linux-stable/mm/backing-dev.c

1078 lines
27 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
#include <linux/wait.h>
#include <linux/rbtree.h>
#include <linux/backing-dev.h>
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
#include <linux/kthread.h>
#include <linux/freezer.h>
#include <linux/fs.h>
#include <linux/pagemap.h>
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
#include <linux/mm.h>
#include <linux/sched/mm.h>
#include <linux/sched.h>
#include <linux/module.h>
#include <linux/writeback.h>
#include <linux/device.h>
#include <trace/events/writeback.h>
struct backing_dev_info noop_backing_dev_info;
EXPORT_SYMBOL_GPL(noop_backing_dev_info);
static struct class *bdi_class;
static const char *bdi_unknown_name = "(unknown)";
/*
* bdi_lock protects bdi_tree and updates to bdi_list. bdi_list has RCU
* reader side locking.
*/
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
DEFINE_SPINLOCK(bdi_lock);
static u64 bdi_id_cursor;
static struct rb_root bdi_tree = RB_ROOT;
LIST_HEAD(bdi_list);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
writeback: replace custom worker pool implementation with unbound workqueue Writeback implements its own worker pool - each bdi can be associated with a worker thread which is created and destroyed dynamically. The worker thread for the default bdi is always present and serves as the "forker" thread which forks off worker threads for other bdis. there's no reason for writeback to implement its own worker pool when using unbound workqueue instead is much simpler and more efficient. This patch replaces custom worker pool implementation in writeback with an unbound workqueue. The conversion isn't too complicated but the followings are worth mentioning. * bdi_writeback->last_active, task and wakeup_timer are removed. delayed_work ->dwork is added instead. Explicit timer handling is no longer necessary. Everything works by either queueing / modding / flushing / canceling the delayed_work item. * bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off bdi_writeback->dwork. On each execution, it processes bdi->work_list and reschedules itself if there are more things to do. The function also handles low-mem condition, which used to be handled by the forker thread. If the function is running off a rescuer thread, it only writes out limited number of pages so that the rescuer can serve other bdis too. This preserves the flusher creation failure behavior of the forker thread. * INIT_LIST_HEAD(&bdi->bdi_list) is used to tell bdi_writeback_workfn() about on-going bdi unregistration so that it always drains work_list even if it's running off the rescuer. Note that the original code was broken in this regard. Under memory pressure, a bdi could finish unregistration with non-empty work_list. * The default bdi is no longer special. It now is treated the same as any other bdi and bdi_cap_flush_forker() is removed. * BDI_pending is no longer used. Removed. * Some tracepoints become non-applicable. The following TPs are removed - writeback_nothread, writeback_wake_thread, writeback_wake_forker_thread, writeback_thread_start, writeback_thread_stop. Everything, including devices coming and going away and rescuer operation under simulated memory pressure, seems to work fine in my test setup. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Fengguang Wu <fengguang.wu@intel.com> Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-02 02:08:06 +00:00
/* bdi_wq serves all asynchronous writeback tasks */
struct workqueue_struct *bdi_wq;
#define K(x) ((x) << (PAGE_SHIFT - 10))
#ifdef CONFIG_DEBUG_FS
#include <linux/debugfs.h>
#include <linux/seq_file.h>
static struct dentry *bdi_debug_root;
static void bdi_debug_init(void)
{
bdi_debug_root = debugfs_create_dir("bdi", NULL);
}
static int bdi_debug_stats_show(struct seq_file *m, void *v)
{
struct backing_dev_info *bdi = m->private;
struct bdi_writeback *wb = &bdi->wb;
unsigned long background_thresh;
unsigned long dirty_thresh;
unsigned long wb_thresh;
unsigned long nr_dirty, nr_io, nr_more_io, nr_dirty_time;
struct inode *inode;
nr_dirty = nr_io = nr_more_io = nr_dirty_time = 0;
writeback: split inode_wb_list_lock into bdi_writeback.list_lock Split the global inode_wb_list_lock into a per-bdi_writeback list_lock, as it's currently the most contended lock in the system for metadata heavy workloads. It won't help for single-filesystem workloads for which we'll need the I/O-less balance_dirty_pages, but at least we can dedicate a cpu to spinning on each bdi now for larger systems. Based on earlier patches from Nick Piggin and Dave Chinner. It reduces lock contentions to 1/4 in this test case: 10 HDD JBOD, 100 dd on each disk, XFS, 6GB ram lock_stat version 0.3 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- vanilla 2.6.39-rc3: inode_wb_list_lock: 42590 44433 0.12 147.74 144127.35 252274 886792 0.08 121.34 917211.23 ------------------ inode_wb_list_lock 2 [<ffffffff81165da5>] bdev_inode_switch_bdi+0x29/0x85 inode_wb_list_lock 34 [<ffffffff8115bd0b>] inode_wb_list_del+0x22/0x49 inode_wb_list_lock 12893 [<ffffffff8115bb53>] __mark_inode_dirty+0x170/0x1d0 inode_wb_list_lock 10702 [<ffffffff8115afef>] writeback_single_inode+0x16d/0x20a ------------------ inode_wb_list_lock 2 [<ffffffff81165da5>] bdev_inode_switch_bdi+0x29/0x85 inode_wb_list_lock 19 [<ffffffff8115bd0b>] inode_wb_list_del+0x22/0x49 inode_wb_list_lock 5550 [<ffffffff8115bb53>] __mark_inode_dirty+0x170/0x1d0 inode_wb_list_lock 8511 [<ffffffff8115b4ad>] writeback_sb_inodes+0x10f/0x157 2.6.39-rc3 + patch: &(&wb->list_lock)->rlock: 11383 11657 0.14 151.69 40429.51 90825 527918 0.11 145.90 556843.37 ------------------------ &(&wb->list_lock)->rlock 10 [<ffffffff8115b189>] inode_wb_list_del+0x5f/0x86 &(&wb->list_lock)->rlock 1493 [<ffffffff8115b1ed>] writeback_inodes_wb+0x3d/0x150 &(&wb->list_lock)->rlock 3652 [<ffffffff8115a8e9>] writeback_sb_inodes+0x123/0x16f &(&wb->list_lock)->rlock 1412 [<ffffffff8115a38e>] writeback_single_inode+0x17f/0x223 ------------------------ &(&wb->list_lock)->rlock 3 [<ffffffff8110b5af>] bdi_lock_two+0x46/0x4b &(&wb->list_lock)->rlock 6 [<ffffffff8115b189>] inode_wb_list_del+0x5f/0x86 &(&wb->list_lock)->rlock 2061 [<ffffffff8115af97>] __mark_inode_dirty+0x173/0x1cf &(&wb->list_lock)->rlock 2629 [<ffffffff8115a8e9>] writeback_sb_inodes+0x123/0x16f hughd@google.com: fix recursive lock when bdi_lock_two() is called with new the same as old akpm@linux-foundation.org: cleanup bdev_inode_switch_bdi() comment Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2011-04-22 00:19:44 +00:00
spin_lock(&wb->list_lock);
list_for_each_entry(inode, &wb->b_dirty, i_io_list)
nr_dirty++;
list_for_each_entry(inode, &wb->b_io, i_io_list)
nr_io++;
list_for_each_entry(inode, &wb->b_more_io, i_io_list)
nr_more_io++;
list_for_each_entry(inode, &wb->b_dirty_time, i_io_list)
if (inode->i_state & I_DIRTY_TIME)
nr_dirty_time++;
writeback: split inode_wb_list_lock into bdi_writeback.list_lock Split the global inode_wb_list_lock into a per-bdi_writeback list_lock, as it's currently the most contended lock in the system for metadata heavy workloads. It won't help for single-filesystem workloads for which we'll need the I/O-less balance_dirty_pages, but at least we can dedicate a cpu to spinning on each bdi now for larger systems. Based on earlier patches from Nick Piggin and Dave Chinner. It reduces lock contentions to 1/4 in this test case: 10 HDD JBOD, 100 dd on each disk, XFS, 6GB ram lock_stat version 0.3 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- vanilla 2.6.39-rc3: inode_wb_list_lock: 42590 44433 0.12 147.74 144127.35 252274 886792 0.08 121.34 917211.23 ------------------ inode_wb_list_lock 2 [<ffffffff81165da5>] bdev_inode_switch_bdi+0x29/0x85 inode_wb_list_lock 34 [<ffffffff8115bd0b>] inode_wb_list_del+0x22/0x49 inode_wb_list_lock 12893 [<ffffffff8115bb53>] __mark_inode_dirty+0x170/0x1d0 inode_wb_list_lock 10702 [<ffffffff8115afef>] writeback_single_inode+0x16d/0x20a ------------------ inode_wb_list_lock 2 [<ffffffff81165da5>] bdev_inode_switch_bdi+0x29/0x85 inode_wb_list_lock 19 [<ffffffff8115bd0b>] inode_wb_list_del+0x22/0x49 inode_wb_list_lock 5550 [<ffffffff8115bb53>] __mark_inode_dirty+0x170/0x1d0 inode_wb_list_lock 8511 [<ffffffff8115b4ad>] writeback_sb_inodes+0x10f/0x157 2.6.39-rc3 + patch: &(&wb->list_lock)->rlock: 11383 11657 0.14 151.69 40429.51 90825 527918 0.11 145.90 556843.37 ------------------------ &(&wb->list_lock)->rlock 10 [<ffffffff8115b189>] inode_wb_list_del+0x5f/0x86 &(&wb->list_lock)->rlock 1493 [<ffffffff8115b1ed>] writeback_inodes_wb+0x3d/0x150 &(&wb->list_lock)->rlock 3652 [<ffffffff8115a8e9>] writeback_sb_inodes+0x123/0x16f &(&wb->list_lock)->rlock 1412 [<ffffffff8115a38e>] writeback_single_inode+0x17f/0x223 ------------------------ &(&wb->list_lock)->rlock 3 [<ffffffff8110b5af>] bdi_lock_two+0x46/0x4b &(&wb->list_lock)->rlock 6 [<ffffffff8115b189>] inode_wb_list_del+0x5f/0x86 &(&wb->list_lock)->rlock 2061 [<ffffffff8115af97>] __mark_inode_dirty+0x173/0x1cf &(&wb->list_lock)->rlock 2629 [<ffffffff8115a8e9>] writeback_sb_inodes+0x123/0x16f hughd@google.com: fix recursive lock when bdi_lock_two() is called with new the same as old akpm@linux-foundation.org: cleanup bdev_inode_switch_bdi() comment Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2011-04-22 00:19:44 +00:00
spin_unlock(&wb->list_lock);
global_dirty_limits(&background_thresh, &dirty_thresh);
wb_thresh = wb_calc_thresh(wb, dirty_thresh);
seq_printf(m,
"BdiWriteback: %10lu kB\n"
"BdiReclaimable: %10lu kB\n"
"BdiDirtyThresh: %10lu kB\n"
"DirtyThresh: %10lu kB\n"
"BackgroundThresh: %10lu kB\n"
"BdiDirtied: %10lu kB\n"
"BdiWritten: %10lu kB\n"
"BdiWriteBandwidth: %10lu kBps\n"
"b_dirty: %10lu\n"
"b_io: %10lu\n"
"b_more_io: %10lu\n"
"b_dirty_time: %10lu\n"
"bdi_list: %10u\n"
"state: %10lx\n",
(unsigned long) K(wb_stat(wb, WB_WRITEBACK)),
(unsigned long) K(wb_stat(wb, WB_RECLAIMABLE)),
K(wb_thresh),
K(dirty_thresh),
K(background_thresh),
(unsigned long) K(wb_stat(wb, WB_DIRTIED)),
(unsigned long) K(wb_stat(wb, WB_WRITTEN)),
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
(unsigned long) K(wb->write_bandwidth),
nr_dirty,
nr_io,
nr_more_io,
nr_dirty_time,
!list_empty(&bdi->bdi_list), bdi->wb.state);
return 0;
}
DEFINE_SHOW_ATTRIBUTE(bdi_debug_stats);
static void bdi_debug_register(struct backing_dev_info *bdi, const char *name)
{
bdi->debug_dir = debugfs_create_dir(name, bdi_debug_root);
debugfs_create_file("stats", 0444, bdi->debug_dir, bdi,
&bdi_debug_stats_fops);
}
static void bdi_debug_unregister(struct backing_dev_info *bdi)
{
debugfs_remove_recursive(bdi->debug_dir);
}
#else
static inline void bdi_debug_init(void)
{
}
static inline void bdi_debug_register(struct backing_dev_info *bdi,
const char *name)
{
}
static inline void bdi_debug_unregister(struct backing_dev_info *bdi)
{
}
#endif
static ssize_t read_ahead_kb_store(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct backing_dev_info *bdi = dev_get_drvdata(dev);
unsigned long read_ahead_kb;
ssize_t ret;
ret = kstrtoul(buf, 10, &read_ahead_kb);
if (ret < 0)
return ret;
bdi->ra_pages = read_ahead_kb >> (PAGE_SHIFT - 10);
return count;
}
#define BDI_SHOW(name, expr) \
static ssize_t name##_show(struct device *dev, \
struct device_attribute *attr, char *buf) \
{ \
struct backing_dev_info *bdi = dev_get_drvdata(dev); \
\
return sysfs_emit(buf, "%lld\n", (long long)expr); \
} \
static DEVICE_ATTR_RW(name);
BDI_SHOW(read_ahead_kb, K(bdi->ra_pages))
