003e.\n\nSigned-off-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nCc: Matt Mackall \u003cmpm@selenic.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8ab22b9abb5c55413802e4adc9aa6223324547c3", "tree": "cff3319e1275e8a7c083d492889ec6bd0c7712d3", "parents": [ "d84a52f62f6a396ed77aa0052da74ca9e760b28a" ], "author": { "name": "Hisashi Hifumi", "email": "hifumi.hisashi@oss.ntt.co.jp", "time": "Mon Jul 28 15:46:36 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 28 16:30:21 2008 -0700" }, "message": "vfs: pagecache usage optimization for pagesize!\u003dblocksize\n\nWhen we read some part of a file through pagecache, if there is a\npagecache of corresponding index but this page is not uptodate, read IO\nis issued and this page will be uptodate.\n\nI think this is good for pagesize \u003d\u003d blocksize environment but there is\nroom for improvement on pagesize !\u003d blocksize environment. Because in\nthis case a page can have multiple buffers and even if a page is not\nuptodate, some buffers can be uptodate.\n\nSo I suggest that when all buffers which correspond to a part of a file\nthat we want to read are uptodate, use this pagecache and copy data from\nthis pagecache to user buffer even if a page is not uptodate. This can\nreduce read IO and improve system throughput.\n\nI wrote a benchmark program and got result number with this program.\n\nThis benchmark do:\n\n 1: mount and open a test file.\n\n 2: create a 512MB file.\n\n 3: close a file and umount.\n\n 4: mount and again open a test file.\n\n 5: pwrite randomly 300000 times on a test file. offset is aligned\n by IO size(1024bytes).\n\n 6: measure time of preading randomly 100000 times on a test file.\n\nThe result was:\n\t2.6.26\n 330 sec\n\n\t2.6.26-patched\n 226 sec\n\nArch:i386\nFilesystem:ext3\nBlocksize:1024 bytes\nMemory: 1GB\n\nOn ext3/4, a file is written through buffer/block. So random read/write\nmixed workloads or random read after random write workloads are optimized\nwith this patch under pagesize !\u003d blocksize environment. This test result\nshowed this.\n\nThe benchmark program is as follows:\n\n#include \u003cstdio.h\u003e\n#include \u003csys/types.h\u003e\n#include \u003csys/stat.h\u003e\n#include \u003cfcntl.h\u003e\n#include \u003cunistd.h\u003e\n#include \u003ctime.h\u003e\n#include \u003cstdlib.h\u003e\n#include \u003cstring.h\u003e\n#include \u003csys/mount.h\u003e\n\n#define LEN 1024\n#define LOOP 1024*512 /* 512MB */\n\nmain(void)\n{\n\tunsigned long i, offset, filesize;\n\tint fd;\n\tchar buf[LEN];\n\ttime_t t1, t2;\n\n\tif (mount(\"/dev/sda1\", \"/root/test1/\", \"ext3\", 0, 0) \u003c 0) {\n\t\tperror(\"cannot mount\\n\");\n\t\texit(1);\n\t}\n\tmemset(buf, 0, LEN);\n\tfd \u003d open(\"/root/test1/testfile\", O_CREAT|O_RDWR|O_TRUNC);\n\tif (fd \u003c 0) {\n\t\tperror(\"cannot open file\\n\");\n\t\texit(1);\n\t}\n\tfor (i \u003d 0; i \u003c LOOP; i++)\n\t\twrite(fd, buf, LEN);\n\tclose(fd);\n\tif (umount(\"/root/test1/\") \u003c 0) {\n\t\tperror(\"cannot umount\\n\");\n\t\texit(1);\n\t}\n\tif (mount(\"/dev/sda1\", \"/root/test1/\", \"ext3\", 0, 0) \u003c 0) {\n\t\tperror(\"cannot mount\\n\");\n\t\texit(1);\n\t}\n\tfd \u003d open(\"/root/test1/testfile\", O_RDWR);\n\tif (fd \u003c 0) {\n\t\tperror(\"cannot open file\\n\");\n\t\texit(1);\n\t}\n\n\tfilesize \u003d LEN * LOOP;\n\tfor (i \u003d 0; i \u003c 300000; i++){\n\t\toffset \u003d (random() % filesize) \u0026 (~(LEN - 1));\n\t\tpwrite(fd, buf, LEN, offset);\n\t}\n\tprintf(\"start test\\n\");\n\ttime(\u0026t1);\n\tfor (i \u003d 0; i \u003c 100000; i++){\n\t\toffset \u003d (random() % filesize) \u0026 (~(LEN - 1));\n\t\tpread(fd, buf, LEN, offset);\n\t}\n\ttime(\u0026t2);\n\tprintf(\"%ld sec\\n\", t2-t1);\n\tclose(fd);\n\tif (umount(\"/root/test1/\") \u003c 0) {\n\t\tperror(\"cannot umount\\n\");\n\t\texit(1);\n\t}\n}\n\nSigned-off-by: Hisashi Hifumi \u003chifumi.hisashi@oss.ntt.co.jp\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nCc: Jan Kara \u003cjack@ucw.cz\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5c752ad9f35910ff1912b3f3ae82878178ddc432", "tree": "4ac627f9cdd7f0e1b0557b9b5c4838587c1c0890", "parents": [ "5cd2b459d326a424671dcd95f038649f7bf7cb96" ], "author": { "name": "Arjan van de Ven", "email": "arjan@linux.intel.com", "time": "Fri Jul 25 19:45:40 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:07 2008 -0700" }, "message": "Use WARN() in fs/\n\nUse WARN() instead of a printk+WARN_ON() pair; this way the message\nbecomes part of the warning section for better reporting/collection.\n\nSigned-off-by: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "51cc50685a4275c6a02653670af9f108a64e01cf", "tree": "819d47bd2b0c8a9d1835d863853804b0a0242b97", "parents": [ "d91958815d214ea365b98cbff6215383897edcb6" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Fri Jul 25 19:45:34 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:07 2008 -0700" }, "message": "SL*B: drop kmem cache argument from constructor\n\nKmem cache passed to constructor is only needed for constructors that are\nthemselves multiplexeres. Nobody uses this \"feature\", nor does anybody uses\npassed kmem cache in non-trivial way, so pass only pointer to object.\n\nNon-trivial places are:\n\tarch/powerpc/mm/init_64.c\n\tarch/powerpc/mm/hugetlbpage.c\n\nThis is flag day, yes.\n\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nAcked-by: Pekka Enberg \u003cpenberg@cs.helsinki.fi\u003e\nAcked-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Jon Tollefson \u003ckniht@linux.vnet.ibm.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nCc: Matt Mackall \u003cmpm@selenic.com\u003e\n[akpm@linux-foundation.org: fix arch/powerpc/mm/hugetlbpage.c]\n[akpm@linux-foundation.org: fix mm/slab.c]\n[akpm@linux-foundation.org: fix ubifs]\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "19fd6231279be3c3bdd02ed99f9b0eb195978064", "tree": "ee09121054262d73c551b57114acd855b82a7a82", "parents": [ "a60637c85893e7191faaafa6a72e197c24386727" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Fri Jul 25 19:45:32 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:06 2008 -0700" }, "message": "mm: spinlock tree_lock\n\nmapping-\u003etree_lock has no read lockers. convert the lock from an rwlock\nto a spinlock.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@us.ibm.com\u003e\nReviewed-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1a781a777b2f6ac46523fe92396215762ced624d", "tree": "4f34bb4aade85c0eb364b53d664ec7f6ab959006", "parents": [ "b9d2252c1e44fa83a4e65fdc9eb93db6297c55af", "42a2f217a5e324ed5f2373ab1b7a0a15187c4d6c" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Jul 15 21:55:59 2008 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Tue Jul 15 21:55:59 2008 +0200" }, "message": "Merge branch \u0027generic-ipi\u0027 into generic-ipi-for-linus\n\nConflicts:\n\n\tarch/powerpc/Kconfig\n\tarch/s390/kernel/time.c\n\tarch/x86/kernel/apic_32.c\n\tarch/x86/kernel/cpu/perfctr-watchdog.c\n\tarch/x86/kernel/i8259_64.c\n\tarch/x86/kernel/ldt.c\n\tarch/x86/kernel/nmi_64.c\n\tarch/x86/kernel/smpboot.c\n\tarch/x86/xen/smp.c\n\tinclude/asm-x86/hw_irq_32.h\n\tinclude/asm-x86/hw_irq_64.h\n\tinclude/asm-x86/mach-default/irq_vectors.h\n\tinclude/asm-x86/mach-voyager/irq_vectors.h\n\tinclude/asm-x86/smp.h\n\tkernel/Makefile\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "29a814d2ee0e43c2980f33f91c1311ec06c0aa35", "tree": "dfc9d45fc3194237192b6cde1069faa70fe4c260", "parents": [ "87c89c232c8f7b3820c33c3b9bc803e9358027da" ], "author": { "name": "Alex Tomas", "email": "alex@clusterfs.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": "vfs: add hooks for ext4\u0027s delayed allocation support\n\nExport mpage_bio_submit() and __mpage_writepage() for the benefit of\next4\u0027s delayed allocation support. Also change __block_write_full_page\nso that if buffers that have the BH_Delay flag set it will call\nget_block() to get the physical block allocated, just as in the\n!BH_Mapped case.