)]}' { "log": [ { "commit": "0e175a1835ffc979e55787774e58ec79e41957d7", "tree": "6ec4b65a8de4e9d1c12d26a1079079ed81d79450", "parents": [ "ad4e38dd6a33bb3a4882c487d7abe621e583b982" ], "author": { "name": "Curt Wohlgemuth", "email": "curtw@google.com", "time": "Fri Oct 07 21:54:10 2011 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Mon Oct 31 00:33:36 2011 +0800" }, "message": "writeback: Add a \u0027reason\u0027 to wb_writeback_work\n\nThis creates a new \u0027reason\u0027 field in a wb_writeback_work\nstructure, which unambiguously identifies who initiates\nwriteback activity. A \u0027wb_reason\u0027 enumeration has been\nadded to writeback.h, to enumerate the possible reasons.\n\nThe \u0027writeback_work_class\u0027 and tracepoint event class and\n\u0027writeback_queue_io\u0027 tracepoints are updated to include the\nsymbolic \u0027reason\u0027 in all trace events.\n\nAnd the \u0027writeback_inodes_sbXXX\u0027 family of routines has had\na wb_stats parameter added to them, so callers can specify\nwhy writeback is being started.\n\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Curt Wohlgemuth \u003ccurtw@google.com\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "7381131cbcf7e15d201a0ffd782a4698efe4e740", "tree": "83f00c40d0a3fcd41ff2e6681a5da70dd155628a", "parents": [ "be3ffa276446e1b691a2bf84e7621e5a6fb49db9" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Fri Aug 26 15:53:24 2011 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Mon Oct 03 21:08:57 2011 +0800" }, "message": "writeback: stabilize bdi-\u003edirty_ratelimit\n\nThere are some imperfections in balanced_dirty_ratelimit.\n\n1) large fluctuations\n\nThe dirty_rate used for computing balanced_dirty_ratelimit is merely\naveraged in the past 200ms (very small comparing to the 3s estimation\nperiod for write_bw), which makes rather dispersed distribution of\nbalanced_dirty_ratelimit.\n\nIt\u0027s pretty hard to average out the singular points by increasing the\nestimation period. Considering that the averaging technique will\nintroduce very undesirable time lags, I give it up totally. (btw, the 3s\nwrite_bw averaging time lag is much more acceptable because its impact\nis one-way and therefore won\u0027t lead to oscillations.)\n\nThe more practical way is filtering -- most singular\nbalanced_dirty_ratelimit points can be filtered out by remembering some\nprev_balanced_rate and prev_prev_balanced_rate. However the more\nreliable way is to guard balanced_dirty_ratelimit with task_ratelimit.\n\n2) due to truncates and fs redirties, the (write_bw \u003c\u003d\u003e dirty_rate)\nmatch could become unbalanced, which may lead to large systematical\nerrors in balanced_dirty_ratelimit. The truncates, due to its possibly\nbumpy nature, can hardly be compensated smoothly. So let\u0027s face it. When\nsome over-estimated balanced_dirty_ratelimit brings dirty_ratelimit\nhigh, dirty pages will go higher than the setpoint. task_ratelimit will\nin turn become lower than dirty_ratelimit. So if we consider both\nbalanced_dirty_ratelimit and task_ratelimit and update dirty_ratelimit\nonly when they are on the same side of dirty_ratelimit, the systematical\nerrors in balanced_dirty_ratelimit won\u0027t be able to bring\ndirty_ratelimit far away.\n\nThe balanced_dirty_ratelimit estimation may also be inaccurate near\n@limit or @freerun, however is less an issue.\n\n3) since we ultimately want to\n\n- keep the fluctuations of task ratelimit as small as possible\n- keep the dirty pages around the setpoint as long time as possible\n\nthe update policy used for (2) also serves the above goals nicely:\nif for some reason the dirty pages are high (task_ratelimit \u003c dirty_ratelimit),\nand dirty_ratelimit is low (dirty_ratelimit \u003c balanced_dirty_ratelimit),\nthere is no point to bring up dirty_ratelimit in a hurry only to hurt\nboth the above two goals.\n\nSo, we make use of task_ratelimit to limit the update of dirty_ratelimit\nin two ways:\n\n1) avoid changing dirty rate when it\u0027s against the position control target\n (the adjusted rate will slow down the progress of dirty pages going\n back to setpoint).\n\n2) limit the step size. task_ratelimit is changing values step by step,\n leaving a consistent trace comparing to the randomly jumping\n balanced_dirty_ratelimit. task_ratelimit also has the nice smaller\n errors in stable state and typically larger errors when there are big\n errors in rate. So it\u0027s a pretty good limiting factor for the step\n size of dirty_ratelimit.\n\nNote that bdi-\u003edirty_ratelimit is always tracking balanced_dirty_ratelimit.\ntask_ratelimit is merely used as a limiting factor.\n\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "be3ffa276446e1b691a2bf84e7621e5a6fb49db9", "tree": "ca1b112195a9a8b63265f3204748cb23cff5b653", "parents": [ "af6a311384bce6c88e15c80ab22ab051a918b4eb" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Sun Jun 12 10:51:31 2011 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Mon Oct 03 21:08:56 2011 +0800" }, "message": "writeback: dirty rate control\n\nIt\u0027s all about bdi-\u003edirty_ratelimit, which aims to be (write_bw / N)\nwhen there are N dd tasks.\n\nOn write() syscall, use bdi-\u003edirty_ratelimit\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\n balance_dirty_pages(pages_dirtied)\n {\n task_ratelimit \u003d bdi-\u003edirty_ratelimit * bdi_position_ratio();\n pause \u003d pages_dirtied / task_ratelimit;\n sleep(pause);\n }\n\nOn every 200ms, update bdi-\u003edirty_ratelimit\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\n bdi_update_dirty_ratelimit()\n {\n task_ratelimit \u003d bdi-\u003edirty_ratelimit * bdi_position_ratio();\n balanced_dirty_ratelimit \u003d task_ratelimit * write_bw / dirty_rate;\n bdi-\u003edirty_ratelimit \u003d balanced_dirty_ratelimit\n }\n\nEstimation of balanced bdi-\u003edirty_ratelimit\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nbalanced task_ratelimit\n-----------------------\n\nbalance_dirty_pages() needs to throttle tasks dirtying pages such that\nthe total amount of dirty pages stays below the specified dirty limit in\norder to avoid memory deadlocks. Furthermore we desire fairness in that\ntasks get throttled proportionally to the amount of pages they dirty.\n\nIOW we want to throttle tasks such that we match the dirty rate to the\nwriteout bandwidth, this yields a stable amount of dirty pages:\n\n dirty_rate \u003d\u003d write_bw (1)\n\nThe fairness requirement gives us:\n\n task_ratelimit \u003d balanced_dirty_ratelimit\n \u003d\u003d write_bw / N (2)\n\nwhere N is the number of dd tasks. We don\u0027t know N beforehand, but\nstill can estimate balanced_dirty_ratelimit within 200ms.