static ssize_t min_ratio_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t count)
{
struct backing_dev_info *bdi = dev_get_drvdata(dev);
unsigned int ratio;
ssize_t ret;
ret = kstrtouint(buf, 10, &ratio);
if (ret < 0)
return ret;
ret = bdi_set_min_ratio(bdi, ratio);
if (!ret)
ret = count;
return ret;
}
BDI_SHOW(min_ratio, bdi->min_ratio)
static ssize_t max_ratio_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t count)
{
struct backing_dev_info *bdi = dev_get_drvdata(dev);
unsigned int ratio;
ssize_t ret;
ret = kstrtouint(buf, 10, &ratio);
if (ret < 0)
return ret;
ret = bdi_set_max_ratio(bdi, ratio);
if (!ret)
ret = count;
return ret;
}
BDI_SHOW(max_ratio, bdi->max_ratio)
bdi: allow block devices to say that they require stable page writes This patchset ("stable page writes, part 2") makes some key modifications to the original 'stable page writes' patchset. First, it provides creators (devices and filesystems) of a backing_dev_info a flag that declares whether or not it is necessary to ensure that page contents cannot change during writeout. It is no longer assumed that this is true of all devices (which was never true anyway). Second, the flag is used to relaxed the wait_on_page_writeback calls so that wait only occurs if the device needs it. Third, it fixes up the remaining disk-backed filesystems to use this improved conditional-wait logic to provide stable page writes on those filesystems. It is hoped that (for people not using checksumming devices, anyway) this patchset will give back unnecessary performance decreases since the original stable page write patchset went into 3.0. Sorry about not fixing it sooner. Complaints were registered by several people about the long write latencies introduced by the original stable page write patchset. Generally speaking, the kernel ought to allocate as little extra memory as possible to facilitate writeout, but for people who simply cannot wait, a second page stability strategy is (re)introduced: snapshotting page contents. The waiting behavior is still the default strategy; to enable page snapshotting, a superblock flag (MS_SNAP_STABLE) must be set. This flag is used to bandaid^Henable stable page writeback on ext3[1], and is not used anywhere else. Given that there are already a few storage devices and network FSes that have rolled their own page stability wait/page snapshot code, it would be nice to move towards consolidating all of these. It seems possible that iscsi and raid5 may wish to use the new stable page write support to enable zero-copy writeout. Thank you to Jan Kara for helping fix a couple more filesystems. Per Andrew Morton's request, here are the result of using dbench to measure latencies on ext2: 3.8.0-rc3: Operation Count AvgLat MaxLat ---------------------------------------- WriteX 109347 0.028 59.817 ReadX 347180 0.004 3.391 Flush 15514 29.828 287.283 Throughput 57.429 MB/sec 4 clients 4 procs max_latency=287.290 ms 3.8.0-rc3 + patches: WriteX 105556 0.029 4.273 ReadX 335004 0.005 4.112 Flush 14982 30.540 298.634 Throughput 55.4496 MB/sec 4 clients 4 procs max_latency=298.650 ms As you can see, for ext2 the maximum write latency decreases from ~60ms on a laptop hard disk to ~4ms. I'm not sure why the flush latencies increase, though I suspect that being able to dirty pages faster gives the flusher more work to do. On ext4, the average write latency decreases as well as all the maximum latencies: 3.8.0-rc3: WriteX 85624 0.152 33.078 ReadX 272090 0.010 61.210 Flush 12129 36.219 168.260 Throughput 44.8618 MB/sec 4 clients 4 procs max_latency=168.276 ms 3.8.0-rc3 + patches: WriteX 86082 0.141 30.928 ReadX 273358 0.010 36.124 Flush 12214 34.800 165.689 Throughput 44.9941 MB/sec 4 clients 4 procs max_latency=165.722 ms XFS seems to exhibit similar latency improvements as ext2: 3.8.0-rc3: WriteX 125739 0.028 104.343 ReadX 399070 0.005 4.115 Flush 17851 25.004 131.390 Throughput 66.0024 MB/sec 4 clients 4 procs max_latency=131.406 ms 3.8.0-rc3 + patches: WriteX 123529 0.028 6.299 ReadX 392434 0.005 4.287 Flush 17549 25.120 188.687 Throughput 64.9113 MB/sec 4 clients 4 procs max_latency=188.704 ms ...and btrfs, just to round things out, also shows some latency decreases: 3.8.0-rc3: WriteX 67122 0.083 82.355 ReadX 212719 0.005 2.828 Flush 9547 47.561 147.418 Throughput 35.3391 MB/sec 4 clients 4 procs max_latency=147.433 ms 3.8.0-rc3 + patches: WriteX 64898 0.101 71.631 ReadX 206673 0.005 7.123 Flush 9190 47.963 219.034 Throughput 34.0795 MB/sec 4 clients 4 procs max_latency=219.044 ms Before this patchset, all filesystems would block, regardless of whether or not it was necessary. ext3 would wait, but still generate occasional checksum errors. The network filesystems were left to do their own thing, so they'd wait too. After this patchset, all the disk filesystems except ext3 and btrfs will wait only if the hardware requires it. ext3 (if necessary) snapshots pages instead of blocking, and btrfs provides its own bdi so the mm will never wait. Network filesystems haven't been touched, so either they provide their own wait code, or they don't block at all. The blocking behavior is back to what it was before 3.0 if you don't have a disk requiring stable page writes. This patchset has been tested on 3.8.0-rc3 on x64 with ext3, ext4, and xfs. I've spot-checked 3.8.0-rc4 and seem to be getting the same results as -rc3. [1] The alternative fixes to ext3 include fixing the locking order and page bit handling like we did for ext4 (but then why not just use ext4?), or setting PG_writeback so early that ext3 becomes extremely slow. I tried that, but the number of write()s I could initiate dropped by nearly an order of magnitude. That was a bit much even for the author of the stable page series! :) This patch: Creates a per-backing-device flag that tracks whether or not pages must be held immutable during writeout. Eventually it will be used to waive wait_for_page_writeback() if nothing requires stable pages. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Artem Bityutskiy <dedekind1@gmail.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-22 00:42:48 +00:00
static ssize_t stable_pages_required_show(struct device *dev,
struct device_attribute *attr,
char *buf)
bdi: allow block devices to say that they require stable page writes This patchset ("stable page writes, part 2") makes some key modifications to the original 'stable page writes' patchset. First, it provides creators (devices and filesystems) of a backing_dev_info a flag that declares whether or not it is necessary to ensure that page contents cannot change during writeout. It is no longer assumed that this is true of all devices (which was never true anyway). Second, the flag is used to relaxed the wait_on_page_writeback calls so that wait only occurs if the device needs it. Third, it fixes up the remaining disk-backed filesystems to use this improved conditional-wait logic to provide stable page writes on those filesystems. It is hoped that (for people not using checksumming devices, anyway) this patchset will give back unnecessary performance decreases since the original stable page write patchset went into 3.0. Sorry about not fixing it sooner. Complaints were registered by several people about the long write latencies introduced by the original stable page write patchset. Generally speaking, the kernel ought to allocate as little extra memory as possible to facilitate writeout, but for people who simply cannot wait, a second page stability strategy is (re)introduced: snapshotting page contents. The waiting behavior is still the default strategy; to enable page snapshotting, a superblock flag (MS_SNAP_STABLE) must be set. This flag is used to bandaid^Henable stable page writeback on ext3[1], and is not used anywhere else. Given that there are already a few storage devices and network FSes that have rolled their own page stability wait/page snapshot code, it would be nice to move towards consolidating all of these. It seems possible that iscsi and raid5 may wish to use the new stable page write support to enable zero-copy writeout. Thank you to Jan Kara for helping fix a couple more filesystems. Per Andrew Morton's request, here are the result of using dbench to measure latencies on ext2: 3.8.0-rc3: Operation Count AvgLat MaxLat ---------------------------------------- WriteX 109347 0.028 59.817 ReadX 347180 0.004 3.391 Flush 15514 29.828 287.283 Throughput 57.429 MB/sec 4 clients 4 procs max_latency=287.290 ms 3.8.0-rc3 + patches: WriteX 105556 0.029 4.273 ReadX 335004 0.005 4.112 Flush 14982 30.540 298.634 Throughput 55.4496 MB/sec 4 clients 4 procs max_latency=298.650 ms As you can see, for ext2 the maximum write latency decreases from ~60ms on a laptop hard disk to ~4ms. I'm not sure why the flush latencies increase, though I suspect that being able to dirty pages faster gives the flusher more work to do. On ext4, the average write latency decreases as well as all the maximum latencies: 3.8.0-rc3: WriteX 85624 0.152 33.078 ReadX 272090 0.010 61.210 Flush 12129 36.219 168.260 Throughput 44.8618 MB/sec 4 clients 4 procs max_latency=168.276 ms 3.8.0-rc3 + patches: WriteX 86082 0.141 30.928 ReadX 273358 0.010 36.124 Flush 12214 34.800 165.689 Throughput 44.9941 MB/sec 4 clients 4 procs max_latency=165.722 ms XFS seems to exhibit similar latency improvements as ext2: 3.8.0-rc3: WriteX 125739 0.028 104.343 ReadX 399070 0.005 4.115 Flush 17851 25.004 131.390 Throughput 66.0024 MB/sec 4 clients 4 procs max_latency=131.406 ms 3.8.0-rc3 + patches: WriteX 123529 0.028 6.299 ReadX 392434 0.005 4.287 Flush 17549 25.120 188.687 Throughput 64.9113 MB/sec 4 clients 4 procs max_latency=188.704 ms ...and btrfs, just to round things out, also shows some latency decreases: 3.8.0-rc3: WriteX 67122 0.083 82.355 ReadX 212719 0.005 2.828 Flush 9547 47.561 147.418 Throughput 35.3391 MB/sec 4 clients 4 procs max_latency=147.433 ms 3.8.0-rc3 + patches: WriteX 64898 0.101 71.631 ReadX 206673 0.005 7.123 Flush 9190 47.963 219.034 Throughput 34.0795 MB/sec 4 clients 4 procs max_latency=219.044 ms Before this patchset, all filesystems would block, regardless of whether or not it was necessary. ext3 would wait, but still generate occasional checksum errors. The network filesystems were left to do their own thing, so they'd wait too. After this patchset, all the disk filesystems except ext3 and btrfs will wait only if the hardware requires it. ext3 (if necessary) snapshots pages instead of blocking, and btrfs provides its own bdi so the mm will never wait. Network filesystems haven't been touched, so either they provide their own wait code, or they don't block at all. The blocking behavior is back to what it was before 3.0 if you don't have a disk requiring stable page writes. This patchset has been tested on 3.8.0-rc3 on x64 with ext3, ext4, and xfs. I've spot-checked 3.8.0-rc4 and seem to be getting the same results as -rc3. [1] The alternative fixes to ext3 include fixing the locking order and page bit handling like we did for ext4 (but then why not just use ext4?), or setting PG_writeback so early that ext3 becomes extremely slow. I tried that, but the number of write()s I could initiate dropped by nearly an order of magnitude. That was a bit much even for the author of the stable page series! :) This patch: Creates a per-backing-device flag that tracks whether or not pages must be held immutable during writeout. Eventually it will be used to waive wait_for_page_writeback() if nothing requires stable pages. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Artem Bityutskiy <dedekind1@gmail.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-22 00:42:48 +00:00