\n\nSigned-off-by: Alex Tomas \u003calex@clusterfs.com\u003e\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\n\n" }, { "commit": "c7d206b3379f7d6462e778b74f475c470ee3dcaf", "tree": "71d11797c72e2e4ee04e6690c048fd26f25ae90d", "parents": [ "2e9ee850355593e311d9a26542290fe51e152f74" ], "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": "vfs: Move mark_inode_dirty() from under page lock in generic_write_end()\n\nThere\u0027s no need to call mark_inode_dirty() under page lock in\ngeneric_write_end(). It unnecessarily makes hold time of page lock longer\nand more importantly it forces locking order of page lock and transaction\nstart for journaling filesystems.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\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\n" }, { "commit": "18ce3751ccd488c78d3827e9f6bf54e6322676fb", "tree": "4bb83c2b963e8ebe918b79f61d1a440fb1d28f8d", "parents": [ "d585d0b9d73ed999cc7b8cf3cac4a5b01abb544e" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Jul 01 09:07:34 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Jul 01 09:07:34 2008 +0200" }, "message": "Properly notify block layer of sync writes\n\nfsync_buffers_list() and sync_dirty_buffer() both issue async writes and\nthen immediately wait on them. Conceptually, that makes them sync writes\nand we should treat them as such so that the IO schedulers can handle\nthem appropriately.\n\nThis patch fixes a write starvation issue that Lin Ming reported, where\nxx is stuck for more than 2 minutes because of a large number of\nsynchronous IO in the system:\n\nINFO: task kjournald:20558 blocked for more than 120 seconds.\n\"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this\nmessage.\nkjournald D ffff810010820978 6712 20558 2\nffff81022ddb1d10 0000000000000046 ffff81022e7baa10 ffffffff803ba6f2\nffff81022ecd0000 ffff8101e6dc9160 ffff81022ecd0348 000000008048b6cb\n0000000000000086 ffff81022c4e8d30 0000000000000000 ffffffff80247537\nCall Trace:\n[\u003cffffffff803ba6f2\u003e] kobject_get+0x12/0x17\n[\u003cffffffff80247537\u003e] getnstimeofday+0x2f/0x83\n[\u003cffffffff8029c1ac\u003e] sync_buffer+0x0/0x3f\n[\u003cffffffff8066d195\u003e] io_schedule+0x5d/0x9f\n[\u003cffffffff8029c1e7\u003e] sync_buffer+0x3b/0x3f\n[\u003cffffffff8066d3f0\u003e] __wait_on_bit+0x40/0x6f\n[\u003cffffffff8029c1ac\u003e] sync_buffer+0x0/0x3f\n[\u003cffffffff8066d48b\u003e] out_of_line_wait_on_bit+0x6c/0x78\n[\u003cffffffff80243909\u003e] wake_bit_function+0x0/0x23\n[\u003cffffffff8029e3ad\u003e] sync_dirty_buffer+0x98/0xcb\n[\u003cffffffff8030056b\u003e] journal_commit_transaction+0x97d/0xcb6\n[\u003cffffffff8023a676\u003e] lock_timer_base+0x26/0x4b\n[\u003cffffffff8030300a\u003e] kjournald+0xc1/0x1fb\n[\u003cffffffff802438db\u003e] autoremove_wake_function+0x0/0x2e\n[\u003cffffffff80302f49\u003e] kjournald+0x0/0x1fb\n[\u003cffffffff802437bb\u003e] kthread+0x47/0x74\n[\u003cffffffff8022de51\u003e] schedule_tail+0x28/0x5d\n[\u003cffffffff8020cac8\u003e] child_rip+0xa/0x12\n[\u003cffffffff80243774\u003e] kthread+0x0/0x74\n[\u003cffffffff8020cabe\u003e] child_rip+0x0/0x12\n\nLin Ming confirms that this patch fixes the issue. I\u0027ve run tests with\nit for the past week and no ill effects have been observed, so I\u0027m\nproposing it for inclusion into 2.6.26.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "15c8b6c1aaaf1c4edd67e2f02e4d8e1bd1a51c0d", "tree": "3658f893c2f89ea0be4c6cc08aa11fa54476d0f4", "parents": [ "8691e5a8f691cc2a4fda0651e8d307aaba0e7d68" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri May 09 09:39:44 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Jun 26 11:24:38 2008 +0200" }, "message": "on_each_cpu(): kill unused \u0027retry\u0027 parameter\n\nIt\u0027s not even passed on to smp_call_function() anymore, since that\nwas removed. So kill it.\n\nAcked-by: Jeremy Fitzhardinge \u003cjeremy.fitzhardinge@citrix.com\u003e\nReviewed-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "8e24eea728068bbeb6a3c500b848f883a20bf225", "tree": "93e79da649723e2766237505b22725fec395f139", "parents": [ "530b6412786d7f83592c1a8e2445541ed73fca76" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Wed Apr 30 00:55:09 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:54 2008 -0700" }, "message": "fs: replace remaining __FUNCTION__ occurrences\n\n__FUNCTION__ is gcc-specific, use __func__\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f1e3af72c10ba74fb15864c354515ec1bd8bf2a5", "tree": "faf2d2f8b0264fd175acf8ed5623c74e0da908ec", "parents": [ "946a57b526a16e5662235cb8f573337bc8ecdc48" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Tue Apr 29 00:59:01 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 29 08:06:01 2008 -0700" }, "message": "make fs/buffer.c:cont_expand_zero() static\n\ncont_expand_zero() can become static.\n\nSigned-off-by: Adrian Bunk \u003cbunk@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": "946a57b526a16e5662235cb8f573337bc8ecdc48", "tree": "e7805833cf724d2bf2b5ebfe0fc7bb4d8cd2c321", "parents": [ "45cc2b96f20fa27088a650587e5d9dc5fa5e32c0" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Tue Apr 29 00:59:00 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 29 08:06:01 2008 -0700" }, "message": "remove generic_commit_write()\n\nRemove the obsolete and no longer used generic_commit_write().\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "061e97469f46f924cf14bbf1dd4805b46986691a", "tree": "f89bae50777ff9d7a8dc917342f55218de132166", "parents": [ "e69be4c9c4ac1b6bb6380512557da0afa6262e98" ], "author": { "name": "OGAWA Hirofumi", "email": "hirofumi@mail.parknet.co.jp", "time": "Mon Apr 28 02:16:28 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:47 2008 -0700" }, "message": "Add balance_dirty_pages_ratelimited() to cont_expand_zero()\n\nOn the systems, ftruncate() which expand size for FAT became the cause\nof OOM. The cont_expand_zero() filled all memory with dirty pages,\nand since disk is very slow, limit of page scanning was exceeded, then\nit triggered OOM.\n\nThis adds balance_dirty_pages_ratelimited() to avoid filling memory\nwith dirty pages.\n\nSigned-off-by: OGAWA Hirofumi \u003chirofumi@mail.parknet.co.jp\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "19770b32609b6bf97a3dece2529089494cbfc549", "tree": "3b5922d1b20aabdf929bde9309f323841717747a", "parents": [ "dd1a239f6f2d4d3eedd318583ec319aa145b324c" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Mon Apr 28 02:12:18 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:19 2008 -0700" }, "message": "mm: filter based on a nodemask as well as a gfp_mask\n\nThe MPOL_BIND policy creates a zonelist that is used for allocations\ncontrolled by that mempolicy. As the per-node zonelist is already being\nfiltered based on a zone id, this patch adds a version of __alloc_pages() that\ntakes a nodemask for further filtering. This eliminates the need for\nMPOL_BIND to create a custom zonelist.\n\nA positive benefit of this is that allocations using MPOL_BIND now use the\nlocal node\u0027s distance-ordered zonelist instead of a custom node-id-ordered\nzonelist. I.e., pages will be allocated from the closest allowed node with\navailable memory.