\n\nStart by throttling each dd task at rate\n\n task_ratelimit \u003d task_ratelimit_0 (3)\n (any non-zero initial value is OK)\n\nAfter 200ms, we measured\n\n dirty_rate \u003d # of pages dirtied by all dd\u0027s / 200ms\n write_bw \u003d # of pages written to the disk / 200ms\n\nFor the aggressive dd dirtiers, the equality holds\n\n dirty_rate \u003d\u003d N * task_rate\n \u003d\u003d N * task_ratelimit_0 (4)\nOr\n task_ratelimit_0 \u003d\u003d dirty_rate / N (5)\n\nNow we conclude that the balanced task ratelimit can be estimated by\n\n write_bw\n balanced_dirty_ratelimit \u003d task_ratelimit_0 * ---------- (6)\n dirty_rate\n\nBecause with (4) and (5) we can get the desired equality (1):\n\n write_bw\n balanced_dirty_ratelimit \u003d\u003d (dirty_rate / N) * ----------\n dirty_rate\n \u003d\u003d write_bw / N\n\nThen using the balanced task ratelimit we can compute task pause times like:\n\n task_pause \u003d task-\u003enr_dirtied / task_ratelimit\n\ntask_ratelimit with position control\n------------------------------------\n\nHowever, while the above gives us means of matching the dirty rate to\nthe writeout bandwidth, it at best provides us with a stable dirty page\ncount (assuming a static system). In order to control the dirty page\ncount such that it is high enough to provide performance, but does not\nexceed the specified limit we need another control.\n\nThe dirty position control works by extending (2) to\n\n task_ratelimit \u003d balanced_dirty_ratelimit * pos_ratio (7)\n\nwhere pos_ratio is a negative feedback function that subjects to\n\n1) f(setpoint) \u003d 1.0\n2) df/dx \u003c 0\n\nThat is, if the dirty pages are ABOVE the setpoint, we throttle each\ntask a bit more HEAVY than balanced_dirty_ratelimit, so that the dirty\npages are created less fast than they are cleaned, thus DROP to the\nsetpoints (and the reverse).\n\nBased on (7) and the assumption that both dirty_ratelimit and pos_ratio\nremains CONSTANT for the past 200ms, we get\n\n task_ratelimit_0 \u003d balanced_dirty_ratelimit * pos_ratio (8)\n\nPutting (8) into (6), we get the formula used in\nbdi_update_dirty_ratelimit():\n\n write_bw\n balanced_dirty_ratelimit *\u003d pos_ratio * ---------- (9)\n dirty_rate\n\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "c8e28ce049faa53a470c132893abbc9f2bde9420", "tree": "40553e72f50a613c93f4175d1d28d06abc90db65", "parents": [ "9b13776977d45505469edc6decc93e9e3799afe2" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Sun Jan 23 10:07:47 2011 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Mon Oct 03 21:08:56 2011 +0800" }, "message": "writeback: account per-bdi accumulated dirtied pages\n\nIntroduce the BDI_DIRTIED counter. It will be used for estimating the\nbdi\u0027s dirty bandwidth.\n\nCC: Jan Kara \u003cjack@suse.cz\u003e\nCC: Michael Rubin \u003cmrubin@google.com\u003e\nCC: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "60063497a95e716c9a689af3be2687d261f115b4", "tree": "6ce0d68db76982c53df46aee5f29f944ebf2c320", "parents": [ "148817ba092f9f6edd35bad3c6c6b8e8f90fe2ed" ], "author": { "name": "Arun Sharma", "email": "asharma@fb.com", "time": "Tue Jul 26 16:09:06 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 16:49:47 2011 -0700" }, "message": "atomic: use \u003clinux/atomic.h\u003e\n\nThis allows us to move duplicated code in \u003casm/atomic.h\u003e\n(atomic_inc_not_zero() for now) to \u003clinux/atomic.h\u003e\n\nSigned-off-by: Arun Sharma \u003casharma@fb.com\u003e\nReviewed-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nCc: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nAcked-by: Mike Frysinger \u003cvapier@gentoo.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e98be2d599207c6b31e9bb340d52a231b2f3662d", "tree": "3ae28e7d621a6e2ddf8e7462f8d282901c113d5c", "parents": [ "f7d2b1ecd0c714adefc7d3a942ef87beb828a763" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Sun Aug 29 11:22:30 2010 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Sat Jul 09 22:09:01 2011 -0700" }, "message": "writeback: bdi write bandwidth estimation\n\nThe estimation value will start from 100MB/s and adapt to the real\nbandwidth in seconds.\n\nIt tries to update the bandwidth only when disk is fully utilized.\nAny inactive period of more than one second will be skipped.\n\nThe estimated bandwidth will be reflecting how fast the device can\nwriteout when _fully utilized_, and won\u0027t drop to 0 when it goes idle.\nThe value will remain constant at disk idle time. At busy write time, if\nnot considering fluctuations, it will also remain high unless be knocked\ndown by possible concurrent reads that compete for the disk time and\nbandwidth with async writes.\n\nThe estimation is not done purely in the flusher because there is no\nguarantee for write_cache_pages() to return timely to update bandwidth.\n\nThe bdi-\u003eavg_write_bandwidth smoothing is very effective for filtering\nout sudden spikes, however may be a little biased in long term.\n\nThe overheads are low because the bdi bandwidth update only occurs at\n200ms intervals.\n\nThe 200ms update interval is suitable, because it\u0027s not possible to get\nthe real bandwidth for the instance at all, due to large fluctuations.\n\nThe NFS commits can be as large as seconds worth of data. One XFS\ncompletion may be as large as half second worth of data if we are going\nto increase the write chunk to half second worth of data. In ext4,\nfluctuations with time period of around 5 seconds is observed. And there\nis another pattern of irregular periods of up to 20 seconds on SSD tests.\n\nThat\u0027s why we are not only doing the estimation at 200ms intervals, but\nalso averaging them over a period of 3 seconds and then go further to do\nanother level of smoothing in avg_write_bandwidth.\n\nCC: Li Shaohua \u003cshaohua.li@intel.com\u003e\nCC: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "f7d2b1ecd0c714adefc7d3a942ef87beb828a763", "tree": "7c5adb0abd73d3ad449b94698dadbaceb573a6f4", "parents": [ "d46db3d58233be4be980eb1e42eebe7808bcabab" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Dec 08 22:44:24 2010 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Sat Jul 09 22:09:01 2011 -0700" }, "message": "writeback: account per-bdi accumulated written pages\n\nIntroduce the BDI_WRITTEN counter. It will be used for estimating the\nbdi\u0027s write bandwidth.\n\nPeter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e:\nMove BDI_WRITTEN accounting into __bdi_writeout_inc().\nThis will cover and fix fuse, which only calls bdi_writeout_inc().\n\nCC: Michael Rubin \u003cmrubin@google.com\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "f758eeabeb96f878c860e8f110f94ec8820822a9", "tree": "fea5a465aa0aa38c6c9263eb264acbeb7f722c02", "parents": [ "424b351fe1901fc909fd0ca4f21dab58f24c1aac" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Apr 21 18:19:44 2011 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Wed Jun 08 08:25:21 2011 +0800" }, "message": "writeback: split inode_wb_list_lock into bdi_writeback.list_lock\n\nSplit the global inode_wb_list_lock into a per-bdi_writeback list_lock,\nas it\u0027s currently the most contended lock in the system for metadata\nheavy workloads. It won\u0027t help for single-filesystem workloads for\nwhich we\u0027ll need the I/O-less balance_dirty_pages, but at least we\ncan dedicate a cpu to spinning on each bdi now for larger systems.\n\nBased on earlier patches from Nick Piggin and Dave Chinner.