{
dev_warn_once(dev,
"the stable_pages_required attribute has been removed. Use the stable_writes queue attribute instead.\n");
return sysfs_emit(buf, "%d\n", 0);
bdi: allow block devices to say that they require stable page writes This patchset ("stable page writes, part 2") makes some key modifications to the original 'stable page writes' patchset. First, it provides creators (devices and filesystems) of a backing_dev_info a flag that declares whether or not it is necessary to ensure that page contents cannot change during writeout. It is no longer assumed that this is true of all devices (which was never true anyway). Second, the flag is used to relaxed the wait_on_page_writeback calls so that wait only occurs if the device needs it. Third, it fixes up the remaining disk-backed filesystems to use this improved conditional-wait logic to provide stable page writes on those filesystems. It is hoped that (for people not using checksumming devices, anyway) this patchset will give back unnecessary performance decreases since the original stable page write patchset went into 3.0. Sorry about not fixing it sooner. Complaints were registered by several people about the long write latencies introduced by the original stable page write patchset. Generally speaking, the kernel ought to allocate as little extra memory as possible to facilitate writeout, but for people who simply cannot wait, a second page stability strategy is (re)introduced: snapshotting page contents. The waiting behavior is still the default strategy; to enable page snapshotting, a superblock flag (MS_SNAP_STABLE) must be set. This flag is used to bandaid^Henable stable page writeback on ext3[1], and is not used anywhere else. Given that there are already a few storage devices and network FSes that have rolled their own page stability wait/page snapshot code, it would be nice to move towards consolidating all of these. It seems possible that iscsi and raid5 may wish to use the new stable page write support to enable zero-copy writeout. Thank you to Jan Kara for helping fix a couple more filesystems. Per Andrew Morton's request, here are the result of using dbench to measure latencies on ext2: 3.8.0-rc3: Operation Count AvgLat MaxLat ---------------------------------------- WriteX 109347 0.028 59.817 ReadX 347180 0.004 3.391 Flush 15514 29.828 287.283 Throughput 57.429 MB/sec 4 clients 4 procs max_latency=287.290 ms 3.8.0-rc3 + patches: WriteX 105556 0.029 4.273 ReadX 335004 0.005 4.112 Flush 14982 30.540 298.634 Throughput 55.4496 MB/sec 4 clients 4 procs max_latency=298.650 ms As you can see, for ext2 the maximum write latency decreases from ~60ms on a laptop hard disk to ~4ms. I'm not sure why the flush latencies increase, though I suspect that being able to dirty pages faster gives the flusher more work to do. On ext4, the average write latency decreases as well as all the maximum latencies: 3.8.0-rc3: WriteX 85624 0.152 33.078 ReadX 272090 0.010 61.210 Flush 12129 36.219 168.260 Throughput 44.8618 MB/sec 4 clients 4 procs max_latency=168.276 ms 3.8.0-rc3 + patches: WriteX 86082 0.141 30.928 ReadX 273358 0.010 36.124 Flush 12214 34.800 165.689 Throughput 44.9941 MB/sec 4 clients 4 procs max_latency=165.722 ms XFS seems to exhibit similar latency improvements as ext2: 3.8.0-rc3: WriteX 125739 0.028 104.343 ReadX 399070 0.005 4.115 Flush 17851 25.004 131.390 Throughput 66.0024 MB/sec 4 clients 4 procs max_latency=131.406 ms 3.8.0-rc3 + patches: WriteX 123529 0.028 6.299 ReadX 392434 0.005 4.287 Flush 17549 25.120 188.687 Throughput 64.9113 MB/sec 4 clients 4 procs max_latency=188.704 ms ...and btrfs, just to round things out, also shows some latency decreases: 3.8.0-rc3: WriteX 67122 0.083 82.355 ReadX 212719 0.005 2.828 Flush 9547 47.561 147.418 Throughput 35.3391 MB/sec 4 clients 4 procs max_latency=147.433 ms 3.8.0-rc3 + patches: WriteX 64898 0.101 71.631 ReadX 206673 0.005 7.123 Flush 9190 47.963 219.034 Throughput 34.0795 MB/sec 4 clients 4 procs max_latency=219.044 ms Before this patchset, all filesystems would block, regardless of whether or not it was necessary. ext3 would wait, but still generate occasional checksum errors. The network filesystems were left to do their own thing, so they'd wait too. After this patchset, all the disk filesystems except ext3 and btrfs will wait only if the hardware requires it. ext3 (if necessary) snapshots pages instead of blocking, and btrfs provides its own bdi so the mm will never wait. Network filesystems haven't been touched, so either they provide their own wait code, or they don't block at all. The blocking behavior is back to what it was before 3.0 if you don't have a disk requiring stable page writes. This patchset has been tested on 3.8.0-rc3 on x64 with ext3, ext4, and xfs. I've spot-checked 3.8.0-rc4 and seem to be getting the same results as -rc3. [1] The alternative fixes to ext3 include fixing the locking order and page bit handling like we did for ext4 (but then why not just use ext4?), or setting PG_writeback so early that ext3 becomes extremely slow. I tried that, but the number of write()s I could initiate dropped by nearly an order of magnitude. That was a bit much even for the author of the stable page series! :) This patch: Creates a per-backing-device flag that tracks whether or not pages must be held immutable during writeout. Eventually it will be used to waive wait_for_page_writeback() if nothing requires stable pages. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Artem Bityutskiy <dedekind1@gmail.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-22 00:42:48 +00:00
}
static DEVICE_ATTR_RO(stable_pages_required);
static struct attribute *bdi_dev_attrs[] = {
&dev_attr_read_ahead_kb.attr,
&dev_attr_min_ratio.attr,
&dev_attr_max_ratio.attr,
&dev_attr_stable_pages_required.attr,
NULL,
};
ATTRIBUTE_GROUPS(bdi_dev);
static __init int bdi_class_init(void)
{
bdi_class = class_create(THIS_MODULE, "bdi");
if (IS_ERR(bdi_class))
return PTR_ERR(bdi_class);
bdi_class->dev_groups = bdi_dev_groups;
bdi_debug_init();
return 0;
}
postcore_initcall(bdi_class_init);
static int bdi_init(struct backing_dev_info *bdi);
static int __init default_bdi_init(void)
{
int err;
bdi: Do not use freezable workqueue A removable block device, such as NVMe or SSD connected over Thunderbolt can be hot-removed any time including when the system is suspended. When device is hot-removed during suspend and the system gets resumed, kernel first resumes devices and then thaws the userspace including freezable workqueues. What happens in that case is that the NVMe driver notices that the device is unplugged and removes it from the system. This ends up calling bdi_unregister() for the gendisk which then schedules wb_workfn() to be run one more time. However, since the bdi_wq is still frozen flush_delayed_work() call in wb_shutdown() blocks forever halting system resume process. User sees this as hang as nothing is happening anymore. Triggering sysrq-w reveals this: Workqueue: nvme-wq nvme_remove_dead_ctrl_work [nvme] Call Trace: ? __schedule+0x2c5/0x630 ? wait_for_completion+0xa4/0x120 schedule+0x3e/0xc0 schedule_timeout+0x1c9/0x320 ? resched_curr+0x1f/0xd0 ? wait_for_completion+0xa4/0x120 wait_for_completion+0xc3/0x120 ? wake_up_q+0x60/0x60 __flush_work+0x131/0x1e0 ? flush_workqueue_prep_pwqs+0x130/0x130 bdi_unregister+0xb9/0x130 del_gendisk+0x2d2/0x2e0 nvme_ns_remove+0xed/0x110 [nvme_core] nvme_remove_namespaces+0x96/0xd0 [nvme_core] nvme_remove+0x5b/0x160 [nvme] pci_device_remove+0x36/0x90 device_release_driver_internal+0xdf/0x1c0 nvme_remove_dead_ctrl_work+0x14/0x30 [nvme] process_one_work+0x1c2/0x3f0 worker_thread+0x48/0x3e0 kthread+0x100/0x140 ? current_work+0x30/0x30 ? kthread_park+0x80/0x80 ret_from_fork+0x35/0x40 This is not limited to NVMes so exactly same issue can be reproduced by hot-removing SSD (over Thunderbolt) while the system is suspended. Prevent this from happening by removing WQ_FREEZABLE from bdi_wq. Reported-by: AceLan Kao <acelan.kao@canonical.com> Link: https://marc.info/?l=linux-kernel&m=138695698516487 Link: https://bugzilla.kernel.org/show_bug.cgi?id=204385 Link: https://lore.kernel.org/lkml/20191002122136.GD2819@lahna.fi.intel.com/#t Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-10-04 10:00:24 +00:00
bdi_wq = alloc_workqueue("writeback", WQ_MEM_RECLAIM | WQ_UNBOUND |
WQ_SYSFS, 0);
writeback: replace custom worker pool implementation with unbound workqueue Writeback implements its own worker pool - each bdi can be associated with a worker thread which is created and destroyed dynamically. The worker thread for the default bdi is always present and serves as the "forker" thread which forks off worker threads for other bdis. there's no reason for writeback to implement its own worker pool when using unbound workqueue instead is much simpler and more efficient. This patch replaces custom worker pool implementation in writeback with an unbound workqueue. The conversion isn't too complicated but the followings are worth mentioning. * bdi_writeback->last_active, task and wakeup_timer are removed. delayed_work ->dwork is added instead. Explicit timer handling is no longer necessary. Everything works by either queueing / modding / flushing / canceling the delayed_work item. * bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off bdi_writeback->dwork. On each execution, it processes bdi->work_list and reschedules itself if there are more things to do. The function also handles low-mem condition, which used to be handled by the forker thread. If the function is running off a rescuer thread, it only writes out limited number of pages so that the rescuer can serve other bdis too. This preserves the flusher creation failure behavior of the forker thread. * INIT_LIST_HEAD(&bdi->bdi_list) is used to tell bdi_writeback_workfn() about on-going bdi unregistration so that it always drains work_list even if it's running off the rescuer. Note that the original code was broken in this regard. Under memory pressure, a bdi could finish unregistration with non-empty work_list. * The default bdi is no longer special. It now is treated the same as any other bdi and bdi_cap_flush_forker() is removed. * BDI_pending is no longer used. Removed. * Some tracepoints become non-applicable. The following TPs are removed - writeback_nothread, writeback_wake_thread, writeback_wake_forker_thread, writeback_thread_start, writeback_thread_stop. Everything, including devices coming and going away and rescuer operation under simulated memory pressure, seems to work fine in my test setup. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Fengguang Wu <fengguang.wu@intel.com> Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-02 02:08:06 +00:00
if (!bdi_wq)
return -ENOMEM;
err = bdi_init(&noop_backing_dev_info);
return err;
}
subsys_initcall(default_bdi_init);
/*
writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:30 +00:00
* This function is used when the first inode for this wb is marked dirty. It
* wakes-up the corresponding bdi thread which should then take care of the
* periodic background write-out of dirty inodes. Since the write-out would
* starts only 'dirty_writeback_interval' centisecs from now anyway, we just
* set up a timer which wakes the bdi thread up later.