\n\n[Lee.Schermerhorn@hp.com: Mempolicy: update stale documentation and comments]\n[Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask]\n[Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask rework]\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "dd1a239f6f2d4d3eedd318583ec319aa145b324c", "tree": "aff4224c96b5e2e67588c3946858a724863eeaf9", "parents": [ "54a6eb5c4765aa573a030ceeba2c14e3d2ea5706" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Mon Apr 28 02:12:17 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:18 2008 -0700" }, "message": "mm: have zonelist contains structs with both a zone pointer and zone_idx\n\nFiltering zonelists requires very frequent use of zone_idx(). This is costly\nas it involves a lookup of another structure and a substraction operation. As\nthe zone_idx is often required, it should be quickly accessible. The node idx\ncould also be stored here if it was found that accessing zone-\u003enode is\nsignificant which may be the case on workloads where nodemasks are heavily\nused.\n\nThis patch introduces a struct zoneref to store a zone pointer and a zone\nindex. The zonelist then consists of an array of these struct zonerefs which\nare looked up as necessary. Helpers are given for accessing the zone index as\nwell as the node index.\n\n[kamezawa.hiroyu@jp.fujitsu.com: Suggested struct zoneref instead of embedding information in pointers]\n[hugh@veritas.com: mm-have-zonelist: fix memcg ooms]\n[hugh@veritas.com: just return do_try_to_free_pages]\n[hugh@veritas.com: do_try_to_free_pages gfp_mask redundant]\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nAcked-by: David Rientjes \u003crientjes@google.com\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "54a6eb5c4765aa573a030ceeba2c14e3d2ea5706", "tree": "547176a090beb787722a153cf2b8b942dc0e68db", "parents": [ "18ea7e710d2452fa726814a406779188028cf1bf" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Mon Apr 28 02:12:16 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:18 2008 -0700" }, "message": "mm: use two zonelist that are filtered by GFP mask\n\nCurrently a node has two sets of zonelists, one for each zone type in the\nsystem and a second set for GFP_THISNODE allocations. Based on the zones\nallowed by a gfp mask, one of these zonelists is selected. All of these\nzonelists consume memory and occupy cache lines.\n\nThis patch replaces the multiple zonelists per-node with two zonelists. The\nfirst contains all populated zones in the system, ordered by distance, for\nfallback allocations when the target/preferred node has no free pages. The\nsecond contains all populated zones in the node suitable for GFP_THISNODE\nallocations.\n\nAn iterator macro is introduced called for_each_zone_zonelist() that interates\nthrough each zone allowed by the GFP flags in the selected zonelist.\n\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0e88460da6ab7bb6a7ef83675412ed5b6315d741", "tree": "1feb4de2362e4998a0deeab66af1efb9c7b8bb34", "parents": [ "dac1d27bc8d5ca636d3014ecfdf94407031d1970" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Mon Apr 28 02:12:14 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:18 2008 -0700" }, "message": "mm: introduce node_zonelist() for accessing the zonelist for a GFP mask\n\nIntroduce a node_zonelist() helper function. It is used to lookup the\nappropriate zonelist given a node and a GFP mask. The patch on its own is a\ncleanup but it helps clarify parts of the two-zonelist-per-node patchset. If\nnecessary, it can be merged with the next patch in this set without problems.\n\nReviewed-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "dac1d27bc8d5ca636d3014ecfdf94407031d1970", "tree": "2d92f6533f91da05c5f3d61e314f36b9313be89a", "parents": [ "9d02dbc8137759e4c2f91db0b7f9c8a1ec2a9276" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Mon Apr 28 02:12:12 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:18 2008 -0700" }, "message": "mm: use zonelists instead of zones when direct reclaiming pages\n\nThe following patches replace multiple zonelists per node with two zonelists\nthat are filtered based on the GFP flags. The patches as a set fix a bug with\nregard to the use of MPOL_BIND and ZONE_MOVABLE. With this patchset, the\nMPOL_BIND will apply to the two highest zones when the highest zone is\nZONE_MOVABLE. This should be considered as an alternative fix for the\nMPOL_BIND+ZONE_MOVABLE in 2.6.23 to the previously discussed hack that filters\nonly custom zonelists.\n\nThe first patch cleans up an inconsistency where direct reclaim uses\nzonelist-\u003ezones where other places use zonelist.\n\nThe second patch introduces a helper function node_zonelist() for looking up\nthe appropriate zonelist for a GFP mask which simplifies patches later in the\nset.\n\nThe third patch defines/remembers the \"preferred zone\" for numa statistics, as\nit is no longer always the first zone in a zonelist.\n\nThe forth patch replaces multiple zonelists with two zonelists that are\nfiltered. The two zonelists are due to the fact that the memoryless patchset\nintroduces a second set of zonelists for __GFP_THISNODE.\n\nThe fifth patch introduces helper macros for retrieving the zone and node\nindices of entries in a zonelist.\n\nThe final patch introduces filtering of the zonelists based on a nodemask.\nTwo zonelists exist per node, one for normal allocations and one for\n__GFP_THISNODE.\n\nPerformance results varied depending on the machine configuration. In real\nworkloads the gain/loss will depend on how much the userspace portion of the\nbenchmark benefits from having more cache available due to reduced referencing\nof zonelists.\n\nThese are the range of performance losses/gains when running against\n2.6.24-rc4-mm1. The set and these machines are a mix of i386, x86_64 and\nppc64 both NUMA and non-NUMA.\n\t\t\t loss to gain\nTotal CPU time on Kernbench: -0.86% to 1.13%\nElapsed time on Kernbench: -0.79% to 0.76%\npage_test from aim9: -4.37% to 0.79%\nbrk_test from aim9: -0.71% to 4.07%\nfork_test from aim9: -1.84% to 4.60%\nexec_test from aim9: -0.71% to 1.08%\n\nThis patch:\n\nThe allocator deals with zonelists which indicate the order in which zones\nshould be targeted for an allocation. Similarly, direct reclaim of pages\niterates over an array of zones. For consistency, this patch converts direct\nreclaim to use a zonelist. No functionality is changed by this patch. This\nsimplifies zonelist iterators in the next patch.\n\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "488514d1798289f56f80ed018e246179fe500383", "tree": "e68d9f703dcbceed0cc08b03014d109d7ec3acd4", "parents": [ "e92adcba261fd391591bb63c1703185a04a41554" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Mon Apr 28 02:12:05 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:17 2008 -0700" }, "message": "Remove set_migrateflags()\n\nMigrate flags must be set on slab creation as agreed upon when the antifrag\nlogic was reviewed. Otherwise some slabs of a slabcache will end up in the\nunmovable and others in the reclaimable section depending on which flag was\nactive when a new slab page was allocated.\n\nThis likely slid in somehow when antifrag was merged. Remove it.\n\nThe buffer_heads are always allocated with __GFP_RECLAIMABLE because the\nSLAB_RECLAIM_ACCOUNT option is set. The set_migrateflags() never had any\neffect there.\n\nRadix tree allocations are not directly reclaimable but they are allocated\nwith __GFP_RECLAIMABLE set on each allocation. We now set\nSLAB_RECLAIM_ACCOUNT on radix tree slab creation making sure that radix\ntree slabs are consistently placed in the reclaimable section. Radix tree\nslabs will also be accounted as such.\n\nThere is then no user left of set_migratepages. So remove it.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.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\n" }, { "commit": "1be62dc190ebaca331038962c873e7967de6cc4b", "tree": "a8eb0a106bf362819d2bb0fc602b0e52df6a4198", "parents": [ "4ed919014eb2b591eb8fdd4dd00226a65faddef4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 04 14:38:17 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 04 14:38:17 2008 -0700" }, "message": "Be more careful about marking buffers dirty\n\nMikulas Patocka noted that the optimization where we check if a buffer\nwas already dirty (and we avoid re-dirtying it) was not really SMP-safe.