\n\nIt reduces lock contentions to 1/4 in this test case:\n10 HDD JBOD, 100 dd on each disk, XFS, 6GB ram\n\nlock_stat version 0.3\n-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total\n-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\nvanilla 2.6.39-rc3:\n inode_wb_list_lock: 42590 44433 0.12 147.74 144127.35 252274 886792 0.08 121.34 917211.23\n ------------------\n inode_wb_list_lock 2 [\u003cffffffff81165da5\u003e] bdev_inode_switch_bdi+0x29/0x85\n inode_wb_list_lock 34 [\u003cffffffff8115bd0b\u003e] inode_wb_list_del+0x22/0x49\n inode_wb_list_lock 12893 [\u003cffffffff8115bb53\u003e] __mark_inode_dirty+0x170/0x1d0\n inode_wb_list_lock 10702 [\u003cffffffff8115afef\u003e] writeback_single_inode+0x16d/0x20a\n ------------------\n inode_wb_list_lock 2 [\u003cffffffff81165da5\u003e] bdev_inode_switch_bdi+0x29/0x85\n inode_wb_list_lock 19 [\u003cffffffff8115bd0b\u003e] inode_wb_list_del+0x22/0x49\n inode_wb_list_lock 5550 [\u003cffffffff8115bb53\u003e] __mark_inode_dirty+0x170/0x1d0\n inode_wb_list_lock 8511 [\u003cffffffff8115b4ad\u003e] writeback_sb_inodes+0x10f/0x157\n\n2.6.39-rc3 + patch:\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock: 11383 11657 0.14 151.69 40429.51 90825 527918 0.11 145.90 556843.37\n ------------------------\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 10 [\u003cffffffff8115b189\u003e] inode_wb_list_del+0x5f/0x86\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 1493 [\u003cffffffff8115b1ed\u003e] writeback_inodes_wb+0x3d/0x150\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 3652 [\u003cffffffff8115a8e9\u003e] writeback_sb_inodes+0x123/0x16f\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 1412 [\u003cffffffff8115a38e\u003e] writeback_single_inode+0x17f/0x223\n ------------------------\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 3 [\u003cffffffff8110b5af\u003e] bdi_lock_two+0x46/0x4b\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 6 [\u003cffffffff8115b189\u003e] inode_wb_list_del+0x5f/0x86\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 2061 [\u003cffffffff8115af97\u003e] __mark_inode_dirty+0x173/0x1cf\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 2629 [\u003cffffffff8115a8e9\u003e] writeback_sb_inodes+0x123/0x16f\n\nhughd@google.com: fix recursive lock when bdi_lock_two() is called with new the same as old\nakpm@linux-foundation.org: cleanup bdev_inode_switch_bdi() comment\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "7eaceaccab5f40bbfda044629a6298616aeaed50", "tree": "33954d12f63e25a47eb6d86ef3d3d0a5e62bf752", "parents": [ "73c101011926c5832e6e141682180c4debe2cf45" ], "author": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:07 2011 +0100" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:07 2011 +0100" }, "message": "block: remove per-queue plugging\n\nCode has been converted over to the new explicit on-stack plugging,\nand delay users have been converted to use the new API for that.\nSo lets kill off the old plugging along with aops-\u003esync_page().\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "92c09c041f15fc88b35f8628e07639f52e1fbb38", "tree": "c9b91ace1405487a7a939229620d415461c76124", "parents": [ "e199b5d1fed13f5e8f47a0ee8216f36244dad1f4" ], "author": { "name": "Namhyung Kim", "email": "namhyung@gmail.com", "time": "Tue Oct 26 14:22:03 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 26 16:52:10 2010 -0700" }, "message": "mm: declare some external symbols\n\nDeclare \u0027bdi_pending_list\u0027 and \u0027tag_pages_for_writeback()\u0027 to remove\nfollowing sparse warnings:\n\n mm/backing-dev.c:46:1: warning: symbol \u0027bdi_pending_list\u0027 was not declared. Should it be static?\n mm/page-writeback.c:825:6: warning: symbol \u0027tag_pages_for_writeback\u0027 was not declared. Should it be static?\n\nSigned-off-by: Namhyung Kim \u003cnamhyung@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": "0e093d99763eb4cea09f8ca4f1d01f34e121d10b", "tree": "fad38f9c3651c81db298521141a79d9468f71986", "parents": [ "08fc468f4eaf6683bae5bdb94743a09d8630cb80" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Tue Oct 26 14:21:45 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 26 16:52:07 2010 -0700" }, "message": "writeback: do not sleep on the congestion queue if there are no congested BDIs or if significant congestion is not being encountered in the current zone\n\nIf congestion_wait() is called with no BDI congested, the caller will\nsleep for the full timeout and this may be an unnecessary sleep. This\npatch adds a wait_iff_congested() that checks congestion and only sleeps\nif a BDI is congested else, it calls cond_resched() to ensure the caller\nis not hogging the CPU longer than its quota but otherwise will not sleep.\n\nThis is aimed at reducing some of the major desktop stalls reported during\nIO. For example, while kswapd is operating, it calls congestion_wait()\nbut it could just have been reclaiming clean page cache pages with no\ncongestion. Without this patch, it would sleep for a full timeout but\nafter this patch, it\u0027ll just call schedule() if it has been on the CPU too\nlong. Similar logic applies to direct reclaimers that are not making\nenough progress.\n\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Jens Axboe \u003caxboe@kernel.dk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "81d73a32d775ae9674ea6edf0b5b721fc3bc57d9", "tree": "c4097c3c7715a128cf54252b2ff906d3d8946ae0", "parents": [ "a50aeb40144982eb766053309b6fc33e14ca46f0" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Aug 11 14:17:44 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Aug 12 08:43:30 2010 -0700" }, "message": "mm: fix writeback_in_progress()\n\nCommit 83ba7b071f3 (\"writeback: simplify the write back thread queue\")\nbroke writeback_in_progress() as in that commit we started to remove work\nitems from the list at the moment we start working on them and not at the\nmoment they are finished. Thus if the flusher thread was doing some work\nbut there was no other work queued, writeback_in_progress() returned\nfalse. This could in particular cause unnecessary queueing of background\nwriteback from balance_dirty_pages() or writeout work from\nwriteback_sb_if_idle().\n\nThis patch fixes the problem by introducing a bit in the bdi state which\nindicates that the flusher thread is processing some work and uses this\nbit for writeback_in_progress() test.\n\nNOTE: Both callsites of writeback_in_progress() (namely,\nwriteback_inodes_sb_if_idle() and balance_dirty_pages()) would actually\nneed a different information than what writeback_in_progress() provides.\nThey would need to know whether *the kind of writeback they are going to\nsubmit* is already queued. But this information isn\u0027t that simple to\nprovide so let\u0027s fix writeback_in_progress() for the time being.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nAcked-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6467716a37673e8d47b4984eb19839bdad0a8353", "tree": "8b2bfd38d53e31c47654162d4ce9220c4424a31d", "parents": [ "253c34e9b10c30d3064be654b5b78fbc1a8b1896" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Sun Jul 25 14:29:22 2010 +0300" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Sat Aug 07 18:53:56 2010 +0200" }, "message": "writeback: optimize periodic bdi thread wakeups\n\nWhe the first inode for a bdi is marked dirty, we wake up the bdi thread which\nshould take care of the periodic background write-out. However, the write-out\nwill actually start only \u0027dirty_writeback_interval\u0027 centisecs later, so we can\ndelay the wake-up.