*
* Note, we wouldn't bother setting up the timer, but this function is on the
* fast-path (used by '__mark_inode_dirty()'), so we save few context switches
* by delaying the wake-up.
backing_dev: fix hung task on sync bdi_wakeup_thread_delayed() used the mod_delayed_work() function to schedule work to writeback dirty inodes. The problem with this is that it can delay work that is scheduled for immediate execution, such as the work from sync_inodes_sb(). This can happen since mod_delayed_work() can now steal work from a work_queue. This fixes the problem by using queue_delayed_work() instead. This is a regression caused by commit 839a8e8660b6 ("writeback: replace custom worker pool implementation with unbound workqueue"). The reason that this causes a problem is that laptop-mode will change the delay, dirty_writeback_centisecs, to 60000 (10 minutes) by default. In the case that bdi_wakeup_thread_delayed() races with sync_inodes_sb(), sync will be stopped for 10 minutes and trigger a hung task. Even if dirty_writeback_centisecs is not long enough to cause a hung task, we still don't want to delay sync for that long. We fix the problem by using queue_delayed_work() when we want to schedule writeback sometime in future. This function doesn't change the timer if it is already armed. For the same reason, we also change bdi_writeback_workfn() to immediately queue the work again in the case that the work_list is not empty. The same problem can happen if the sync work is run on the rescue worker. [jack@suse.cz: update changelog, add comment, use bdi_wakeup_thread_delayed()] Signed-off-by: Derek Basehore <dbasehore@chromium.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zento.linux.org.uk> Reviewed-by: Tejun Heo <tj@kernel.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Darrick J. Wong" <darrick.wong@oracle.com> Cc: Derek Basehore <dbasehore@chromium.org> Cc: Kees Cook <keescook@chromium.org> Cc: Benson Leung <bleung@chromium.org> Cc: Sonny Rao <sonnyrao@chromium.org> Cc: Luigi Semenzato <semenzato@chromium.org> Cc: Jens Axboe <axboe@kernel.dk> Cc: Dave Chinner <david@fromorbit.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:46:22 +00:00
*
* We have to be careful not to postpone flush work if it is scheduled for
* earlier. Thus we use queue_delayed_work().
*/
writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:30 +00:00
void wb_wakeup_delayed(struct bdi_writeback *wb)
{
unsigned long timeout;
timeout = msecs_to_jiffies(dirty_writeback_interval * 10);
writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:30 +00:00
spin_lock_bh(&wb->work_lock);
if (test_bit(WB_registered, &wb->state))
queue_delayed_work(bdi_wq, &wb->dwork, timeout);
spin_unlock_bh(&wb->work_lock);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
}
/*
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
* Initial write bandwidth: 100 MB/s
*/
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
#define INIT_BW (100 << (20 - PAGE_SHIFT))
static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi,
gfp_t gfp)
{
int i, err;
memset(wb, 0, sizeof(*wb));
if (wb != &bdi->wb)
bdi_get(bdi);
wb->bdi = bdi;
wb->last_old_flush = jiffies;
INIT_LIST_HEAD(&wb->b_dirty);
INIT_LIST_HEAD(&wb->b_io);
INIT_LIST_HEAD(&wb->b_more_io);
INIT_LIST_HEAD(&wb->b_dirty_time);
writeback: split inode_wb_list_lock into bdi_writeback.list_lock Split the global inode_wb_list_lock into a per-bdi_writeback list_lock, as it's currently the most contended lock in the system for metadata heavy workloads. It won't help for single-filesystem workloads for which we'll need the I/O-less balance_dirty_pages, but at least we can dedicate a cpu to spinning on each bdi now for larger systems. Based on earlier patches from Nick Piggin and Dave Chinner. It reduces lock contentions to 1/4 in this test case: 10 HDD JBOD, 100 dd on each disk, XFS, 6GB ram lock_stat version 0.3 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- vanilla 2.6.39-rc3: inode_wb_list_lock: 42590 44433 0.12 147.74 144127.35 252274 886792 0.08 121.34 917211.23 ------------------ inode_wb_list_lock 2 [<ffffffff81165da5>] bdev_inode_switch_bdi+0x29/0x85 inode_wb_list_lock 34 [<ffffffff8115bd0b>] inode_wb_list_del+0x22/0x49 inode_wb_list_lock 12893 [<ffffffff8115bb53>] __mark_inode_dirty+0x170/0x1d0 inode_wb_list_lock 10702 [<ffffffff8115afef>] writeback_single_inode+0x16d/0x20a ------------------ inode_wb_list_lock 2 [<ffffffff81165da5>] bdev_inode_switch_bdi+0x29/0x85 inode_wb_list_lock 19 [<ffffffff8115bd0b>] inode_wb_list_del+0x22/0x49 inode_wb_list_lock 5550 [<ffffffff8115bb53>] __mark_inode_dirty+0x170/0x1d0 inode_wb_list_lock 8511 [<ffffffff8115b4ad>] writeback_sb_inodes+0x10f/0x157 2.6.39-rc3 + patch: &(&wb->list_lock)->rlock: 11383 11657 0.14 151.69 40429.51 90825 527918 0.11 145.90 556843.37 ------------------------ &(&wb->list_lock)->rlock 10 [<ffffffff8115b189>] inode_wb_list_del+0x5f/0x86 &(&wb->list_lock)->rlock 1493 [<ffffffff8115b1ed>] writeback_inodes_wb+0x3d/0x150 &(&wb->list_lock)->rlock 3652 [<ffffffff8115a8e9>] writeback_sb_inodes+0x123/0x16f &(&wb->list_lock)->rlock 1412 [<ffffffff8115a38e>] writeback_single_inode+0x17f/0x223 ------------------------ &(&wb->list_lock)->rlock 3 [<ffffffff8110b5af>] bdi_lock_two+0x46/0x4b &(&wb->list_lock)->rlock 6 [<ffffffff8115b189>] inode_wb_list_del+0x5f/0x86 &(&wb->list_lock)->rlock 2061 [<ffffffff8115af97>] __mark_inode_dirty+0x173/0x1cf &(&wb->list_lock)->rlock 2629 [<ffffffff8115a8e9>] writeback_sb_inodes+0x123/0x16f hughd@google.com: fix recursive lock when bdi_lock_two() is called with new the same as old akpm@linux-foundation.org: cleanup bdev_inode_switch_bdi() comment Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2011-04-22 00:19:44 +00:00
spin_lock_init(&wb->list_lock);
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
wb->bw_time_stamp = jiffies;
wb->balanced_dirty_ratelimit = INIT_BW;
wb->dirty_ratelimit = INIT_BW;
wb->write_bandwidth = INIT_BW;
wb->avg_write_bandwidth = INIT_BW;
writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:30 +00:00
spin_lock_init(&wb->work_lock);
INIT_LIST_HEAD(&wb->work_list);
INIT_DELAYED_WORK(&wb->dwork, wb_workfn);
wb->dirty_sleep = jiffies;
err = fprop_local_init_percpu(&wb->completions, gfp);
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
if (err)
goto out_put_bdi;
for (i = 0; i < NR_WB_STAT_ITEMS; i++) {
err = percpu_counter_init(&wb->stat[i], 0, gfp);
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
if (err)
goto out_destroy_stat;
}
return 0;
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
out_destroy_stat:
while (i--)
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
percpu_counter_destroy(&wb->stat[i]);
fprop_local_destroy_percpu(&wb->completions);
out_put_bdi:
if (wb != &bdi->wb)
bdi_put(bdi);
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
return err;
}
static void cgwb_remove_from_bdi_list(struct bdi_writeback *wb);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
/*
* Remove bdi from the global list and shutdown any threads we have running
*/
static void wb_shutdown(struct bdi_writeback *wb)
{
/* Make sure nobody queues further work */
spin_lock_bh(&wb->work_lock);
if (!test_and_clear_bit(WB_registered, &wb->state)) {
spin_unlock_bh(&wb->work_lock);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
return;
}
spin_unlock_bh(&wb->work_lock);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
cgwb_remove_from_bdi_list(wb);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
/*
* Drain work list and shutdown the delayed_work. !WB_registered
* tells wb_workfn() that @wb is dying and its work_list needs to
* be drained no matter what.
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
*/
mod_delayed_work(bdi_wq, &wb->dwork, 0);
flush_delayed_work(&wb->dwork);
WARN_ON(!list_empty(&wb->work_list));
}
writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:30 +00:00
static void wb_exit(struct bdi_writeback *wb)
{
int i;
WARN_ON(delayed_work_pending(&wb->dwork));
for (i = 0; i < NR_WB_STAT_ITEMS; i++)
percpu_counter_destroy(&wb->stat[i]);
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
fprop_local_destroy_percpu(&wb->completions);
if (wb != &wb->bdi->wb)
bdi_put(wb->bdi);
writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:28 +00:00
}
writeback: bdi write bandwidth estimation The estimation value will start from 100MB/s and adapt to the real bandwidth in seconds. It tries to update the bandwidth only when disk is fully utilized. Any inactive period of more than one second will be skipped. The estimated bandwidth will be reflecting how fast the device can writeout when _fully utilized_, and won't drop to 0 when it goes idle. The value will remain constant at disk idle time. At busy write time, if not considering fluctuations, it will also remain high unless be knocked down by possible concurrent reads that compete for the disk time and bandwidth with async writes. The estimation is not done purely in the flusher because there is no guarantee for write_cache_pages() to return timely to update bandwidth. The bdi->avg_write_bandwidth smoothing is very effective for filtering out sudden spikes, however may be a little biased in long term. The overheads are low because the bdi bandwidth update only occurs at 200ms intervals. The 200ms update interval is suitable, because it's not possible to get the real bandwidth for the instance at all, due to large fluctuations. The NFS commits can be as large as seconds worth of data. One XFS completion may be as large as half second worth of data if we are going to increase the write chunk to half second worth of data. In ext4, fluctuations with time period of around 5 seconds is observed. And there is another pattern of irregular periods of up to 20 seconds on SSD tests. That's why we are not only doing the estimation at 200ms intervals, but also averaging them over a period of 3 seconds and then go further to do another level of smoothing in avg_write_bandwidth. CC: Li Shaohua <shaohua.li@intel.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-08-29 17:22:30 +00:00
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
#ifdef CONFIG_CGROUP_WRITEBACK
#include <linux/memcontrol.h>
/*
writeback, cgroup: release dying cgwbs by switching attached inodes Asynchronously try to release dying cgwbs by switching attached inodes to the nearest living ancestor wb. It helps to get rid of per-cgroup writeback structures themselves and of pinned memory and block cgroups, which are significantly larger structures (mostly due to large per-cpu statistics data). This prevents memory waste and helps to avoid different scalability problems caused by large piles of dying cgroups. Reuse the existing mechanism of inode switching used for foreign inode detection. To speed things up batch up to 115 inode switching in a single operation (the maximum number is selected so that the resulting struct inode_switch_wbs_context can fit into 1024 bytes). Because every switching consists of two steps divided by an RCU grace period, it would be too slow without batching. Please note that the whole batch counts as a single operation (when increasing/decreasing isw_nr_in_flight). This allows to keep umounting working (flush the switching queue), however prevents cleanups from consuming the whole switching quota and effectively blocking the frn switching. A cgwb cleanup operation can fail due to different reasons (e.g. not enough memory, the cgwb has an in-flight/pending io, an attached inode in a wrong state, etc). In this case the next scheduled cleanup will make a new attempt. An attempt is made each time a new cgwb is offlined (in other words a memcg and/or a blkcg is deleted by a user). In the future an additional attempt scheduled by a timer can be implemented. [guro@fb.com: replace open-coded "115" with arithmetic] Link: https://lkml.kernel.org/r/YMEcSBcq/VXMiPPO@carbon.dhcp.thefacebook.com [guro@fb.com: add smp_mb() to inode_prepare_wbs_switch()] Link: https://lkml.kernel.org/r/YMFa+guFw7OFjf3X@carbon.dhcp.thefacebook.com [willy@infradead.org: fix documentation] Link: https://lkml.kernel.org/r/20210615200242.1716568-2-willy@infradead.org Link: https://lkml.kernel.org/r/20210608230225.2078447-9-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:36:03 +00:00
* cgwb_lock protects bdi->cgwb_tree, blkcg->cgwb_list, offline_cgwbs and
* memcg->cgwb_list. bdi->cgwb_tree is also RCU protected.