\n\nSince the read of the old status was not synchronized with anything, an\naggressive CPU re-ordering of memory accesses might have moved that read\nup to before the data was even written to the buffer, and another CPU\nthat cleaned it again, causing the newly dirty state to never actually\nhit the disk.\n\nAdmittedly this would probably never trigger in practice, but it\u0027s still\nwrong.\n\nMikulas sent a patch that fixed the problem, but I dislike the subtlety\nof the whole optimization, so this is an alternate fix that is more\nexplicit about the particular SMP ordering for the optimization, and\nseparates out the speculative reads of the buffer state into its own\nconditional (and makes the memory barrier only happen if we are likely\nto actually hit the optimized case in the first place).\n\nI considered removing the optimization entirely, but Andrew argued for\nit\u0027s continued existence. I\u0027m a push-over.\n\nCc: Mikulas Patocka \u003cmikulas@artax.karlin.mff.cuni.cz\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5b41e74ad1b0bf7bc51765ae74e5dc564afc3e48", "tree": "e0edafb1d44154b258e013f745d123326183acbd", "parents": [ "1d4a788f15302877ff2cb08f22009e290a36a209" ], "author": { "name": "Dmitri Monakhov", "email": "dmonakhov@openvz.org", "time": "Fri Mar 28 14:15:52 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 28 14:45:21 2008 -0700" }, "message": "vfs: fix data leak in nobh_write_end()\n\nCurrent nobh_write_end() implementation ignore partial writes(copied \u003c len)\ncase if page was fully mapped and simply mark page as Uptodate, which is\ntotally wrong because area [pos+copied, pos+len) wasn\u0027t updated explicitly in\nprevious write_begin call. It simply contains garbage from pagecache and\nresult in data leakage.\n\n#TEST_CASE_BEGIN:\n~~~~~~~~~~~~~~~~\nIn fact issue triggered by classical testcase\n\topen(\"/mnt/test\", O_RDWR|O_CREAT|O_TRUNC, 0666) \u003d 3\n\tftruncate(3, 409600) \u003d 0\n\twritev(3, [{\"a\", 1}, {NULL, 4095}], 2) \u003d 1\n##TESTCASE_SOURCE:\n~~~~~~~~~~~~~~~~~\n#include \u003cstdio.h\u003e\n#include \u003cstdlib.h\u003e\n#include \u003cfcntl.h\u003e\n#include \u003csys/uio.h\u003e\n#include \u003csys/mman.h\u003e\n#include \u003cerrno.h\u003e\nint main(int argc, char **argv)\n{\n\tint fd, ret;\n\tvoid* p;\n\tstruct iovec iov[2];\n\tfd \u003d open(argv[1], O_RDWR|O_CREAT|O_TRUNC, 0666);\n\tftruncate(fd, 409600);\n\tiov[0].iov_base\u003d\"a\";\n\tiov[0].iov_len\u003d1;\n\tiov[1].iov_base\u003dNULL;\n\tiov[1].iov_len\u003d4096;\n\tret \u003d writev(fd, iov, sizeof(iov)/sizeof(struct iovec));\n\tprintf(\"writev \u003d %d, err \u003d %d\\n\", ret, errno);\n\treturn 0;\n}\n##TESTCASE RESULT:\n~~~~~~~~~~~~~~~~~~\n[root@ts63 ~]# mount | grep mnt2\n/dev/mapper/test on /mnt2 type ext2 (rw,nobh)\n[root@ts63 ~]# /tmp/writev /mnt2/test\nwritev \u003d 1, err \u003d 0\n[root@ts63 ~]# hexdump -C /mnt2/test\n\n00000000 61 65 62 6f 6f 74 00 00 f0 b9 b4 59 3a 00 00 00 |aeboot.....Y:...|\n00000010 20 00 00 00 00 00 00 00 21 00 00 00 00 00 00 00 | .......!.......|\n00000020 df df df df df df df df df df df df df df df df |................|\n00000030 3a 00 00 00 2a 00 00 00 21 00 00 00 00 00 00 00 |:...*...!.......|\n00000040 60 c0 8c 00 00 00 00 00 40 4a 8d 00 00 00 00 00 |`.......@J......|\n00000050 00 00 00 00 00 00 00 00 41 00 00 00 00 00 00 00 |........A.......|\n00000060 74 69 6d 65 20 64 64 20 69 66 3d 2f 64 65 76 2f |time dd if\u003d/dev/|\n00000070 6c 6f 6f 70 30 20 20 6f 66 3d 2f 64 65 76 2f 6e |loop0 of\u003d/dev/n|\nskip..\n00000f50 00 00 00 00 00 00 00 00 31 00 00 00 00 00 00 00 |........1.......|\n00000f60 6d 6b 66 73 2e 65 78 74 33 20 2f 64 65 76 2f 76 |mkfs.ext3 /dev/v|\n00000f70 7a 76 67 2f 74 65 73 74 20 2d 62 34 30 39 36 00 |zvg/test -b4096.|\n00000f80 a0 fe 8c 00 00 00 00 00 21 00 00 00 00 00 00 00 |........!.......|\n00000f90 23 31 32 30 35 39 35 30 34 30 34 00 3a 00 00 00 |#1205950404.:...|\n00000fa0 20 00 8d 00 00 00 00 00 21 00 00 00 00 00 00 00 | .......!.......|\n00000fb0 d0 cf 8c 00 00 00 00 00 10 d0 8c 00 00 00 00 00 |................|\n00000fc0 00 00 00 00 00 00 00 00 41 00 00 00 00 00 00 00 |........A.......|\n00000fd0 6d 6f 75 6e 74 20 2f 64 65 76 2f 76 7a 76 67 2f |mount /dev/vzvg/|\n00000fe0 74 65 73 74 20 20 2f 76 7a 20 2d 6f 20 64 61 74 |test /vz -o dat|\n00000ff0 61 3d 77 72 69 74 65 62 61 63 6b 00 00 00 00 00 |a\u003dwriteback.....|\n00001000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|\n\nAs you can see file\u0027s page contains garbage from pagecache instead of zeros.\n#TEST_CASE_END\n\nAttached patch:\n- Add sanity check BUG_ON in order to prevent incorrect usage by caller,\n This is function invariant because page can has buffers and in no zero\n *fadata pointer at the same time.\n- Always attach buffers to page is it is partial write case.\n- Always switch back to generic_write_end if page has buffers.\n This is reasonable because if page already has buffer then generic_write_begin\n was called previously.\n\nSigned-off-by: Dmitri Monakhov \u003cdmonakhov@openvz.org\u003e\nReviewed-by: Nick Piggin \u003cnpiggin@suse.de\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": "a6b91919e0881a0d0a4ae5211d5c879a8c7ca92b", "tree": "148aabc0a92a297f9220174435e822f30c951215", "parents": [ "5366dc9fd1cb74023790a8edf29e4fe42c886dcf" ], "author": { "name": "Randy Dunlap", "email": "randy.dunlap@oracle.com", "time": "Wed Mar 19 17:01:00 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 19 18:53:36 2008 -0700" }, "message": "fs: fix kernel-doc notation warnings\n\nFix kernel-doc notation warnings in fs/.\n\nWarning(mmotm-2008-0314-1449//fs/super.c:560): missing initial short description on line:\n *\tmark_files_ro\nWarning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line:\n *\tlease_get_mtime\nWarning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line:\n *\tlease_get_mtime\nWarning(mmotm-2008-0314-1449//fs/namei.c:1368): missing initial short description on line:\n * lookup_one_len: filesystem helper to lookup single pathname component\nWarning(mmotm-2008-0314-1449//fs/buffer.c:3221): missing initial short description on line:\n * bh_uptodate_or_lock: Test whether the buffer is uptodate\nWarning(mmotm-2008-0314-1449//fs/buffer.c:3240): missing initial short description on line:\n * bh_submit_read: Submit a locked buffer for reading\nWarning(mmotm-2008-0314-1449//fs/fs-writeback.c:30): missing initial short description on line:\n * writeback_acquire: attempt to get exclusive writeback access to a device\nWarning(mmotm-2008-0314-1449//fs/fs-writeback.c:47): missing initial short description on line:\n * writeback_in_progress: determine whether there is writeback in progress\nWarning(mmotm-2008-0314-1449//fs/fs-writeback.c:58): missing initial short description on line:\n * writeback_release: relinquish exclusive writeback access against a device.\nWarning(mmotm-2008-0314-1449//include/linux/jbd.h:351): contents before sections\nWarning(mmotm-2008-0314-1449//include/linux/jbd.h:561): contents before sections\nWarning(mmotm-2008-0314-1449//fs/jbd/transaction.c:1935): missing initial short description on line:\n * void journal_invalidatepage()\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": "e3892296de632e3f9299d9fabe0c746740004891", "tree": "a6162f566e5ec6b0e099a79e7a59005fcfe9be4b", "parents": [ "10ed273f5016c582413dfbc468dd084957d847e1" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Tue Mar 04 14:28:33 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Mar 04 16:35:10 2008 -0800" }, "message": "vfs: fix NULL pointer dereference in fsync_buffers_list()\n\nFix NULL pointer dereference in fsync_buffers_list() introduced by recent fix\nof races in private_list handling. Since bh-\u003eb_assoc_map has been cleared in\n__remove_assoc_queue() we should really use original value stored in the\n\u0027mapping\u0027 variable.