\n\nThis change was requested by Nick Piggin who pointed out that if we delay the\nwake-up, we weed out 2 unnecessary contex switches, which matters because\n\u0027__mark_inode_dirty()\u0027 is a hot-path function.\n\nThis patch introduces a new function - \u0027bdi_wakeup_thread_delayed()\u0027, which\nsets up a timer to wake-up the bdi thread and returns. So the wake-up is\ndelayed.\n\nWe also delete the timer in bdi threads just before writing-back. And\nsynchronously delete it when unregistering bdi. At the unregister point the bdi\ndoes not have any users, so no one can arm it again.\n\nSince now we take \u0027bdi-\u003ewb_lock\u0027 in the timer, which can execute in softirq\ncontext, we have to use \u0027spin_lock_bh()\u0027 for \u0027bdi-\u003ewb_lock\u0027. This patch makes\nthis change as well.\n\nThis patch also moves the \u0027bdi_wb_init()\u0027 function down in the file to avoid\nforward-declaration of \u0027bdi_wakeup_thread_delayed()\u0027.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "ecd584030da67ede1bf17955746a6ce834d9fc6b", "tree": "3f26260102a84e8cb047249e65cdde6b9aadcb96", "parents": [ "78c40cb6581a74adc48821f3de6b864a54d4c34d" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Sun Jul 25 14:29:18 2010 +0300" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Sat Aug 07 18:53:56 2010 +0200" }, "message": "writeback: move last_active to bdi\n\nCurrently bdi threads use local variable \u0027last_active\u0027 which stores last time\nwhen the bdi thread did some useful work. Move this local variable to \u0027struct\nbdi_writeback\u0027. This is just a preparation for the further patches which will\nmake the forker thread decide when bdi threads should be killed.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "080dcec41709be72613133f695be75b98dd43e88", "tree": "8f453a9fe9b2a144662aad4e06944722610a1056", "parents": [ "297252c81de8043ca6c36e5984c24fdb5aab9013" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Sun Jul 25 14:29:16 2010 +0300" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Sat Aug 07 18:53:56 2010 +0200" }, "message": "writeback: simplify bdi code a little\n\nThis patch simplifies bdi code a little by removing the \u0027pending_list\u0027 which is\nredundant. Indeed, currently the forker thread (\u0027bdi_forker_thread()\u0027) is\nworking like this:\n\n1. In a loop, fetch all bdi\u0027s which have works but have no writeback thread and\n move them to the \u0027pending_list\u0027.\n2. If the list is empty, sleep for 5 sec.\n3. Otherwise, take one bdi from the list, fork the writeback thread for this\n bdi, and repeat the loop.\n\nIOW, it first moves everything to the \u0027pending_list\u0027, then process only one\nelement, and so on. This patch simplifies the algorithm, which is now as\nfollows.\n\n1. Find the first bdi which has a work and remove it from the global list of\n bdi\u0027s (bdi_list).\n2. If there was not such bdi, sleep 5 sec.\n3. Fork the writeback thread for this bdi and repeat the loop.\n\nIOW, now we find the first bdi to process, process it, and so on. This is\nsimpler and involves less lists.\n\nThe bonus now is that we can get rid of a couple of functions, as well as\nremove complications which involve \u0027rcu_call()\u0027 and \u0027bdi-\u003ercu_head\u0027.\n\nThis patch also makes sure we use \u0027list_add_tail_rcu()\u0027, instead of plain\n\u0027list_add_tail()\u0027, but this piece of code is going to be removed in the next\npatch anyway.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "6f904ff0e39ea88f81eb77e8dfb4e1238492f0a8", "tree": "8a22ff5ffe31e221554915ac9135da15d3adad8d", "parents": [ "4aeefdc69f7b6f3f287e6fd8d4b213953b9e92d8" ], "author": { "name": "Artem Bityutskiy", "email": "Artem.Bityutskiy@nokia.com", "time": "Sun Jul 25 14:29:11 2010 +0300" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Sat Aug 07 18:53:16 2010 +0200" }, "message": "writeback: harmonize writeback threads naming\n\nThe write-back code mixes words \"thread\" and \"task\" for the same things. This\nis not a big deal, but still an inconsistency.\n\nhch: a convention I tend to use and I\u0027ve seen in various places\nis to always use _task for the storage of the task_struct pointer,\nand thread everywhere else. This especially helps with having\nfoo_thread for the actual thread and foo_task for a global\nvariable keeping the task_struct pointer\n\nThis patch renames:\n* \u0027bdi_add_default_flusher_task()\u0027 -\u003e \u0027bdi_add_default_flusher_thread()\u0027\n* \u0027bdi_forker_task()\u0027 -\u003e \u0027bdi_forker_thread()\u0027\n\nbecause bdi threads are \u0027bdi_writeback_thread()\u0027, so these names are more\nconsistent.\n\nThis patch also amends commentaries and makes them refer the forker and bdi\nthreads as \"thread\", not \"task\".\n\nAlso, while on it, make \u0027bdi_add_default_flusher_thread()\u0027 declaration use\n\u0027static void\u0027 instead of \u0027void static\u0027 and make checkpatch.pl happy.\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "082439004b31adc146e96e5f1c574dd2b57dcd93", "tree": "19ef2ccab9588c2aa77d0b09c2bae760b23ae110", "parents": [ "c1955ce32fdb0877b7a1b22feb2669358f65be76" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Sat Jun 19 23:08:22 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Sat Aug 07 18:23:06 2010 +0200" }, "message": "writeback: merge bdi_writeback_task and bdi_start_fn\n\nMove all code for the writeback thread into fs/fs-writeback.c instead of\nsplitting it over two functions in two files.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "c1955ce32fdb0877b7a1b22feb2669358f65be76", "tree": "805b4da7168b0b16de890ea248323f3cdf5dc0b7", "parents": [ "4c4762d10faf93167b84ee03e4b58dd62012b06f" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Sat Jun 19 23:08:06 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Sat Aug 07 18:23:03 2010 +0200" }, "message": "writeback: remove wb_list\n\nThe wb_list member of struct backing_device_info always has exactly one\nelement. Just use the direct bdi-\u003ewb pointer instead and simplify some\ncode.\n\nAlso remove bdi_task_init which is now trivial to prepare for the next\npatch.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "83ba7b071f30f7c01f72518ad72d5cd203c27502", "tree": "4737320dcce72cfff4d87d835e4f78428eca7ef5", "parents": [ "edadfb10ba35da7253541e4155aa92eff758ebe6" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Tue Jul 06 08:59:53 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Tue Jul 06 08:59:53 2010 +0200" }, "message": "writeback: simplify the write back thread queue\n\nFirst remove items from work_list as soon as we start working on them. This\nmeans we don\u0027t have to track any pending or visited state and can get\nrid of all the RCU magic freeing the work items - we can simply free\nthem once the operation has finished. Second use a real completion for\ntracking synchronous requests - if the caller sets the completion pointer\nwe complete it, otherwise use it as a boolean indicator that we can free\nthe work item directly. Third unify struct wb_writeback_args and struct\nbdi_work into a single data structure, wb_writeback_work. Previous we\nset all parameters into a struct wb_writeback_args, copied it into\nstruct bdi_work, copied it again on the stack to use it there. Instead\nof just allocate one structure dynamically or on the stack and use it\nall the way through the stack.