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
*/
static DEFINE_SPINLOCK(cgwb_lock);
bdi: Move cgroup bdi_writeback to a dedicated low concurrency workqueue From 0aa2e9b921d6db71150633ff290199554f0842a8 Mon Sep 17 00:00:00 2001 From: Tejun Heo <tj@kernel.org> Date: Wed, 23 May 2018 10:29:00 -0700 cgwb_release() punts the actual release to cgwb_release_workfn() on system_wq. Depending on the number of cgroups or block devices, there can be a lot of cgwb_release_workfn() in flight at the same time. We're periodically seeing close to 256 kworkers getting stuck with the following stack trace and overtime the entire system gets stuck. [<ffffffff810ee40c>] _synchronize_rcu_expedited.constprop.72+0x2fc/0x330 [<ffffffff810ee634>] synchronize_rcu_expedited+0x24/0x30 [<ffffffff811ccf23>] bdi_unregister+0x53/0x290 [<ffffffff811cd1e9>] release_bdi+0x89/0xc0 [<ffffffff811cd645>] wb_exit+0x85/0xa0 [<ffffffff811cdc84>] cgwb_release_workfn+0x54/0xb0 [<ffffffff810a68d0>] process_one_work+0x150/0x410 [<ffffffff810a71fd>] worker_thread+0x6d/0x520 [<ffffffff810ad3dc>] kthread+0x12c/0x160 [<ffffffff81969019>] ret_from_fork+0x29/0x40 [<ffffffffffffffff>] 0xffffffffffffffff The events leading to the lockup are... 1. A lot of cgwb_release_workfn() is queued at the same time and all system_wq kworkers are assigned to execute them. 2. They all end up calling synchronize_rcu_expedited(). One of them wins and tries to perform the expedited synchronization. 3. However, that invovles queueing rcu_exp_work to system_wq and waiting for it. Because #1 is holding all available kworkers on system_wq, rcu_exp_work can't be executed. cgwb_release_workfn() is waiting for synchronize_rcu_expedited() which in turn is waiting for cgwb_release_workfn() to free up some of the kworkers. We shouldn't be scheduling hundreds of cgwb_release_workfn() at the same time. There's nothing to be gained from that. This patch updates cgwb release path to use a dedicated percpu workqueue with @max_active of 1. While this resolves the problem at hand, it might be a good idea to isolate rcu_exp_work to its own workqueue too as it can be used from various paths and is prone to this sort of indirect A-A deadlocks. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Cc: stable@vger.kernel.org Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-05-23 17:56:32 +00:00
static struct workqueue_struct *cgwb_release_wq;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
writeback, cgroup: release dying cgwbs by switching attached inodes Asynchronously try to release dying cgwbs by switching attached inodes to the nearest living ancestor wb. It helps to get rid of per-cgroup writeback structures themselves and of pinned memory and block cgroups, which are significantly larger structures (mostly due to large per-cpu statistics data). This prevents memory waste and helps to avoid different scalability problems caused by large piles of dying cgroups. Reuse the existing mechanism of inode switching used for foreign inode detection. To speed things up batch up to 115 inode switching in a single operation (the maximum number is selected so that the resulting struct inode_switch_wbs_context can fit into 1024 bytes). Because every switching consists of two steps divided by an RCU grace period, it would be too slow without batching. Please note that the whole batch counts as a single operation (when increasing/decreasing isw_nr_in_flight). This allows to keep umounting working (flush the switching queue), however prevents cleanups from consuming the whole switching quota and effectively blocking the frn switching. A cgwb cleanup operation can fail due to different reasons (e.g. not enough memory, the cgwb has an in-flight/pending io, an attached inode in a wrong state, etc). In this case the next scheduled cleanup will make a new attempt. An attempt is made each time a new cgwb is offlined (in other words a memcg and/or a blkcg is deleted by a user). In the future an additional attempt scheduled by a timer can be implemented. [guro@fb.com: replace open-coded "115" with arithmetic] Link: https://lkml.kernel.org/r/YMEcSBcq/VXMiPPO@carbon.dhcp.thefacebook.com [guro@fb.com: add smp_mb() to inode_prepare_wbs_switch()] Link: https://lkml.kernel.org/r/YMFa+guFw7OFjf3X@carbon.dhcp.thefacebook.com [willy@infradead.org: fix documentation] Link: https://lkml.kernel.org/r/20210615200242.1716568-2-willy@infradead.org Link: https://lkml.kernel.org/r/20210608230225.2078447-9-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:36:03 +00:00
static LIST_HEAD(offline_cgwbs);
static void cleanup_offline_cgwbs_workfn(struct work_struct *work);
static DECLARE_WORK(cleanup_offline_cgwbs_work, cleanup_offline_cgwbs_workfn);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
static void cgwb_release_workfn(struct work_struct *work)
{
struct bdi_writeback *wb = container_of(work, struct bdi_writeback,
release_work);
blkcg: delay blkg destruction until after writeback has finished Currently, blkcg destruction relies on a sequence of events: 1. Destruction starts. blkcg_css_offline() is called and blkgs release their reference to the blkcg. This immediately destroys the cgwbs (writeback). 2. With blkgs giving up their reference, the blkcg ref count should become zero and eventually call blkcg_css_free() which finally frees the blkcg. Jiufei Xue reported that there is a race between blkcg_bio_issue_check() and cgroup_rmdir(). To remedy this, blkg destruction becomes contingent on the completion of all writeback associated with the blkcg. A count of the number of cgwbs is maintained and once that goes to zero, blkg destruction can follow. This should prevent premature blkg destruction related to writeback. The new process for blkcg cleanup is as follows: 1. Destruction starts. blkcg_css_offline() is called which offlines writeback. Blkg destruction is delayed on the cgwb_refcnt count to avoid punting potentially large amounts of outstanding writeback to root while maintaining any ongoing policies. Here, the base cgwb_refcnt is put back. 2. When the cgwb_refcnt becomes zero, blkcg_destroy_blkgs() is called and handles destruction of blkgs. This is where the css reference held by each blkg is released. 3. Once the blkcg ref count goes to zero, blkcg_css_free() is called. This finally frees the blkg. It seems in the past blk-throttle didn't do the most understandable things with taking data from a blkg while associating with current. So, the simplification and unification of what blk-throttle is doing caused this. Fixes: 08e18eab0c579 ("block: add bi_blkg to the bio for cgroups") Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Dennis Zhou <dennisszhou@gmail.com> Cc: Jiufei Xue <jiufei.xue@linux.alibaba.com> Cc: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Tejun Heo <tj@kernel.org> Cc: Josef Bacik <josef@toxicpanda.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-08-31 20:22:43 +00:00
struct blkcg *blkcg = css_to_blkcg(wb->blkcg_css);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
bdi: Fix another oops in wb_workfn() syzbot is reporting NULL pointer dereference at wb_workfn() [1] due to wb->bdi->dev being NULL. And Dmitry confirmed that wb->state was WB_shutting_down after wb->bdi->dev became NULL. This indicates that unregister_bdi() failed to call wb_shutdown() on one of wb objects. The problem is in cgwb_bdi_unregister() which does cgwb_kill() and thus drops bdi's reference to wb structures before going through the list of wbs again and calling wb_shutdown() on each of them. This way the loop iterating through all wbs can easily miss a wb if that wb has already passed through cgwb_remove_from_bdi_list() called from wb_shutdown() from cgwb_release_workfn() and as a result fully shutdown bdi although wb_workfn() for this wb structure is still running. In fact there are also other ways cgwb_bdi_unregister() can race with cgwb_release_workfn() leading e.g. to use-after-free issues: CPU1 CPU2 cgwb_bdi_unregister() cgwb_kill(*slot); cgwb_release() queue_work(cgwb_release_wq, &wb->release_work); cgwb_release_workfn() wb = list_first_entry(&bdi->wb_list, ...) spin_unlock_irq(&cgwb_lock); wb_shutdown(wb); ... kfree_rcu(wb, rcu); wb_shutdown(wb); -> oops use-after-free We solve these issues by synchronizing writeback structure shutdown from cgwb_bdi_unregister() with cgwb_release_workfn() using a new mutex. That way we also no longer need synchronization using WB_shutting_down as the mutex provides it for CONFIG_CGROUP_WRITEBACK case and without CONFIG_CGROUP_WRITEBACK wb_shutdown() can be called only once from bdi_unregister(). Reported-by: syzbot <syzbot+4a7438e774b21ddd8eca@syzkaller.appspotmail.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-06-18 13:46:58 +00:00
mutex_lock(&wb->bdi->cgwb_release_mutex);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
wb_shutdown(wb);
css_put(wb->memcg_css);
css_put(wb->blkcg_css);
bdi: Fix another oops in wb_workfn() syzbot is reporting NULL pointer dereference at wb_workfn() [1] due to wb->bdi->dev being NULL. And Dmitry confirmed that wb->state was WB_shutting_down after wb->bdi->dev became NULL. This indicates that unregister_bdi() failed to call wb_shutdown() on one of wb objects. The problem is in cgwb_bdi_unregister() which does cgwb_kill() and thus drops bdi's reference to wb structures before going through the list of wbs again and calling wb_shutdown() on each of them. This way the loop iterating through all wbs can easily miss a wb if that wb has already passed through cgwb_remove_from_bdi_list() called from wb_shutdown() from cgwb_release_workfn() and as a result fully shutdown bdi although wb_workfn() for this wb structure is still running. In fact there are also other ways cgwb_bdi_unregister() can race with cgwb_release_workfn() leading e.g. to use-after-free issues: CPU1 CPU2 cgwb_bdi_unregister() cgwb_kill(*slot); cgwb_release() queue_work(cgwb_release_wq, &wb->release_work); cgwb_release_workfn() wb = list_first_entry(&bdi->wb_list, ...) spin_unlock_irq(&cgwb_lock); wb_shutdown(wb); ... kfree_rcu(wb, rcu); wb_shutdown(wb); -> oops use-after-free We solve these issues by synchronizing writeback structure shutdown from cgwb_bdi_unregister() with cgwb_release_workfn() using a new mutex. That way we also no longer need synchronization using WB_shutting_down as the mutex provides it for CONFIG_CGROUP_WRITEBACK case and without CONFIG_CGROUP_WRITEBACK wb_shutdown() can be called only once from bdi_unregister(). Reported-by: syzbot <syzbot+4a7438e774b21ddd8eca@syzkaller.appspotmail.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-06-18 13:46:58 +00:00
mutex_unlock(&wb->bdi->cgwb_release_mutex);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
/* triggers blkg destruction if no online users left */
blkcg_unpin_online(blkcg);
blkcg: delay blkg destruction until after writeback has finished Currently, blkcg destruction relies on a sequence of events: 1. Destruction starts. blkcg_css_offline() is called and blkgs release their reference to the blkcg. This immediately destroys the cgwbs (writeback). 2. With blkgs giving up their reference, the blkcg ref count should become zero and eventually call blkcg_css_free() which finally frees the blkcg. Jiufei Xue reported that there is a race between blkcg_bio_issue_check() and cgroup_rmdir(). To remedy this, blkg destruction becomes contingent on the completion of all writeback associated with the blkcg. A count of the number of cgwbs is maintained and once that goes to zero, blkg destruction can follow. This should prevent premature blkg destruction related to writeback. The new process for blkcg cleanup is as follows: 1. Destruction starts. blkcg_css_offline() is called which offlines writeback. Blkg destruction is delayed on the cgwb_refcnt count to avoid punting potentially large amounts of outstanding writeback to root while maintaining any ongoing policies. Here, the base cgwb_refcnt is put back. 2. When the cgwb_refcnt becomes zero, blkcg_destroy_blkgs() is called and handles destruction of blkgs. This is where the css reference held by each blkg is released. 3. Once the blkcg ref count goes to zero, blkcg_css_free() is called. This finally frees the blkg. It seems in the past blk-throttle didn't do the most understandable things with taking data from a blkg while associating with current. So, the simplification and unification of what blk-throttle is doing caused this. Fixes: 08e18eab0c579 ("block: add bi_blkg to the bio for cgroups") Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Dennis Zhou <dennisszhou@gmail.com> Cc: Jiufei Xue <jiufei.xue@linux.alibaba.com> Cc: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Tejun Heo <tj@kernel.org> Cc: Josef Bacik <josef@toxicpanda.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-08-31 20:22:43 +00:00
fprop_local_destroy_percpu(&wb->memcg_completions);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
percpu_ref_exit(&wb->refcnt);