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "78a4a50a86b0a54f7ecbc164267b6c762760254c", "tree": "a50503b14fc4f21abdcc0e2c6e49380029bb5e1c", "parents": [ "64e6269071fb1888e5ddc9ad557368cc5fa4f538" ], "author": { "name": "Randy Dunlap", "email": "randy.dunlap@oracle.com", "time": "Fri Feb 29 22:02:31 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Mar 03 10:47:13 2008 -0800" }, "message": "docbook: fix filesystems.tmpl source files\n\nFix docbook problems in filesystems.tmpl.\nThese cause the generated docbook to be incorrect.\n\nSigned-off-by: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "535ee2fbf79ab52d26bce3d2e127c9007503581e", "tree": "9d7cdbd98e481ec2fb6d5ff6eebf65f4d422ec8c", "parents": [ "f6f21c81464ce52dbeec921bdc2e8b288c491920" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Fri Feb 08 04:21:59 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:42 2008 -0800" }, "message": "buffer_head: fix private_list handling\n\nThere are two possible races in handling of private_list in buffer cache.\n\n1) When fsync_buffers_list() processes a private_list, it clears\n b_assoc_mapping and moves buffer to its private list. Now\n drop_buffers() comes, sees a buffer is on list so it calls\n __remove_assoc_queue() which complains about b_assoc_mapping being\n cleared (as it cannot propagate possible IO error). This race has been\n actually observed in the wild.\n\n2) When fsync_buffers_list() processes a private_list,\n mark_buffer_dirty_inode() can be called on bh which is already on the\n private list of fsync_buffers_list(). As buffer is on some list (note\n that the check is performed without private_lock), it is not readded to\n the mapping\u0027s private_list and after fsync_buffers_list() finishes, we\n have a dirty buffer which should be on private_list but it isn\u0027t. This\n race has not been reported, probably because most (but not all) callers\n of mark_buffer_dirty_inode() hold i_mutex and thus are serialized with\n fsync().\n\nFix these issues by not clearing b_assoc_map when fsync_buffers_list()\nmoves buffer to a dedicated list and by reinserting buffer in private_list\nwhen it is found dirty after we have submitted buffer for IO. We also\nchange the tests whether a buffer is on a private list from\n!list_empty(\u0026bh-\u003eb_assoc_buffers) to bh-\u003eb_assoc_map so that they are\nsingle word reads and hence lockless checks are safe.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fc9b52cd8f5f459b88adcf67c47668425ae31a78", "tree": "c29924eaf60d2e73641bf11fa906a23fa81f46c9", "parents": [ "75acb9cd2ef0bbb463098fdd40cbcdda79d45fa3" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Fri Feb 08 04:19:52 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:31 2008 -0800" }, "message": "fs: remove fastcall, it is always empty\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9db5579be4bb5320c3248f6acf807aedf05ae143", "tree": "fde09bbeb427946b30d2e0fb6b00494a42488052", "parents": [ "daeb51e62cacde31c8245866e1096ff79a0c83fe" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Fri Feb 08 04:19:49 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Feb 08 09:22:30 2008 -0800" }, "message": "rewrite rd\n\nThis is a rewrite of the ramdisk block device driver.\n\nThe old one is really difficult because it effectively implements a block\ndevice which serves data out of its own buffer cache. It relies on the dirty\nbit being set, to pin its backing store in cache, however there are non\ntrivial paths which can clear the dirty bit (eg. try_to_free_buffers()),\nwhich had recently lead to data corruption. And in general it is completely\nwrong for a block device driver to do this.\n\nThe new one is more like a regular block device driver. It has no idea about\nvm/vfs stuff. It\u0027s backing store is similar to the buffer cache (a simple\nradix-tree of pages), but it doesn\u0027t know anything about page cache (the pages\nin the radix tree are not pagecache pages).\n\nThere is one slight downside -- direct block device access and filesystem\nmetadata access goes through an extra copy and gets stored in RAM twice.\nHowever, this downside is only slight, because the real buffercache of the\ndevice is now reclaimable (because we\u0027re not playing crazy games with it), so\nunder memory intensive situations, footprint should effectively be the same --\nmaybe even a slight advantage to the new driver because it can also reclaim\nbuffer heads.\n\nThe fact that it now goes through all the regular vm/fs paths makes it\nmuch more useful for testing, too.\n\n text data bss dec hex filename\n 2837 849 384 4070 fe6 drivers/block/rd.o\n 3528 371 12 3911 f47 drivers/block/brd.o\n\nText is larger, but data and bss are smaller, making total size smaller.\n\nA few other nice things about it:\n- Similar structure and layout to the new loop device handlinag.\n- Dynamic ramdisk creation.\n- Runtime flexible buffer head size (because it is no longer part of the\n ramdisk code).\n- Boot / load time flexible ramdisk size, which could easily be extended\n to a per-ramdisk runtime changeable size (eg. with an ioctl).\n- Can use highmem for the backing store.\n\n[akpm@linux-foundation.org: fix build]\n[byron.bbradley@gmail.com: make rd_size non-static]\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Byron Bradley \u003cbyron.bbradley@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b98938c373117043598002f197200d7ed08acd49", "tree": "da9356af43085db78cafec82505c3341d6291201", "parents": [ "aec2c3ed01ed54d0cdf7f6b7c4be217c045ac5ea" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Mon Feb 04 22:28:36 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Feb 05 09:44:14 2008 -0800" }, "message": "bufferhead: revert constructor removal\n\nThe constructor for buffer_head slabs was removed recently. We need the\nconstructor back in slab defrag in order to insure that slab objects always\nhave a definite state even before we allocated them.\n\nI think we mistakenly merged the removal of the constuctor into a cleanup\npatch. You (ie: akpm) had a test that showed that the removal of the\nconstructor led to a small regression. The prior state makes things easier\nfor slab defrag.\n\n[akpm@linux-foundation.org: coding-style fixes]\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "eebd2aa355692afaf9906f62118620f1a1c19dbb", "tree": "207eead3a736963c3e50942038c463f2f611ccce", "parents": [ "b98348bdd08dc4ec11828aa98a78edde15c53cfa" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Mon Feb 04 22:28:29 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Feb 05 09:44:13 2008 -0800" }, "message": "Pagecache zeroing: zero_user_segment, zero_user_segments and zero_user\n\nSimplify page cache zeroing of segments of pages through 3 functions\n\nzero_user_segments(page, start1, end1, start2, end2)\n\n Zeros two segments of the page. It takes the position where to\n start and end the zeroing which avoids length calculations and\n\tmakes code clearer.\n\nzero_user_segment(page, start, end)\n\n Same for a single segment.\n\nzero_user(page, start, length)\n\n Length variant for the case where we know the length.\n\nWe remove the zero_user_page macro. Issues:\n\n1. Its a macro. Inline functions are preferable.\n\n2. The KM_USER0 macro is only defined for HIGHMEM.\n\n Having to treat this special case everywhere makes the\n code needlessly complex. The parameter for zeroing is always\n KM_USER0 except in one single case that we open code.