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "c5444198ca210498e8ac0ba121b4cd3537aa12f7", "tree": "c423d38fe1ac7f51a48e455a19ecbe2354811fca", "parents": [ "b8c2f3474f1077599ec6e90c2f263f17055cc3d8" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Tue Jun 08 18:15:15 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Fri Jun 11 12:58:08 2010 +0200" }, "message": "writeback: simplify and split bdi_start_writeback\n\nbdi_start_writeback now never gets a superblock passed, so we can just remove\nthat case. And to further untangle the code and flatten the call stack\nsplit it into two trivial helpers for it\u0027s two callers.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "0e3c9a2284f5417f196e327c254d0b84c9ee8929", "tree": "e3fb40ebe7d042b4b3c1042bc7f2edaf7fb6eee0", "parents": [ "f17625b318d9b151e7bd41e31223e9d89b2aaa77" ], "author": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Tue Jun 01 11:08:43 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Tue Jun 01 11:08:43 2010 +0200" }, "message": "Revert \"writeback: fix WB_SYNC_NONE writeback from umount\"\n\nThis reverts commit e913fc825dc685a444cb4c1d0f9d32f372f59861.\n\nWe are investigating a hang associated with the WB_SYNC_NONE changes,\nso revert them for now.\n\nConflicts:\n\n\tfs/fs-writeback.c\n\tmm/page-writeback.c\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "6423104b6a1e6f0c18be60e8c33f02d263331d5e", "tree": "e22957400e9679bf82b62e03d6bd831181053945", "parents": [ "f9eadbbd424c083b8005c7b738f644611b9ef489" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri May 21 20:00:35 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri May 21 20:00:35 2010 +0200" }, "message": "writeback: fixups for !dirty_writeback_centisecs\n\nCommit 69b62d01 fixed up most of the places where we would enter\nbusy schedule() spins when disabling the periodic background\nwriteback. This fixes up the sb timer so that it doesn\u0027t get\nhammered on with the delay disabled, and ensures that it gets\nrearmed if needed when /proc/sys/vm/dirty_writeback_centisecs\ngets modified.\n\nbdi_forker_task() also needs to check for !dirty_writeback_centisecs\nand use schedule() appropriately, fix that up too.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "e913fc825dc685a444cb4c1d0f9d32f372f59861", "tree": "e470697e43ffe4028ac81c17d3ef90ee9f30bcfb", "parents": [ "69b62d01ec44fe0d505d89917392347732135a4d" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon May 17 12:55:07 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon May 17 12:55:07 2010 +0200" }, "message": "writeback: fix WB_SYNC_NONE writeback from umount\n\nWhen umount calls sync_filesystem(), we first do a WB_SYNC_NONE\nwriteback to kick off writeback of pending dirty inodes, then follow\nthat up with a WB_SYNC_ALL to wait for it. Since umount already holds\nthe sb s_umount mutex, WB_SYNC_NONE ends up doing nothing and all\nwriteback happens as WB_SYNC_ALL. This can greatly slow down umount,\nsince WB_SYNC_ALL writeback is a data integrity operation and thus\na bigger hammer than simple WB_SYNC_NONE. For barrier aware file systems\nit\u0027s a lot slower.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "7407cf355fdf5500430be966dbbde84a27293bad", "tree": "922861288ff38558ed721a79653f52b17b13bb95", "parents": [ "6a47dc1418682c83d603b491df1d048f73aa973e", "79dba2eaa771c3173957eccfd288e0e0d12e4d3f" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Apr 29 09:36:24 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Apr 29 09:36:24 2010 +0200" }, "message": "Merge branch \u0027master\u0027 into for-2.6.35\n\nConflicts:\n\tfs/block_dev.c\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "5129a469a91a91427334c40e29e64c6d0ab68caf", "tree": "141a17cbea94c7c9c038187cc7081e1c688eac55", "parents": [ "7e2455c1a123ceadbb35150a610d61e8443fd340" ], "author": { "name": "Jörn Engel", "email": "joern@logfs.org", "time": "Sun Apr 25 08:54:42 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Sun Apr 25 08:54:42 2010 +0200" }, "message": "Catch filesystems lacking s_bdi\n\nnoop_backing_dev_info is used only as a flag to mark filesystems that\ndon\u0027t have any backing store, like tmpfs, procfs, spufs, etc.\n\nSigned-off-by: Joern Engel \u003cjoern@logfs.org\u003e\n\nChanged the BUG_ON() to a WARN_ON(). Note that adding dirty inodes\nto the noop_backing_dev_info is not legal and will not result in\nthem being flushed, but we already catch this condition in\n__mark_inode_dirty() when checking for a registered bdi.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "c3c532061e46156e8aab1268f38d66cfb63aeb2d", "tree": "8c5b027873b823b04f5564d442477e80e82e8edb", "parents": [ "a534dbe96e9929c7245924d8252d89048c23d569" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Apr 22 11:37:01 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Apr 22 11:39:36 2010 +0200" }, "message": "bdi: add helper function for doing init and register of a bdi for a file system\n\nPretty trivial helper, just sets up the bdi and registers it. An atomic\nsequence count is used to ensure that the registered sysfs names are\nunique.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "31373d09da5b7fe21fe6f781e92bd534a3495f00", "tree": "38cd9896cfc6ce106a03431658a9b98a09129034", "parents": [ "9195291e5f05e01d67f9a09c756b8aca8f009089" ], "author": { "name": "Matthew Garrett", "email": "mjg@redhat.com", "time": "Tue Apr 06 14:25:14 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Apr 06 14:25:14 2010 +0200" }, "message": "laptop-mode: Make flushes per-device\n\nOne of the features of laptop-mode is that it forces a writeout of dirty\npages if something else triggers a physical read or write from a device.\nThe current implementation flushes pages on all devices, rather than only\nthe one that triggered the flush. This patch alters the behaviour so that\nonly the recently accessed block device is flushed, preventing other\ndisks being spun up for no terribly good reason.\n\nSigned-off-by: Matthew Garrett \u003cmjg@redhat.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "b9d128f1088ea5245109dfc9bbceb128b6371a77", "tree": "63e85f281dd372e77af4022e2235a4e5a46287de", "parents": [ "a870a3a485ddf7c0dec549269ed71d169556d61c" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 29 13:59:26 2009 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 29 13:59:26 2009 +0100" }, "message": "block: move bdi/address_space unplug functions to backing-dev.h\n\nThere\u0027s nothing block related about them, the backing device\nis used by things like NFS etc as well. This gets rid of the\nneed to protect such calls by CONFIG_BLOCK.