writeback, cgroup: release dying cgwbs by switching attached inodes Asynchronously try to release dying cgwbs by switching attached inodes to the nearest living ancestor wb. It helps to get rid of per-cgroup writeback structures themselves and of pinned memory and block cgroups, which are significantly larger structures (mostly due to large per-cpu statistics data). This prevents memory waste and helps to avoid different scalability problems caused by large piles of dying cgroups. Reuse the existing mechanism of inode switching used for foreign inode detection. To speed things up batch up to 115 inode switching in a single operation (the maximum number is selected so that the resulting struct inode_switch_wbs_context can fit into 1024 bytes). Because every switching consists of two steps divided by an RCU grace period, it would be too slow without batching. Please note that the whole batch counts as a single operation (when increasing/decreasing isw_nr_in_flight). This allows to keep umounting working (flush the switching queue), however prevents cleanups from consuming the whole switching quota and effectively blocking the frn switching. A cgwb cleanup operation can fail due to different reasons (e.g. not enough memory, the cgwb has an in-flight/pending io, an attached inode in a wrong state, etc). In this case the next scheduled cleanup will make a new attempt. An attempt is made each time a new cgwb is offlined (in other words a memcg and/or a blkcg is deleted by a user). In the future an additional attempt scheduled by a timer can be implemented. [guro@fb.com: replace open-coded "115" with arithmetic] Link: https://lkml.kernel.org/r/YMEcSBcq/VXMiPPO@carbon.dhcp.thefacebook.com [guro@fb.com: add smp_mb() to inode_prepare_wbs_switch()] Link: https://lkml.kernel.org/r/YMFa+guFw7OFjf3X@carbon.dhcp.thefacebook.com [willy@infradead.org: fix documentation] Link: https://lkml.kernel.org/r/20210615200242.1716568-2-willy@infradead.org Link: https://lkml.kernel.org/r/20210608230225.2078447-9-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:36:03 +00:00
spin_lock_irq(&cgwb_lock);
list_del(&wb->offline_node);
spin_unlock_irq(&cgwb_lock);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
wb_exit(wb);
writeback, cgroup: keep list of inodes attached to bdi_writeback Currently there is no way to iterate over inodes attached to a specific cgwb structure. It limits the ability to efficiently reclaim the writeback structure itself and associated memory and block cgroup structures without scanning all inodes belonging to a sb, which can be prohibitively expensive. While dirty/in-active-writeback an inode belongs to one of the bdi_writeback's io lists: b_dirty, b_io, b_more_io and b_dirty_time. Once cleaned up, it's removed from all io lists. So the inode->i_io_list can be reused to maintain the list of inodes, attached to a bdi_writeback structure. This patch introduces a new wb->b_attached list, which contains all inodes which were dirty at least once and are attached to the given cgwb. Inodes attached to the root bdi_writeback structures are never placed on such list. The following patch will use this list to try to release cgwbs structures more efficiently. Link: https://lkml.kernel.org/r/20210608230225.2078447-6-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Suggested-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:35:53 +00:00
WARN_ON_ONCE(!list_empty(&wb->b_attached));
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
kfree_rcu(wb, rcu);
}
static void cgwb_release(struct percpu_ref *refcnt)
{
struct bdi_writeback *wb = container_of(refcnt, struct bdi_writeback,
refcnt);
bdi: Move cgroup bdi_writeback to a dedicated low concurrency workqueue From 0aa2e9b921d6db71150633ff290199554f0842a8 Mon Sep 17 00:00:00 2001 From: Tejun Heo <tj@kernel.org> Date: Wed, 23 May 2018 10:29:00 -0700 cgwb_release() punts the actual release to cgwb_release_workfn() on system_wq. Depending on the number of cgroups or block devices, there can be a lot of cgwb_release_workfn() in flight at the same time. We're periodically seeing close to 256 kworkers getting stuck with the following stack trace and overtime the entire system gets stuck. [<ffffffff810ee40c>] _synchronize_rcu_expedited.constprop.72+0x2fc/0x330 [<ffffffff810ee634>] synchronize_rcu_expedited+0x24/0x30 [<ffffffff811ccf23>] bdi_unregister+0x53/0x290 [<ffffffff811cd1e9>] release_bdi+0x89/0xc0 [<ffffffff811cd645>] wb_exit+0x85/0xa0 [<ffffffff811cdc84>] cgwb_release_workfn+0x54/0xb0 [<ffffffff810a68d0>] process_one_work+0x150/0x410 [<ffffffff810a71fd>] worker_thread+0x6d/0x520 [<ffffffff810ad3dc>] kthread+0x12c/0x160 [<ffffffff81969019>] ret_from_fork+0x29/0x40 [<ffffffffffffffff>] 0xffffffffffffffff The events leading to the lockup are... 1. A lot of cgwb_release_workfn() is queued at the same time and all system_wq kworkers are assigned to execute them. 2. They all end up calling synchronize_rcu_expedited(). One of them wins and tries to perform the expedited synchronization. 3. However, that invovles queueing rcu_exp_work to system_wq and waiting for it. Because #1 is holding all available kworkers on system_wq, rcu_exp_work can't be executed. cgwb_release_workfn() is waiting for synchronize_rcu_expedited() which in turn is waiting for cgwb_release_workfn() to free up some of the kworkers. We shouldn't be scheduling hundreds of cgwb_release_workfn() at the same time. There's nothing to be gained from that. This patch updates cgwb release path to use a dedicated percpu workqueue with @max_active of 1. While this resolves the problem at hand, it might be a good idea to isolate rcu_exp_work to its own workqueue too as it can be used from various paths and is prone to this sort of indirect A-A deadlocks. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Cc: stable@vger.kernel.org Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-05-23 17:56:32 +00:00
queue_work(cgwb_release_wq, &wb->release_work);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
}
static void cgwb_kill(struct bdi_writeback *wb)
{
lockdep_assert_held(&cgwb_lock);
WARN_ON(!radix_tree_delete(&wb->bdi->cgwb_tree, wb->memcg_css->id));
list_del(&wb->memcg_node);
list_del(&wb->blkcg_node);
writeback, cgroup: release dying cgwbs by switching attached inodes Asynchronously try to release dying cgwbs by switching attached inodes to the nearest living ancestor wb. It helps to get rid of per-cgroup writeback structures themselves and of pinned memory and block cgroups, which are significantly larger structures (mostly due to large per-cpu statistics data). This prevents memory waste and helps to avoid different scalability problems caused by large piles of dying cgroups. Reuse the existing mechanism of inode switching used for foreign inode detection. To speed things up batch up to 115 inode switching in a single operation (the maximum number is selected so that the resulting struct inode_switch_wbs_context can fit into 1024 bytes). Because every switching consists of two steps divided by an RCU grace period, it would be too slow without batching. Please note that the whole batch counts as a single operation (when increasing/decreasing isw_nr_in_flight). This allows to keep umounting working (flush the switching queue), however prevents cleanups from consuming the whole switching quota and effectively blocking the frn switching. A cgwb cleanup operation can fail due to different reasons (e.g. not enough memory, the cgwb has an in-flight/pending io, an attached inode in a wrong state, etc). In this case the next scheduled cleanup will make a new attempt. An attempt is made each time a new cgwb is offlined (in other words a memcg and/or a blkcg is deleted by a user). In the future an additional attempt scheduled by a timer can be implemented. [guro@fb.com: replace open-coded "115" with arithmetic] Link: https://lkml.kernel.org/r/YMEcSBcq/VXMiPPO@carbon.dhcp.thefacebook.com [guro@fb.com: add smp_mb() to inode_prepare_wbs_switch()] Link: https://lkml.kernel.org/r/YMFa+guFw7OFjf3X@carbon.dhcp.thefacebook.com [willy@infradead.org: fix documentation] Link: https://lkml.kernel.org/r/20210615200242.1716568-2-willy@infradead.org Link: https://lkml.kernel.org/r/20210608230225.2078447-9-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:36:03 +00:00
list_add(&wb->offline_node, &offline_cgwbs);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
percpu_ref_kill(&wb->refcnt);
}
static void cgwb_remove_from_bdi_list(struct bdi_writeback *wb)
{
spin_lock_irq(&cgwb_lock);
list_del_rcu(&wb->bdi_node);
spin_unlock_irq(&cgwb_lock);
}
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
static int cgwb_create(struct backing_dev_info *bdi,
struct cgroup_subsys_state *memcg_css, gfp_t gfp)
{
struct mem_cgroup *memcg;
struct cgroup_subsys_state *blkcg_css;
struct blkcg *blkcg;
struct list_head *memcg_cgwb_list, *blkcg_cgwb_list;
struct bdi_writeback *wb;
unsigned long flags;
int ret = 0;
memcg = mem_cgroup_from_css(memcg_css);
blkcg: rename subsystem name from blkio to io blkio interface has become messy over time and is currently the largest. In addition to the inconsistent naming scheme, it has multiple stat files which report more or less the same thing, a number of debug stat files which expose internal details which shouldn't have been part of the public interface in the first place, recursive and non-recursive stats and leaf and non-leaf knobs. Both recursive vs. non-recursive and leaf vs. non-leaf distinctions don't make any sense on the unified hierarchy as only leaf cgroups can contain processes. cgroups is going through a major interface revision with the unified hierarchy involving significant fundamental usage changes and given that a significant portion of the interface doesn't make sense anymore, it's a good time to reorganize the interface. As the first step, this patch renames the external visible subsystem name from "blkio" to "io". This is more concise, matches the other two major subsystem names, "cpu" and "memory", and better suited as blkcg will be involved in anything writeback related too whether an actual block device is involved or not. As the subsystem legacy_name is set to "blkio", the only userland visible change outside the unified hierarchy is that blkcg is reported as "io" instead of "blkio" in the subsystem initialized message during boot. On the unified hierarchy, blkcg now appears as "io". Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: cgroups@vger.kernel.org Signed-off-by: Jens Axboe <axboe@fb.com>
2015-08-18 21:55:29 +00:00
blkcg_css = cgroup_get_e_css(memcg_css->cgroup, &io_cgrp_subsys);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
blkcg = css_to_blkcg(blkcg_css);
memcg_cgwb_list = &memcg->cgwb_list;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
blkcg_cgwb_list = &blkcg->cgwb_list;
/* look up again under lock and discard on blkcg mismatch */
spin_lock_irqsave(&cgwb_lock, flags);
wb = radix_tree_lookup(&bdi->cgwb_tree, memcg_css->id);
if (wb && wb->blkcg_css != blkcg_css) {
cgwb_kill(wb);
wb = NULL;
}
spin_unlock_irqrestore(&cgwb_lock, flags);
if (wb)
goto out_put;
/* need to create a new one */
wb = kmalloc(sizeof(*wb), gfp);
if (!wb) {
ret = -ENOMEM;
goto out_put;
}
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
ret = wb_init(wb, bdi, gfp);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
if (ret)
goto err_free;
ret = percpu_ref_init(&wb->refcnt, cgwb_release, 0, gfp);
if (ret)
goto err_wb_exit;
ret = fprop_local_init_percpu(&wb->memcg_completions, gfp);
if (ret)
goto err_ref_exit;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
wb->memcg_css = memcg_css;
wb->blkcg_css = blkcg_css;
writeback, cgroup: keep list of inodes attached to bdi_writeback Currently there is no way to iterate over inodes attached to a specific cgwb structure. It limits the ability to efficiently reclaim the writeback structure itself and associated memory and block cgroup structures without scanning all inodes belonging to a sb, which can be prohibitively expensive. While dirty/in-active-writeback an inode belongs to one of the bdi_writeback's io lists: b_dirty, b_io, b_more_io and b_dirty_time. Once cleaned up, it's removed from all io lists. So the inode->i_io_list can be reused to maintain the list of inodes, attached to a bdi_writeback structure. This patch introduces a new wb->b_attached list, which contains all inodes which were dirty at least once and are attached to the given cgwb. Inodes attached to the root bdi_writeback structures are never placed on such list. The following patch will use this list to try to release cgwbs structures more efficiently. Link: https://lkml.kernel.org/r/20210608230225.2078447-6-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Suggested-by: Jan Kara <jack@suse.cz> Reviewed-by: Jan Kara <jack@suse.cz> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:35:53 +00:00
INIT_LIST_HEAD(&wb->b_attached);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
INIT_WORK(&wb->release_work, cgwb_release_workfn);
set_bit(WB_registered, &wb->state);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
/*
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
* The root wb determines the registered state of the whole bdi and
* memcg_cgwb_list and blkcg_cgwb_list's next pointers indicate
* whether they're still online. Don't link @wb if any is dead.