\n\nAvoiding KM_USER0 makes a lot of code not having to be dealing\nwith the special casing for HIGHMEM anymore. Dealing with\nkmap is only necessary for HIGHMEM configurations. In those\nconfigurations we use KM_USER0 like we do for a series of other\nfunctions defined in highmem.h.\n\nSince KM_USER0 is depends on HIGHMEM the existing zero_user_page\nfunction could not be a macro. zero_user_* functions introduced\nhere can be be inline because that constant is not used when these\nfunctions are called.\n\nAlso extract the flushing of the caches to be outside of the kmap.\n\n[akpm@linux-foundation.org: fix nfs and ntfs build]\n[akpm@linux-foundation.org: fix ntfs build some more]\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Steven French \u003csfrench@us.ibm.com\u003e\nCc: Michael Halcrow \u003cmhalcrow@us.ibm.com\u003e\nCc: \u003clinux-ext4@vger.kernel.org\u003e\nCc: Steven Whitehouse \u003cswhiteho@redhat.com\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nCc: \"J. Bruce Fields\" \u003cbfields@fieldses.org\u003e\nCc: Anton Altaparmakov \u003caia21@cantab.net\u003e\nCc: Mark Fasheh \u003cmark.fasheh@oracle.com\u003e\nCc: David Chinner \u003cdgc@sgi.com\u003e\nCc: Michael Halcrow \u003cmhalcrow@us.ibm.com\u003e\nCc: Steven French \u003csfrench@us.ibm.com\u003e\nCc: Steven Whitehouse \u003cswhiteho@redhat.com\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "389d1b083c767a360ec84b27a95da06244becec8", "tree": "a315c892f615545fd0ce72f7e0d660f5772fb4b2", "parents": [ "bb4f397a1a7f2330cb173233599aa159f5780f58" ], "author": { "name": "Aneesh Kumar K.V", "email": "aneesh.kumar@linux.vnet.ibm.com", "time": "Mon Jan 28 23:58:26 2008 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Mon Jan 28 23:58:26 2008 -0500" }, "message": "Add buffer head related helper functions\n\nAdd buffer head related helper function bh_uptodate_or_lock and\nbh_submit_read which can be used by file system\n\nSigned-off-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.vnet.ibm.com\u003e\n" }, { "commit": "efdc31319d43050a5742fb690b1a4beb68092a94", "tree": "697e9db6db5675c46d559d8982e34a1f8df4f3d1", "parents": [ "cfa76f024f7c9e65169425804e5b32e71f66d0ee" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Sun Oct 21 06:57:41 2007 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sun Oct 21 08:54:05 2007 -0700" }, "message": "nobh: nobh_write_end fix\n\nThis path mustn\u0027t have been tested :( I did attempt to exercise it\nby injecting failures here, but I suspect PageMappedToDisk may have\nbeen getting in the way. Will need more of a look, although I think\nnobh mode is OK for an -rc1 (it shouldn\u0027t eat anyone\u0027s data).\n\nCommit 03158cd7eb3374843de68421142ca5900df845d9 (\"fs: restore nobh\")\nintrocduced a NULL deref. Spotted by the Coverity checker.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Badari Pulavarty \u003cpbadari@us.ibm.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1f7decf6d9f06dac008b8d66935c0c3b18e564f9", "tree": "e0f8430bd4e3ed5425aa55b7fa28483341bbfe87", "parents": [ "08d8e9749e7f0435ba4683b620e8d30d59276b4c" ], "author": { "name": "Fengguang Wu", "email": "wfg@mail.ustc.edu.cn", "time": "Tue Oct 16 23:30:42 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 08:43:02 2007 -0700" }, "message": "writeback: remove pages_skipped accounting in __block_write_full_page()\n\nMiklos Szeredi \u003cmiklos@szeredi.hu\u003e and me identified a writeback bug:\n\n\u003e The following strange behavior can be observed:\n\u003e\n\u003e 1. large file is written\n\u003e 2. after 30 seconds, nr_dirty goes down by 1024\n\u003e 3. then for some time (\u003c 30 sec) nothing happens (disk idle)\n\u003e 4. then nr_dirty again goes down by 1024\n\u003e 5. repeat from 3. until whole file is written\n\u003e\n\u003e So basically a 4Mbyte chunk of the file is written every 30 seconds.\n\u003e I\u0027m quite sure this is not the intended behavior.\n\nIt can be produced by the following test scheme:\n\n# cat bin/test-writeback.sh\ngrep nr_dirty /proc/vmstat\necho 1 \u003e /proc/sys/fs/inode_debug\ndd if\u003d/dev/zero of\u003d/var/x bs\u003d1K count\u003d204800\u0026\nwhile true; do grep nr_dirty /proc/vmstat; sleep 1; done\n\n# bin/test-writeback.sh\nnr_dirty 19207\nnr_dirty 19207\nnr_dirty 30924\n204800+0 records in\n204800+0 records out\n209715200 bytes (210 MB) copied, 1.58363 seconds, 132 MB/s\nnr_dirty 47150\nnr_dirty 47141\nnr_dirty 47142\nnr_dirty 47142\nnr_dirty 47142\nnr_dirty 47142\nnr_dirty 47205\nnr_dirty 47214\nnr_dirty 47214\nnr_dirty 47214\nnr_dirty 47214\nnr_dirty 47214\nnr_dirty 47215\nnr_dirty 47216\nnr_dirty 47216\nnr_dirty 47216\nnr_dirty 47154\nnr_dirty 47143\nnr_dirty 47143\nnr_dirty 47143\nnr_dirty 47143\nnr_dirty 47143\nnr_dirty 47142\nnr_dirty 47142\nnr_dirty 47142\nnr_dirty 47142\nnr_dirty 47134\nnr_dirty 47134\nnr_dirty 47135\nnr_dirty 47135\nnr_dirty 47135\nnr_dirty 46097 \u003c\u003d\u003d -1038\nnr_dirty 46098\nnr_dirty 46098\nnr_dirty 46098\n[...]\nnr_dirty 46091\nnr_dirty 46092\nnr_dirty 46092\nnr_dirty 45069 \u003c\u003d\u003d -1023\nnr_dirty 45056\nnr_dirty 45056\nnr_dirty 45056\n[...]\nnr_dirty 37822\nnr_dirty 36799 \u003c\u003d\u003d -1023\n[...]\nnr_dirty 36781\nnr_dirty 35758 \u003c\u003d\u003d -1023\n[...]\nnr_dirty 34708\nnr_dirty 33672 \u003c\u003d\u003d -1024\n[...]\nnr_dirty 33692\nnr_dirty 32669 \u003c\u003d\u003d -1023\n\n% ls -li /var/x\n847824 -rw-r--r-- 1 root root 200M 2007-08-12 04:12 /var/x\n\n% dmesg|grep 847824 # generated by a debug printk\n[ 529.263184] redirtied inode 847824 line 548\n[ 564.250872] redirtied inode 847824 line 548\n[ 594.272797] redirtied inode 847824 line 548\n[ 629.231330] redirtied inode 847824 line 548\n[ 659.224674] redirtied inode 847824 line 548\n[ 689.219890] redirtied inode 847824 line 548\n[ 724.226655] redirtied inode 847824 line 548\n[ 759.198568] redirtied inode 847824 line 548\n\n# line 548 in fs/fs-writeback.c:\n543 if (wbc-\u003epages_skipped !\u003d pages_skipped) {\n544 /*\n545 * writeback is not making progress due to locked\n546 * buffers. Skip this inode for now.\n547 */\n548 redirty_tail(inode);\n549 }\n\nMore debug efforts show that __block_write_full_page()\nnever has the chance to call submit_bh() for that big dirty file:\nthe buffer head is *clean*. So basicly no page io is issued by\n__block_write_full_page(), hence pages_skipped goes up.\n\nAlso the comment in generic_sync_sb_inodes():\n\n544 /*\n545 * writeback is not making progress due to locked\n546 * buffers. Skip this inode for now.\n547 */\n\nand the comment in __block_write_full_page():\n\n1713 /*\n1714 * The page was marked dirty, but the buffers were\n1715 * clean. Someone wrote them back by hand with\n1716 * ll_rw_block/submit_bh. A rare case.\n1717 */\n\ndo not quite agree with each other. The page writeback should be skipped for\n\u0027locked buffer\u0027, but here it is \u0027clean buffer\u0027!\n\nThis patch fixes this bug. Though I\u0027m not sure why __block_write_full_page()\nis called only to do nothing and who actually issued the writeback for us.\n\nThis is the two possible new behaviors after the patch:\n\n1) pretty nice: wait 30s and write ALL:)\n2) not so good:\n\t- during the dd: ~16M\n\t- after 30s: ~4M\n\t- after 5s: ~4M\n\t- after 5s: ~176M\n\nThe next patch will fix case (2).\n\nCc: David Chinner \u003cdgc@sgi.com\u003e\nCc: Ken Chen \u003ckenchen@google.com\u003e\nSigned-off-by: Fengguang Wu \u003cwfg@mail.ustc.edu.cn\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c9e51e4180696aa67915ec5665e4ec74125565de", "tree": "d66e1427b16e56bc7da0dfb583a6d8edb55a2c3e", "parents": [ "b2e8fb6efa209c82203c79b491b5bc952d44aa57" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue Oct 16 23:25:47 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 08:42:45 2007 -0700" }, "message": "mm: count reclaimable pages per BDI\n\nCount per BDI reclaimable pages; nr_reclaimable \u003d nr_dirty + nr_unstable.