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "a72bfd4dea053bb8e2233902c3f1893ef5485802", "tree": "1246fc000adfee6d2874b9324eaf7383ad4413bb", "parents": [ "6d7f18f6ea3a13af95bdf507fc54d42b165e1712" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Sat Sep 26 00:07:46 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Sat Sep 26 00:10:40 2009 +0200" }, "message": "writeback: pass in super_block to bdi_start_writeback()\n\nSometimes we only want to write pages from a specific super_block,\nso allow that to be passed in.\n\nThis fixes a problem with commit 56a131dcf7ed36c3c6e36bea448b674ea85ed5bb\ncausing writeback on all super_blocks on a bdi, where we only really\nwant to sync a specific sb from writeback_inodes_sb().\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "b6e51316daede0633e9274e1e30391cfa4747877", "tree": "664476bb4e7c05dcce3ad908363b482134c68429", "parents": [ "bcddc3f01c9122882c8b9f12ab94a934e55aef97" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 16 15:13:54 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 16 15:18:52 2009 +0200" }, "message": "writeback: separate starting of sync vs opportunistic writeback\n\nbdi_start_writeback() is currently split into two paths, one for\nWB_SYNC_NONE and one for WB_SYNC_ALL. Add bdi_sync_writeback()\nfor WB_SYNC_ALL writeback and let bdi_start_writeback() handle\nonly WB_SYNC_NONE.\n\nPush down the writeback_control allocation and only accept the\nparameters that make sense for each function. This cleans up\nthe API considerably.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "cfc4ba5365449cb6b5c9f68d755a142f17da1e47", "tree": "08770de9bb0e658f2e65abd4d10187b3e9f6bb1b", "parents": [ "f11fcae8401a3175f528e2f7917362645d570111" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Sep 14 13:12:40 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 16 15:18:51 2009 +0200" }, "message": "writeback: use RCU to protect bdi_list\n\nNow that bdi_writeback_all() no longer handles integrity writeback,\nit doesn\u0027t have to block anymore. This means that we can switch\nbdi_list reader side protection to RCU.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "500b067c5e6ceea49cf280a02597b1169320e08c", "tree": "5e2e9bfd66f825f845fdf395b6303b884769ff6d", "parents": [ "d993831fa7ffeb89e994f046f93eeb09ec91df08" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 09 09:10:25 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Sep 11 09:20:26 2009 +0200" }, "message": "writeback: check for registered bdi in flusher add and inode dirty\n\nAlso a debugging aid. We want to catch dirty inodes being added to\nbacking devices that don\u0027t do writeback.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "d993831fa7ffeb89e994f046f93eeb09ec91df08", "tree": "da4f94bbf022c83988bda71adf1f1b3a88cb4592", "parents": [ "f09b00d3e789a88fa6c7c03cedc62cb65c1de0cb" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Jun 12 14:45:52 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Sep 11 09:20:26 2009 +0200" }, "message": "writeback: add name to backing_dev_info\n\nThis enables us to track who does what and print info. Its main use\nis catching dirty inodes on the default_backing_dev_info, so we can\nfix that up.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\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": "66f3b8e2e103a0b93b945764d98e9ba46cb926dd", "tree": "442bf5664214f0a1448e4010b09868cc58fdd3d1", "parents": [ "d8a8559cd7a9ccac98d5f6f13297a2ff68a43627" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 02 09:19:46 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Sep 11 09:20:25 2009 +0200" }, "message": "writeback: move dirty inodes from super_block to backing_dev_info\n\nThis is a first step at introducing per-bdi flusher threads. We should\nhave no change in behaviour, although sb_has_dirty_inodes() is now\nridiculously expensive, as there\u0027s no easy way to answer that question.\nNot a huge problem, since it\u0027ll be deleted in subsequent patches.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "373c0a7ed3ea3b34efedb7c83ffb521adff7c894", "tree": "1fcc8cdcf8deab3ed1d16ff3c0f2dfa38bcf7915", "parents": [ "f00caa76297a534e30c3ec85382fbc3fd0dcbc69" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Jul 11 10:06:54 2009 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 11 11:22:26 2009 -0700" }, "message": "Fix compile error due to congestion_wait() changes\n\nMove the definition of BLK_RW_ASYNC/BLK_RW_SYNC into linux/backing-dev.h\nso that it is available to all callers of set/clear_bdi_congested().\n\nThis replaces commit 097041e576ee3a50d92dd643ee8ca65bf6a62e21 (\"fuse:\nFix build error\"), which will be reverted.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nAcked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nCc: Miklos Szeredi \u003cmiklos@szeredi.hu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8aa7e847d834ed937a9ad37a0f2ad5b8584c1ab0", "tree": "76c8b4f1362a928d426f2201790ab5d128f57724", "parents": [ "c2cc49a2f8a479dde96a599646d30b6cc9dbed78" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Jul 09 14:52:32 2009 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@carl.(none)", "time": "Fri Jul 10 20:31:53 2009 +0200" }, "message": "Fix congestion_wait() sync/async vs read/write confusion\n\nCommit 1faa16d22877f4839bd433547d770c676d1d964c accidentally broke\nthe bdi congestion wait queue logic, causing us to wait on congestion\nfor WRITE (\u003d\u003d 1) when we really wanted BLK_RW_ASYNC (\u003d\u003d 0) instead.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "1faa16d22877f4839bd433547d770c676d1d964c", "tree": "9a0d50be1ef0358c1f53d7107413100904e7d526", "parents": [ "0221c81b1b8eb0cbb6b30a0ced52ead32d2b4e4c" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Apr 06 14:48:01 2009 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 06 08:04:53 2009 -0700" }, "message": "block: change the request allocation/congestion logic to be sync/async based\n\nThis makes sure that we never wait on async IO for sync requests, instead\nof doing the split on writes vs reads.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "4f98a2fee8acdb4ac84545df98cccecfd130f8db", "tree": "035a2937f4c3e2f7b4269412041c073ac646937c", "parents": [ "b2e185384f534781fd22f5ce170b2ad26f97df70" ], "author": { "name": "Rik van Riel", "email": "riel@redhat.com", "time": "Sat Oct 18 20:26:32 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Oct 20 08:50:25 2008 -0700" }, "message": "vmscan: split LRU lists into anon \u0026 file sets\n\nSplit the LRU lists in two, one set for pages that are backed by real file\nsystems (\"file\") and one for pages that are backed by memory and swap\n(\"anon\"). The latter includes tmpfs.\n\nThe advantage of doing this is that the VM will not have to scan over lots\nof anonymous pages (which we generally do not want to swap out), just to\nfind the page cache pages that it should evict.\n\nThis patch has the infrastructure and a basic policy to balance how much\nwe scan the anon lists and how much we scan the file lists. The big\npolicy changes are in separate patches.