* See wb_memcg_offline() and wb_blkcg_offline().
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
*/
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
ret = -ENODEV;
spin_lock_irqsave(&cgwb_lock, flags);
if (test_bit(WB_registered, &bdi->wb.state) &&
blkcg_cgwb_list->next && memcg_cgwb_list->next) {
/* we might have raced another instance of this function */
ret = radix_tree_insert(&bdi->cgwb_tree, memcg_css->id, wb);
if (!ret) {
writeback: bdi_writeback iteration must not skip dying ones bdi_for_each_wb() is used in several places to wake up or issue writeback work items to all wb's (bdi_writeback's) on a given bdi. The iteration is performed by walking bdi->cgwb_tree; however, the tree only indexes wb's which are currently active. For example, when a memcg gets associated with a different blkcg, the old wb is removed from the tree so that the new one can be indexed. The old wb starts dying from then on but will linger till all its inodes are drained. As these dying wb's may still host dirty inodes, writeback operations which affect all wb's must include them. bdi_for_each_wb() skipping dying wb's led to sync(2) missing and failing to sync the inodes belonging to those wb's. This patch adds a RCU protected @bdi->wb_list which lists all wb's beloinging to that bdi. wb's are added on creation and removed on release rather than on the start of destruction. bdi_for_each_wb() usages are replaced with list_for_each[_continue]_rcu() iterations over @bdi->wb_list and bdi_for_each_wb() and its helpers are removed. v2: Updated as per Jan. last_wb ref leak in bdi_split_work_to_wbs() fixed and unnecessary list head severing in cgwb_bdi_destroy() removed. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-and-tested-by: Artem Bityutskiy <dedekind1@gmail.com> Fixes: ebe41ab0c79d ("writeback: implement bdi_for_each_wb()") Link: http://lkml.kernel.org/g/1443012552.19983.209.camel@gmail.com Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-02 18:47:05 +00:00
list_add_tail_rcu(&wb->bdi_node, &bdi->wb_list);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
list_add(&wb->memcg_node, memcg_cgwb_list);
list_add(&wb->blkcg_node, blkcg_cgwb_list);
blkcg_pin_online(blkcg);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
css_get(memcg_css);
css_get(blkcg_css);
}
}
spin_unlock_irqrestore(&cgwb_lock, flags);
if (ret) {
if (ret == -EEXIST)
ret = 0;
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
goto err_fprop_exit;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
}
goto out_put;
err_fprop_exit:
fprop_local_destroy_percpu(&wb->memcg_completions);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
err_ref_exit:
percpu_ref_exit(&wb->refcnt);
err_wb_exit:
wb_exit(wb);
err_free:
kfree(wb);
out_put:
css_put(blkcg_css);
return ret;
}
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
/**
* wb_get_lookup - get wb for a given memcg
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
* @bdi: target bdi
* @memcg_css: cgroup_subsys_state of the target memcg (must have positive ref)
*
* Try to get the wb for @memcg_css on @bdi. The returned wb has its
* refcount incremented.
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
*
* This function uses css_get() on @memcg_css and thus expects its refcnt
* to be positive on invocation. IOW, rcu_read_lock() protection on
* @memcg_css isn't enough. try_get it before calling this function.
*
* A wb is keyed by its associated memcg. As blkcg implicitly enables
* memcg on the default hierarchy, memcg association is guaranteed to be
* more specific (equal or descendant to the associated blkcg) and thus can
* identify both the memcg and blkcg associations.
*
* Because the blkcg associated with a memcg may change as blkcg is enabled
* and disabled closer to root in the hierarchy, each wb keeps track of
* both the memcg and blkcg associated with it and verifies the blkcg on
* each lookup. On mismatch, the existing wb is discarded and a new one is
* created.
*/
struct bdi_writeback *wb_get_lookup(struct backing_dev_info *bdi,
struct cgroup_subsys_state *memcg_css)
{
struct bdi_writeback *wb;
if (!memcg_css->parent)
return &bdi->wb;
rcu_read_lock();
wb = radix_tree_lookup(&bdi->cgwb_tree, memcg_css->id);
if (wb) {
struct cgroup_subsys_state *blkcg_css;
/* see whether the blkcg association has changed */
blkcg_css = cgroup_get_e_css(memcg_css->cgroup, &io_cgrp_subsys);
if (unlikely(wb->blkcg_css != blkcg_css || !wb_tryget(wb)))
wb = NULL;
css_put(blkcg_css);
}
rcu_read_unlock();
return wb;
}
/**
* wb_get_create - get wb for a given memcg, create if necessary
* @bdi: target bdi
* @memcg_css: cgroup_subsys_state of the target memcg (must have positive ref)
* @gfp: allocation mask to use
*
* Try to get the wb for @memcg_css on @bdi. If it doesn't exist, try to
* create one. See wb_get_lookup() for more details.
*/
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi,
struct cgroup_subsys_state *memcg_css,
gfp_t gfp)
{
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
struct bdi_writeback *wb;
might_alloc(gfp);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
if (!memcg_css->parent)
return &bdi->wb;
do {
wb = wb_get_lookup(bdi, memcg_css);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
} while (!wb && !cgwb_create(bdi, memcg_css, gfp));
return wb;
}
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
static int cgwb_bdi_init(struct backing_dev_info *bdi)
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
{
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
int ret;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
INIT_RADIX_TREE(&bdi->cgwb_tree, GFP_ATOMIC);
bdi: Fix another oops in wb_workfn() syzbot is reporting NULL pointer dereference at wb_workfn() [1] due to wb->bdi->dev being NULL. And Dmitry confirmed that wb->state was WB_shutting_down after wb->bdi->dev became NULL. This indicates that unregister_bdi() failed to call wb_shutdown() on one of wb objects. The problem is in cgwb_bdi_unregister() which does cgwb_kill() and thus drops bdi's reference to wb structures before going through the list of wbs again and calling wb_shutdown() on each of them. This way the loop iterating through all wbs can easily miss a wb if that wb has already passed through cgwb_remove_from_bdi_list() called from wb_shutdown() from cgwb_release_workfn() and as a result fully shutdown bdi although wb_workfn() for this wb structure is still running. In fact there are also other ways cgwb_bdi_unregister() can race with cgwb_release_workfn() leading e.g. to use-after-free issues: CPU1 CPU2 cgwb_bdi_unregister() cgwb_kill(*slot); cgwb_release() queue_work(cgwb_release_wq, &wb->release_work); cgwb_release_workfn() wb = list_first_entry(&bdi->wb_list, ...) spin_unlock_irq(&cgwb_lock); wb_shutdown(wb); ... kfree_rcu(wb, rcu); wb_shutdown(wb); -> oops use-after-free We solve these issues by synchronizing writeback structure shutdown from cgwb_bdi_unregister() with cgwb_release_workfn() using a new mutex. That way we also no longer need synchronization using WB_shutting_down as the mutex provides it for CONFIG_CGROUP_WRITEBACK case and without CONFIG_CGROUP_WRITEBACK wb_shutdown() can be called only once from bdi_unregister(). Reported-by: syzbot <syzbot+4a7438e774b21ddd8eca@syzkaller.appspotmail.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-06-18 13:46:58 +00:00
mutex_init(&bdi->cgwb_release_mutex);
init_rwsem(&bdi->wb_switch_rwsem);
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
ret = wb_init(&bdi->wb, bdi, GFP_KERNEL);
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
if (!ret) {
bdi->wb.memcg_css = &root_mem_cgroup->css;
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
bdi->wb.blkcg_css = blkcg_root_css;
}
return ret;
}
static void cgwb_bdi_unregister(struct backing_dev_info *bdi)
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
{
struct radix_tree_iter iter;
void **slot;
struct bdi_writeback *wb;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
WARN_ON(test_bit(WB_registered, &bdi->wb.state));
spin_lock_irq(&cgwb_lock);
radix_tree_for_each_slot(slot, &bdi->cgwb_tree, &iter, 0)
cgwb_kill(*slot);
bdi: Fix another oops in wb_workfn() syzbot is reporting NULL pointer dereference at wb_workfn() [1] due to wb->bdi->dev being NULL. And Dmitry confirmed that wb->state was WB_shutting_down after wb->bdi->dev became NULL. This indicates that unregister_bdi() failed to call wb_shutdown() on one of wb objects. The problem is in cgwb_bdi_unregister() which does cgwb_kill() and thus drops bdi's reference to wb structures before going through the list of wbs again and calling wb_shutdown() on each of them. This way the loop iterating through all wbs can easily miss a wb if that wb has already passed through cgwb_remove_from_bdi_list() called from wb_shutdown() from cgwb_release_workfn() and as a result fully shutdown bdi although wb_workfn() for this wb structure is still running. In fact there are also other ways cgwb_bdi_unregister() can race with cgwb_release_workfn() leading e.g. to use-after-free issues: CPU1 CPU2 cgwb_bdi_unregister() cgwb_kill(*slot); cgwb_release() queue_work(cgwb_release_wq, &wb->release_work); cgwb_release_workfn() wb = list_first_entry(&bdi->wb_list, ...) spin_unlock_irq(&cgwb_lock); wb_shutdown(wb); ... kfree_rcu(wb, rcu); wb_shutdown(wb); -> oops use-after-free We solve these issues by synchronizing writeback structure shutdown from cgwb_bdi_unregister() with cgwb_release_workfn() using a new mutex. That way we also no longer need synchronization using WB_shutting_down as the mutex provides it for CONFIG_CGROUP_WRITEBACK case and without CONFIG_CGROUP_WRITEBACK wb_shutdown() can be called only once from bdi_unregister(). Reported-by: syzbot <syzbot+4a7438e774b21ddd8eca@syzkaller.appspotmail.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-06-18 13:46:58 +00:00
spin_unlock_irq(&cgwb_lock);
bdi: Fix another oops in wb_workfn() syzbot is reporting NULL pointer dereference at wb_workfn() [1] due to wb->bdi->dev being NULL. And Dmitry confirmed that wb->state was WB_shutting_down after wb->bdi->dev became NULL. This indicates that unregister_bdi() failed to call wb_shutdown() on one of wb objects. The problem is in cgwb_bdi_unregister() which does cgwb_kill() and thus drops bdi's reference to wb structures before going through the list of wbs again and calling wb_shutdown() on each of them. This way the loop iterating through all wbs can easily miss a wb if that wb has already passed through cgwb_remove_from_bdi_list() called from wb_shutdown() from cgwb_release_workfn() and as a result fully shutdown bdi although wb_workfn() for this wb structure is still running. In fact there are also other ways cgwb_bdi_unregister() can race with cgwb_release_workfn() leading e.g. to use-after-free issues: CPU1 CPU2 cgwb_bdi_unregister() cgwb_kill(*slot); cgwb_release() queue_work(cgwb_release_wq, &wb->release_work); cgwb_release_workfn() wb = list_first_entry(&bdi->wb_list, ...) spin_unlock_irq(&cgwb_lock); wb_shutdown(wb); ... kfree_rcu(wb, rcu); wb_shutdown(wb); -> oops use-after-free We solve these issues by synchronizing writeback structure shutdown from cgwb_bdi_unregister() with cgwb_release_workfn() using a new mutex. That way we also no longer need synchronization using WB_shutting_down as the mutex provides it for CONFIG_CGROUP_WRITEBACK case and without CONFIG_CGROUP_WRITEBACK wb_shutdown() can be called only once from bdi_unregister(). Reported-by: syzbot <syzbot+4a7438e774b21ddd8eca@syzkaller.appspotmail.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-06-18 13:46:58 +00:00