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e12ba74d8ff3e2f73a583500d7095e406df4d093", "tree": "a0d3385b65f0b3e1e00b0bbf11b75e7538a93edb", "parents": [ "c361be55b3128474aa66d31092db330b07539103" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Tue Oct 16 01:25:52 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:43:00 2007 -0700" }, "message": "Group short-lived and reclaimable kernel allocations\n\nThis patch marks a number of allocations that are either short-lived such as\nnetwork buffers or are reclaimable such as inode allocations. When something\nlike updatedb is called, long-lived and unmovable kernel allocations tend to\nbe spread throughout the address space which increases fragmentation.\n\nThis patch groups these allocations together as much as possible by adding a\nnew MIGRATE_TYPE. The MIGRATE_RECLAIMABLE type is for allocations that can be\nreclaimed on demand, but not moved. i.e. they can be migrated by deleting\nthem and re-reading the information from elsewhere.\n\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Andy Whitcroft \u003capw@shadowen.org\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "03158cd7eb3374843de68421142ca5900df845d9", "tree": "5e1bfc1f981651e7f7d7cf30afd15d7bd96b2a57", "parents": [ "b6af1bcd8720cb3062c8c4d4c8ba02bee10ff03f" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Oct 16 01:25:25 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:58 2007 -0700" }, "message": "fs: restore nobh\n\nImplement nobh in new aops. This is a bit tricky. FWIW, nobh_truncate is\nnow implemented in a way that does not create blocks in sparse regions,\nwhich is a silly thing for it to have been doing (isn\u0027t it?)\n\next2 survives fsx and fsstress. jfs is converted as well... ext3\nshould be easy to do (but not done yet).\n\n[akpm@linux-foundation.org: coding-style fixes]\nCc: Badari Pulavarty \u003cpbadari@us.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": "a20fa20c549ed569885d871f689a59cfd2f6ff77", "tree": "bc6f7119cec83032eb43c27d6ec7bbbebd2fdddb", "parents": [ "f7557e8f7ff785d6c2b5bc914cd1675314ff0fcf" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Oct 16 01:25:15 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:56 2007 -0700" }, "message": "With reiserfs no longer using the weird generic_cont_expand, remove it completely.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "89e107877b65bf6eff1d63a1302dee9a091586f5", "tree": "8a120a04c46f19229d1cf9b9c546f1818cf84c44", "parents": [ "7765ec26ae1c01bb29bedf910e4efcced8cc81d2" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Oct 16 01:25:07 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:55 2007 -0700" }, "message": "fs: new cont helpers\n\nRework the generic block \"cont\" routines to handle the new aops. Supporting\ncont_prepare_write would take quite a lot of code to support, so remove it\ninstead (and we later convert all filesystems to use it).\n\nwrite_begin gets passed AOP_FLAG_CONT_EXPAND when called from\ngeneric_cont_expand, so filesystems can avoid the old hacks they used.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: OGAWA Hirofumi \u003chirofumi@mail.parknet.co.jp\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "afddba49d18f346e5cc2938b6ed7c512db18ca68", "tree": "4726e3d3b0e9e8e5b5d3b2b0cccb36446bbdf3ca", "parents": [ "637aff46f94a754207c80c8c64bf1b74f24b967d" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Oct 16 01:25:01 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:55 2007 -0700" }, "message": "fs: introduce write_begin, write_end, and perform_write aops\n\nThese are intended to replace prepare_write and commit_write with more\nflexible alternatives that are also able to avoid the buffered write\ndeadlock problems efficiently (which prepare_write is unable to do).\n\n[mark.fasheh@oracle.com: API design contributions, code review and fixes]\n[akpm@linux-foundation.org: various fixes]\n[dmonakhov@sw.ru: new aop block_write_begin fix]\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Mark Fasheh \u003cmark.fasheh@oracle.com\u003e\nSigned-off-by: Dmitriy Monakhov \u003cdmonakhov@openvz.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "637aff46f94a754207c80c8c64bf1b74f24b967d", "tree": "837fb3196236998c3c3ae44762ed3167ef0256f4", "parents": [ "2f718ffc16c43a435d12919c75dbfad518abd056" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Oct 16 01:25:00 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:55 2007 -0700" }, "message": "fs: fix data-loss on error\n\nNew buffers against uptodate pages are simply be marked uptodate, while the\nbuffer_new bit remains set. This causes error-case code to zero out parts of\nthose buffers because it thinks they contain stale data: wrong, they are\nactually uptodate so this is a data loss situation.\n\nFix this by actually clearning buffer_new and marking the buffer dirty. It\nmakes sense to always clear buffer_new before setting a buffer uptodate.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a4b0672db3a698d0684ee6e54f44e2e162a3da1b", "tree": "d53c52b251856520b0fd02969bfcdd4b5d332da6", "parents": [ "68671f35fe8d785277118a333c88768a4f894917" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Oct 16 01:24:48 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:54 2007 -0700" }, "message": "fs: fix nobh error handling\n\nnobh mode error handling is not just pretty slack, it\u0027s wrong.\n\nOne cannot zero out the whole page to ensure new blocks are zeroed, because\nit just brings the whole page \"uptodate\" with zeroes even if that may not\nbe the correct uptodate data. Also, other parts of the page may already\ncontain dirty data which would get lost by zeroing it out. Thirdly, the\nwriteback of zeroes to the new blocks will also erase existing blocks. All\nthese conditions are pagecache and/or filesystem corruption.\n\nThe problem comes about because we didn\u0027t keep track of which buffers\nactually are new or old. However it is not enough just to keep only this\nstate, because at the point we start dirtying parts of the page (new\nblocks, with zeroes), the handling of IO errors becomes impossible without\nbuffers because the page may only be partially uptodate, in which case the\npage flags allone cannot capture the state of the parts of the page.\n\nSo allocate all buffers for the page upfront, but leave them unattached so\nthat they don\u0027t pick up any other references and can be freed when we\u0027re\ndone. If the error path is hit, then zero the new buffers as the regular\nbuffer path does, then attach the buffers to the page so that it can\nactually be written out correctly and be subject to the normal IO error\nhandling paths.\n\nAs an upshot, we save 1K of kernel stack on ia64 or powerpc 64K page\nsystems.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "68671f35fe8d785277118a333c88768a4f894917", "tree": "50cb5a378db1b021c2ec3573388af3480df49791", "parents": [ "ef8b4520bd9f8294ffce9abd6158085bde5dc902" ], "author": { "name": "Dmitry Monakhov", "email": "dmonakhov@openvz.org", "time": "Tue Oct 16 01:24:47 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:53 2007 -0700" }, "message": "mm: add end_buffer_read helper function\n\nMove duplicated code from end_buffer_read_XXX methods to separate helper\nfunction.