\n\n[lee.schermerhorn@hp.com: collect lru meminfo statistics from correct offset]\n[kosaki.motohiro@jp.fujitsu.com: prevent incorrect oom under split_lru]\n[kosaki.motohiro@jp.fujitsu.com: fix pagevec_move_tail() doesn\u0027t treat unevictable page]\n[hugh@veritas.com: memcg swapbacked pages active]\n[hugh@veritas.com: splitlru: BDI_CAP_SWAP_BACKED]\n[akpm@linux-foundation.org: fix /proc/vmstat units]\n[nishimura@mxp.nes.nec.co.jp: memcg: fix handling of shmem migration]\n[kosaki.motohiro@jp.fujitsu.com: adjust Quicklists field of /proc/meminfo]\n[kosaki.motohiro@jp.fujitsu.com: fix style issue of get_scan_ratio()]\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Daisuke Nishimura \u003cnishimura@mxp.nes.nec.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": "dd5656e59ca7b25fb60a22f9079905ed0da5ed0c", "tree": "47c156e6023e93f3028c63a2594bbb3e27e968a4", "parents": [ "e4ad08fe64afca4ef79ecc4c624e6e871688da0d" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Wed Apr 30 00:54:37 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:50 2008 -0700" }, "message": "mm: bdi: export bdi_writeout_inc()\n\nFuse needs this for writable mmap support.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nCc: 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": "e4ad08fe64afca4ef79ecc4c624e6e871688da0d", "tree": "5b8b390b874700041dc0c095e8ba9ac3ed42ea77", "parents": [ "76f1418b485da2707531178e517bbb5cf06b3c76" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Wed Apr 30 00:54:37 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:50 2008 -0700" }, "message": "mm: bdi: add separate writeback accounting capability\n\nAdd a new BDI capability flag: BDI_CAP_NO_ACCT_WB. If this flag is\nset, then don\u0027t update the per-bdi writeback stats from\ntest_set_page_writeback() and test_clear_page_writeback().\n\nMisc cleanups:\n\n - convert bdi_cap_writeback_dirty() and friends to static inline functions\n - create a flag that includes all three dirty/writeback related flags,\n since almst all users will want to have them toghether\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nCc: 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": "76f1418b485da2707531178e517bbb5cf06b3c76", "tree": "138a012d965ebc284163f42e0ff5786c6de1cda5", "parents": [ "a42dde04152750426cc620fd277e80fffae2f65a" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Wed Apr 30 00:54:36 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:50 2008 -0700" }, "message": "mm: bdi: move statistics to debugfs\n\nMove BDI statistics to debugfs:\n\n /sys/kernel/debug/bdi/\u003cbdi\u003e/stats\n\nUse postcore_initcall() to initialize the sysfs class and debugfs,\nbecause debugfs is initialized in core_initcall().\n\nUpdate descriptions in ABI documentation.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@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": "a42dde04152750426cc620fd277e80fffae2f65a", "tree": "98cdf246280c59e0bf044ae6d2c229fc27b73203", "parents": [ "189d3c4a94ef19fca2a71a6a336e9fda900e25e7" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Apr 30 00:54:36 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:50 2008 -0700" }, "message": "mm: bdi: allow setting a maximum for the bdi dirty limit\n\nAdd \"max_ratio\" to /sys/class/bdi. This indicates the maximum percentage of\nthe global dirty threshold allocated to this bdi.\n\n[mszeredi@suse.cz]\n\n - fix parsing in max_ratio_store().\n - export bdi_set_max_ratio() to modules\n - limit bdi_dirty with bdi-\u003emax_ratio\n - document new sysfs attribute\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@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": "189d3c4a94ef19fca2a71a6a336e9fda900e25e7", "tree": "75c0de871fe9922885a2fa073f15806d829342fa", "parents": [ "b6f2fcbcfca9db2bd7aa24940224fcd3bbdbb8aa" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Apr 30 00:54:35 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:50 2008 -0700" }, "message": "mm: bdi: allow setting a minimum for the bdi dirty limit\n\nUnder normal circumstances each device is given a part of the total write-back\ncache that relates to its current avg writeout speed in relation to the other\ndevices.\n\nmin_ratio - allows one to assign a minimum portion of the write-back cache to\na particular device. This is useful in situations where you might want to\nprovide a minimum QoS. (One request for this feature came from flash based\nstorage people who wanted to avoid writing out at all costs - they of course\nneeded some pdflush hacks as well)\n\nmax_ratio - allows one to assign a maximum portion of the dirty limit to a\nparticular device. This is useful in situations where you want to avoid one\ndevice taking all or most of the write-back cache. Eg. an NFS mount that is\nprone to get stuck, or a FUSE mount which you don\u0027t trust to play fair.\n\nAdd \"min_ratio\" to /sys/class/bdi. This indicates the minimum percentage of\nthe global dirty threshold allocated to this bdi.\n\n[mszeredi@suse.cz]\n\n - fix parsing in min_ratio_store()\n - document new sysfs attribute\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@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": "cf0ca9fe5dd9e3693d935757a7b2fc50fc576554", "tree": "c795c5271eda9fc67579fa3176c646b892dfdb41", "parents": [ "caafa4324335aeb11bc233d5f87aca8cce30beba" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Wed Apr 30 00:54:32 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 30 08:29:49 2008 -0700" }, "message": "mm: bdi: export BDI attributes in sysfs\n\nProvide a place in sysfs (/sys/class/bdi) for the backing_dev_info object.\nThis allows us to see and set the various BDI specific variables.\n\nIn particular this properly exposes the read-ahead window for all relevant\nusers and /sys/block/\u003cblock\u003e/queue/read_ahead_kb should be deprecated.\n\nWith patient help from Kay Sievers and Greg KH\n\n[mszeredi@suse.cz]\n\n - split off NFS and FUSE changes into separate patches\n - document new sysfs attributes under Documentation/ABI\n - do bdi_class_init as a core_initcall, otherwise the \"default\" BDI\n won\u0027t be initialized\n - remove bdi_init_fmt macro, it\u0027s not used very much\n\n[akpm@linux-foundation.org: fix ia64 warning]\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nAcked-by: Greg KH \u003cgreg@kroah.com\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@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": "f11b00f3bd89c91c684d56b2082d1b0241ff20ae", "tree": "59f9be385f05d12d88d0ce70306687b64cbab3b8", "parents": [ "67cde595374dd0e4e4a537dbf9dff70fd3d7bd7b" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Tue Apr 29 00:58:56 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 29 08:06:00 2008 -0700" }, "message": "fs/fs-writeback.c: make 2 functions static\n\nMake the following needlessly global functions static:\n\n- writeback_acquire()\n- writeback_release()\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": "04fbfdc14e5f48463820d6b9807daa5e9c92c51f", "tree": "c62905212c8c6373b2258c7f528398d3c831b075", "parents": [ "145ca25eb2fbd20d4faf1bad4628c7650332058f" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue Oct 16 23:25:50 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 08:42:45 2007 -0700" }, "message": "mm: per device dirty threshold\n\nScale writeback cache per backing device, proportional to its writeout speed.\n\nBy decoupling the BDI dirty thresholds a number of problems we currently have\nwill go away, namely:\n\n - mutual interference starvation (for any number of BDIs);\n - deadlocks with stacked BDIs (loop, FUSE and local NFS mounts).\n\nIt might be that all dirty pages are for a single BDI while other BDIs are\nidling. By giving each BDI a \u0027fair\u0027 share of the dirty limit, each one can have\ndirty pages outstanding and make progress.\n\nA global threshold also creates a deadlock for stacked BDIs; when A writes to\nB, and A generates enough dirty pages to get throttled, B will never start\nwriteback until the dirty pages go away. Again, by giving each BDI its own\n\u0027independent\u0027 dirty limit, this problem is avoided.\n\nSo the problem is to determine how to distribute the total dirty limit across\nthe BDIs fairly and efficiently. A DBI that has a large dirty limit but does\nnot have any dirty pages outstanding is a waste.\n\nWhat is done is to keep a floating proportion between the DBIs based on\nwriteback completions. This way faster/more active devices get a larger share\nthan slower/idle devices.\n\n[akpm@linux-foundation.org: fix warnings]\n[hugh@veritas.com: Fix occasional hang when a task couldn\u0027t get out of balance_dirty_pages]\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\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": "69cb51d18c1ed593009d9a620cac49d0dcf15dc8", "tree": "f26a1245b638e7421a0afd93305fd1c556867d49", "parents": [ "c9e51e4180696aa67915ec5665e4ec74125565de" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue Oct 16 23:25:48 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 08:42:45 2007 -0700" }, "message": "mm: count writeback pages per BDI\n\nCount per BDI writeback pages.\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": "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": "b2e8fb6efa209c82203c79b491b5bc952d44aa57", "tree": "e4c82e3f2c70a188ced7126435bf153f5f6c4a1c", "parents": [ "e0bf68ddec4f4f90e5871404be4f1854c17f3120" ], "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: scalable bdi statistics counters\n\nProvide scalable per backing_dev_info statistics counters.\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": "e0bf68ddec4f4f90e5871404be4f1854c17f3120", "tree": "36203a3558cbe26d698bed18be69b3822fb5eef2", "parents": [ "dc62a30e274d003a4d08fb888f1520add4b21373" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue Oct 16 23:25:46 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 08:42:45 2007 -0700" }, "message": "mm: bdi init hooks\n\nprovide BDI constructor/destructor hooks\n\n[akpm@linux-foundation.org: compile fix]\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": "c4dc4beed23827e155d7cbc2a1ffa3949eddd194", "tree": "b7f115c159fe26b90b7e5b7d5dcf2f9318b69bf1", "parents": [ "1f7d6668c29b1dfa307a44844f9bb38356fc989b" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue Oct 16 23:25:41 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 08:42:44 2007 -0700" }, "message": "nfs: remove congestion_end()\n\nThese patches aim to improve balance_dirty_pages() and directly address three\nissues:\n 1) inter device starvation\n 2) stacked device deadlocks\n 3) inter process starvation\n\n1 and 2 are a direct result from removing the global dirty limit and using\nper device dirty limits. By giving each device its own dirty limit is will\nno longer starve another device, and the cyclic dependancy on the dirty limit\nis broken.\n\nIn order to efficiently distribute the dirty limit across the independant\ndevices a floating proportion is used, this will allocate a share of the total\nlimit proportional to the device\u0027s recent activity.\n\n3 is done by also scaling the dirty limit proportional to the current task\u0027s\nrecent dirty rate.\n\nThis patch:\n\nnfs: remove congestion_end(). It\u0027s redundant, clear_bdi_congested() already\nwakes the waiters.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nCc: \"J. Bruce Fields\" \u003cbfields@fieldses.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8f8a68ee486e1c81eaead3c521822bf86142d380", "tree": "8435b491a52ea22918f621892970536957352fa0", "parents": [ "1492192b4a0bb84dd9b792cc0bd30583220a28a7" ], "author": { "name": "Adrian Bunk", "email": "bunk@stusta.de", "time": "Sun Jul 15 23:41:52 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Jul 16 09:05:52 2007 -0700" }, "message": "remove mm/backing-dev.c:congestion_wait_interruptible()\n\ncongestion_wait_interruptible() is no longer used.\n\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\nAcked-by: 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": "89a09141df6ac1c3821fbe44ca8384eb37692965", "tree": "ccb21055fca86ac2657b3262ac37eb3e5c44bea0", "parents": [ "b74a2f0913694556a027795d2954d30523fac4c5" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Fri Mar 16 13:38:26 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Mar 16 19:25:05 2007 -0700" }, "message": "[PATCH] nfs: fix congestion control\n\nThe current NFS client congestion logic is severly broken, it marks the\nbacking device congested during each nfs_writepages() call but doesn\u0027t\nmirror this in nfs_writepage() which makes for deadlocks. Also it\nimplements its own waitqueue.\n\nReplace this by a more regular congestion implementation that puts a cap on\nthe number of active writeback pages and uses the bdi congestion waitqueue.\n\nAlso always use an interruptible wait since it makes sense to be able to\nSIGKILL the process even for mounts without \u0027intr\u0027.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nAcked-by: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nCc: Christoph Lameter \u003cclameter@engr.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": "3fcfab16c5b86eaa3db3a9a31adba550c5b67141", "tree": "bd348fa081b8fbec2c79fbf8f173a306d70b2b2c", "parents": [ "79e2de4bc53d7ca2a8eedee49e4a92479b4b530e" ], "author": { "name": "Andrew Morton", "email": "akpm@osdl.org", "time": "Thu Oct 19 23:28:16 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Oct 20 10:26:35 2006 -0700" }, "message": "[PATCH] separate bdi congestion functions from queue congestion functions\n\nSeparate out the concept of \"queue congestion\" from \"backing-dev congestion\".\nCongestion is a backing-dev concept, not a queue concept.\n\nThe blk_* congestion functions are retained, as wrappers around the core\nbacking-dev congestion functions.\n\nThis proper layering is needed so that NFS can cleanly use the congestion\nfunctions, and so that CONFIG_BLOCK\u003dn actually links.\n\nCc: \"Thomas Maier\" \u003cbalagi@justmail.de\u003e\nCc: \"Jens Axboe\" \u003cjens.axboe@oracle.com\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Peter Osterlund \u003cpetero2@telia.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2", "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d", "parents": [], "author": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n" } ] }