\n\nSigned-off-by: Dmitry Monakhov \u003cdmonakhov@openvz.org\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6712ecf8f648118c3363c142196418f89a510b90", "tree": "347d39a7d5a7ed96d3b1afecd28de2a0f98b98c9", "parents": [ "5bb23a688b2de23d7765a1dd439d89c038378978" ], "author": { "name": "NeilBrown", "email": "neilb@suse.de", "time": "Thu Sep 27 12:47:43 2007 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@carl.home.kernel.dk", "time": "Wed Oct 10 09:25:57 2007 +0200" }, "message": "Drop \u0027size\u0027 argument from bio_endio and bi_end_io\n\nAs bi_end_io is only called once when the reqeust is complete,\nthe \u0027size\u0027 argument is now redundant. Remove it.\n\nNow there is no need for bio_endio to subtract the size completed\nfrom bi_size. So don\u0027t do that either.\n\nWhile we are at it, change bi_end_io to return void.\n\nSigned-off-by: Neil Brown \u003cneilb@suse.de\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "1833633803c7ef4d8f09877d3f1549cbd252f477", "tree": "ce69505de705701f838bd77c6b6b6bc18529bed0", "parents": [ "e228929bc257b963523ed75aa60d2ad77ece2189" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Fri Jul 20 00:31:45 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Jul 20 08:44:19 2007 -0700" }, "message": "fix some conversion overflows\n\nFix page index to offset conversion overflows in buffer layer, ecryptfs,\nand ocfs2.\n\nIt would be nice to convert the whole tree to page_offset, but for now\njust fix the bugs.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Michael Halcrow \u003cmhalcrow@us.ibm.com\u003e\nCc: Mark Fasheh \u003cmark.fasheh@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": "5417169026c3df151adf5a65eb061278b0a72e69", "tree": "e4872a4243e12fe931ff6ac789df181b579b884c", "parents": [ "589f1e81bde732dd0b1bc5d01b6bddd4bcb4527b" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Jul 19 17:39:55 2007 +1000" }, "committer": { "name": "Tim Shimmin", "email": "tes@chook.melbourne.sgi.com", "time": "Thu Jul 19 19:50:50 2007 +1000" }, "message": "[FS] Implement block_page_mkwrite.\n\nMany filesystems need a -\u003epage-mkwrite callout to correctly\nset up pages that have been written to by mmap. This is especially\nimportant when mmap is writing into holes as it allows filesystems\nto correctly account for and allocate space before the mmap\nwrite is allowed to proceed.\n\nProtection against truncate races is provided by locking the page\nand checking to see whether the page mapping is correct and whether\nit is beyond EOF so we don\u0027t end up allowing allocations beyond\nthe current EOF or changing EOF as a result of a mmap write.\n\nSGI-PV: 940392\nSGI-Modid: 2.6.x-xfs-melb:linux:29146a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Tim Shimmin \u003ctes@sgi.com\u003e\n" }, { "commit": "787d2214c19bcc9b6ac48af0ce098277a801eded", "tree": "a040604fdf9620a66dc83a0cde4f2140e2ec25b3", "parents": [ "a1ed3dda0ad181532f1e0f0d548067fb9fdddac4" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Tue Jul 17 04:03:34 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Jul 17 10:23:02 2007 -0700" }, "message": "fs: introduce some page/buffer invariants\n\nIt is a bug to set a page dirty if it is not uptodate unless it has\nbuffers. If the page has buffers, then the page may be dirty (some buffers\ndirty) but not uptodate (some buffers not uptodate). The exception to this\nrule is if the set_page_dirty caller is racing with truncate or invalidate.\n\nA buffer can not be set dirty if it is not uptodate.\n\nIf either of these situations occurs, it indicates there could be some data\nloss problem. Some of these warnings could be a harmless one where the\npage or buffer is set uptodate immediately after it is dirtied, however we\nshould fix those up, and enforce this ordering.\n\nBring the order of operations for truncate into line with those of\ninvalidate. This will prevent a page from being able to go !uptodate while\nwe\u0027re holding the tree_lock, which is probably a good thing anyway.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5ad333eb66ff1e52a87639822ae088577669dcf9", "tree": "addae6bbd19585f19328f309924d06d647e8f2b7", "parents": [ "7e63efef857575320fb413fbc3d0ee704b72845f" ], "author": { "name": "Andy Whitcroft", "email": "apw@shadowen.org", "time": "Tue Jul 17 04:03:16 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Jul 17 10:22:59 2007 -0700" }, "message": "Lumpy Reclaim V4\n\nWhen we are out of memory of a suitable size we enter reclaim. The current\nreclaim algorithm targets pages in LRU order, which is great for fairness at\norder-0 but highly unsuitable if you desire pages at higher orders. To get\npages of higher order we must shoot down a very high proportion of memory;\n\u003e95% in a lot of cases.\n\nThis patch set adds a lumpy reclaim algorithm to the allocator. It targets\ngroups of pages at the specified order anchored at the end of the active and\ninactive lists. This encourages groups of pages at the requested orders to\nmove from active to inactive, and active to free lists. This behaviour is\nonly triggered out of direct reclaim when higher order pages have been\nrequested.\n\nThis patch set is particularly effective when utilised with an\nanti-fragmentation scheme which groups pages of similar reclaimability\ntogether.\n\nThis patch set is based on Peter Zijlstra\u0027s lumpy reclaim V2 patch which forms\nthe foundation. Credit to Mel Gorman for sanitity checking.\n\nMel said:\n\n The patches have an application with hugepage pool resizing.\n\n When lumpy-reclaim is used used with ZONE_MOVABLE, the hugepages pool can\n be resized with greater reliability. Testing on a desktop machine with 2GB\n of RAM showed that growing the hugepage pool with ZONE_MOVABLE on it\u0027s own\n was very slow as the success rate was quite low. Without lumpy-reclaim,\n each attempt to grow the pool by 100 pages would yield 1 or 2 hugepages.\n With lumpy-reclaim, getting 40 to 70 hugepages on each attempt was typical.\n\n[akpm@osdl.org: ia64 pfn_to_nid fixes and loop cleanup]\n[bunk@stusta.de: static declarations for internal functions]\n[a.p.zijlstra@chello.nl: initial lumpy V2 implementation]\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "769848c03895b63e5662eb7e4ec8c4866f7d0183", "tree": "8911c7c312c8b8b172795fa2874c8162e1d3d15a", "parents": [ "a32ea1e1f925399e0d81ca3f7394a44a6dafa12c" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Tue Jul 17 04:03:05 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Jul 17 10:22:59 2007 -0700" }, "message": "Add __GFP_MOVABLE for callers to flag allocations from high memory that may be migrated\n\nIt is often known at allocation time whether a page may be migrated or not.\nThis patch adds a flag called __GFP_MOVABLE and a new mask called\nGFP_HIGH_MOVABLE. Allocations using the __GFP_MOVABLE can be either migrated\nusing the page migration mechanism or reclaimed by syncing with backing\nstorage and discarding.\n\nAn API function very similar to alloc_zeroed_user_highpage() is added for\n__GFP_MOVABLE allocations called alloc_zeroed_user_highpage_movable(). The\nflags used by alloc_zeroed_user_highpage() are not changed because it would\nchange the semantics of an existing API. After this patch is applied there\nare no in-kernel users of alloc_zeroed_user_highpage() so it probably should\nbe marked deprecated if this patch is merged.\n\nNote that this patch includes a minor cleanup to the use of __GFP_ZERO in\nshmem.c to keep all flag modifications to inode-\u003emapping in the\nshmem_dir_alloc() helper function. This clean-up suggestion is courtesy of\nHugh Dickens.\n\nAdditional credit goes to Christoph Lameter and Linus Torvalds for shaping the\nconcept. Credit to Hugh Dickens for catching issues with shmem swap vector\nand ramfs allocations.\n\n[akpm@linux-foundation.org: build fix]\n[hugh@veritas.com: __GFP_ZERO cleanup]\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Andy Whitcroft \u003capw@shadowen.org\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ], "next": "4a2d44590a603be292addce9c263982043416666" }