mutex_lock(&bdi->cgwb_release_mutex);
spin_lock_irq(&cgwb_lock);
while (!list_empty(&bdi->wb_list)) {
wb = list_first_entry(&bdi->wb_list, struct bdi_writeback,
bdi_node);
spin_unlock_irq(&cgwb_lock);
wb_shutdown(wb);
spin_lock_irq(&cgwb_lock);
}
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
spin_unlock_irq(&cgwb_lock);
bdi: Fix another oops in wb_workfn() syzbot is reporting NULL pointer dereference at wb_workfn() [1] due to wb->bdi->dev being NULL. And Dmitry confirmed that wb->state was WB_shutting_down after wb->bdi->dev became NULL. This indicates that unregister_bdi() failed to call wb_shutdown() on one of wb objects. The problem is in cgwb_bdi_unregister() which does cgwb_kill() and thus drops bdi's reference to wb structures before going through the list of wbs again and calling wb_shutdown() on each of them. This way the loop iterating through all wbs can easily miss a wb if that wb has already passed through cgwb_remove_from_bdi_list() called from wb_shutdown() from cgwb_release_workfn() and as a result fully shutdown bdi although wb_workfn() for this wb structure is still running. In fact there are also other ways cgwb_bdi_unregister() can race with cgwb_release_workfn() leading e.g. to use-after-free issues: CPU1 CPU2 cgwb_bdi_unregister() cgwb_kill(*slot); cgwb_release() queue_work(cgwb_release_wq, &wb->release_work); cgwb_release_workfn() wb = list_first_entry(&bdi->wb_list, ...) spin_unlock_irq(&cgwb_lock); wb_shutdown(wb); ... kfree_rcu(wb, rcu); wb_shutdown(wb); -> oops use-after-free We solve these issues by synchronizing writeback structure shutdown from cgwb_bdi_unregister() with cgwb_release_workfn() using a new mutex. That way we also no longer need synchronization using WB_shutting_down as the mutex provides it for CONFIG_CGROUP_WRITEBACK case and without CONFIG_CGROUP_WRITEBACK wb_shutdown() can be called only once from bdi_unregister(). Reported-by: syzbot <syzbot+4a7438e774b21ddd8eca@syzkaller.appspotmail.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-06-18 13:46:58 +00:00
mutex_unlock(&bdi->cgwb_release_mutex);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
}
writeback, cgroup: release dying cgwbs by switching attached inodes Asynchronously try to release dying cgwbs by switching attached inodes to the nearest living ancestor wb. It helps to get rid of per-cgroup writeback structures themselves and of pinned memory and block cgroups, which are significantly larger structures (mostly due to large per-cpu statistics data). This prevents memory waste and helps to avoid different scalability problems caused by large piles of dying cgroups. Reuse the existing mechanism of inode switching used for foreign inode detection. To speed things up batch up to 115 inode switching in a single operation (the maximum number is selected so that the resulting struct inode_switch_wbs_context can fit into 1024 bytes). Because every switching consists of two steps divided by an RCU grace period, it would be too slow without batching. Please note that the whole batch counts as a single operation (when increasing/decreasing isw_nr_in_flight). This allows to keep umounting working (flush the switching queue), however prevents cleanups from consuming the whole switching quota and effectively blocking the frn switching. A cgwb cleanup operation can fail due to different reasons (e.g. not enough memory, the cgwb has an in-flight/pending io, an attached inode in a wrong state, etc). In this case the next scheduled cleanup will make a new attempt. An attempt is made each time a new cgwb is offlined (in other words a memcg and/or a blkcg is deleted by a user). In the future an additional attempt scheduled by a timer can be implemented. [guro@fb.com: replace open-coded "115" with arithmetic] Link: https://lkml.kernel.org/r/YMEcSBcq/VXMiPPO@carbon.dhcp.thefacebook.com [guro@fb.com: add smp_mb() to inode_prepare_wbs_switch()] Link: https://lkml.kernel.org/r/YMFa+guFw7OFjf3X@carbon.dhcp.thefacebook.com [willy@infradead.org: fix documentation] Link: https://lkml.kernel.org/r/20210615200242.1716568-2-willy@infradead.org Link: https://lkml.kernel.org/r/20210608230225.2078447-9-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:36:03 +00:00
/*
* cleanup_offline_cgwbs_workfn - try to release dying cgwbs
*
* Try to release dying cgwbs by switching attached inodes to the nearest
* living ancestor's writeback. Processed wbs are placed at the end
* of the list to guarantee the forward progress.
*/
static void cleanup_offline_cgwbs_workfn(struct work_struct *work)
{
struct bdi_writeback *wb;
LIST_HEAD(processed);
spin_lock_irq(&cgwb_lock);
while (!list_empty(&offline_cgwbs)) {
wb = list_first_entry(&offline_cgwbs, struct bdi_writeback,
offline_node);
list_move(&wb->offline_node, &processed);
/*
* If wb is dirty, cleaning up the writeback by switching
* attached inodes will result in an effective removal of any
* bandwidth restrictions, which isn't the goal. Instead,
* it can be postponed until the next time, when all io
* will be likely completed. If in the meantime some inodes
* will get re-dirtied, they should be eventually switched to
* a new cgwb.
*/
if (wb_has_dirty_io(wb))
continue;
if (!wb_tryget(wb))
continue;
spin_unlock_irq(&cgwb_lock);
while (cleanup_offline_cgwb(wb))
cond_resched();
spin_lock_irq(&cgwb_lock);
wb_put(wb);
}
if (!list_empty(&processed))
list_splice_tail(&processed, &offline_cgwbs);
spin_unlock_irq(&cgwb_lock);
}
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
/**
* wb_memcg_offline - kill all wb's associated with a memcg being offlined
* @memcg: memcg being offlined
*
* Also prevents creation of any new wb's associated with @memcg.
writeback: bdi write bandwidth estimation The estimation value will start from 100MB/s and adapt to the real bandwidth in seconds. It tries to update the bandwidth only when disk is fully utilized. Any inactive period of more than one second will be skipped. The estimated bandwidth will be reflecting how fast the device can writeout when _fully utilized_, and won't drop to 0 when it goes idle. The value will remain constant at disk idle time. At busy write time, if not considering fluctuations, it will also remain high unless be knocked down by possible concurrent reads that compete for the disk time and bandwidth with async writes. The estimation is not done purely in the flusher because there is no guarantee for write_cache_pages() to return timely to update bandwidth. The bdi->avg_write_bandwidth smoothing is very effective for filtering out sudden spikes, however may be a little biased in long term. The overheads are low because the bdi bandwidth update only occurs at 200ms intervals. The 200ms update interval is suitable, because it's not possible to get the real bandwidth for the instance at all, due to large fluctuations. The NFS commits can be as large as seconds worth of data. One XFS completion may be as large as half second worth of data if we are going to increase the write chunk to half second worth of data. In ext4, fluctuations with time period of around 5 seconds is observed. And there is another pattern of irregular periods of up to 20 seconds on SSD tests. That's why we are not only doing the estimation at 200ms intervals, but also averaging them over a period of 3 seconds and then go further to do another level of smoothing in avg_write_bandwidth. CC: Li Shaohua <shaohua.li@intel.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-08-29 17:22:30 +00:00
*/
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
void wb_memcg_offline(struct mem_cgroup *memcg)
{
struct list_head *memcg_cgwb_list = &memcg->cgwb_list;
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
struct bdi_writeback *wb, *next;
spin_lock_irq(&cgwb_lock);
list_for_each_entry_safe(wb, next, memcg_cgwb_list, memcg_node)
cgwb_kill(wb);
memcg_cgwb_list->next = NULL; /* prevent new wb's */
spin_unlock_irq(&cgwb_lock);
writeback, cgroup: release dying cgwbs by switching attached inodes Asynchronously try to release dying cgwbs by switching attached inodes to the nearest living ancestor wb. It helps to get rid of per-cgroup writeback structures themselves and of pinned memory and block cgroups, which are significantly larger structures (mostly due to large per-cpu statistics data). This prevents memory waste and helps to avoid different scalability problems caused by large piles of dying cgroups. Reuse the existing mechanism of inode switching used for foreign inode detection. To speed things up batch up to 115 inode switching in a single operation (the maximum number is selected so that the resulting struct inode_switch_wbs_context can fit into 1024 bytes). Because every switching consists of two steps divided by an RCU grace period, it would be too slow without batching. Please note that the whole batch counts as a single operation (when increasing/decreasing isw_nr_in_flight). This allows to keep umounting working (flush the switching queue), however prevents cleanups from consuming the whole switching quota and effectively blocking the frn switching. A cgwb cleanup operation can fail due to different reasons (e.g. not enough memory, the cgwb has an in-flight/pending io, an attached inode in a wrong state, etc). In this case the next scheduled cleanup will make a new attempt. An attempt is made each time a new cgwb is offlined (in other words a memcg and/or a blkcg is deleted by a user). In the future an additional attempt scheduled by a timer can be implemented. [guro@fb.com: replace open-coded "115" with arithmetic] Link: https://lkml.kernel.org/r/YMEcSBcq/VXMiPPO@carbon.dhcp.thefacebook.com [guro@fb.com: add smp_mb() to inode_prepare_wbs_switch()] Link: https://lkml.kernel.org/r/YMFa+guFw7OFjf3X@carbon.dhcp.thefacebook.com [willy@infradead.org: fix documentation] Link: https://lkml.kernel.org/r/20210615200242.1716568-2-willy@infradead.org Link: https://lkml.kernel.org/r/20210608230225.2078447-9-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Acked-by: Dennis Zhou <dennis@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-06-29 02:36:03 +00:00
queue_work(system_unbound_wq, &cleanup_offline_cgwbs_work);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
}
/**
* wb_blkcg_offline - kill all wb's associated with a blkcg being offlined
* @blkcg: blkcg being offlined
*
* Also prevents creation of any new wb's associated with @blkcg.
*/
void wb_blkcg_offline(struct blkcg *blkcg)
{
struct bdi_writeback *wb, *next;
spin_lock_irq(&cgwb_lock);
list_for_each_entry_safe(wb, next, &blkcg->cgwb_list, blkcg_node)
cgwb_kill(wb);
blkcg->cgwb_list.next = NULL; /* prevent new wb's */
spin_unlock_irq(&cgwb_lock);
}
static void cgwb_bdi_register(struct backing_dev_info *bdi)
{
spin_lock_irq(&cgwb_lock);
list_add_tail_rcu(&bdi->wb.bdi_node, &bdi->wb_list);
spin_unlock_irq(&cgwb_lock);
}
bdi: Move cgroup bdi_writeback to a dedicated low concurrency workqueue From 0aa2e9b921d6db71150633ff290199554f0842a8 Mon Sep 17 00:00:00 2001 From: Tejun Heo <tj@kernel.org> Date: Wed, 23 May 2018 10:29:00 -0700 cgwb_release() punts the actual release to cgwb_release_workfn() on system_wq. Depending on the number of cgroups or block devices, there can be a lot of cgwb_release_workfn() in flight at the same time. We're periodically seeing close to 256 kworkers getting stuck with the following stack trace and overtime the entire system gets stuck. [<ffffffff810ee40c>] _synchronize_rcu_expedited.constprop.72+0x2fc/0x330 [<ffffffff810ee634>] synchronize_rcu_expedited+0x24/0x30 [<ffffffff811ccf23>] bdi_unregister+0x53/0x290 [<ffffffff811cd1e9>] release_bdi+0x89/0xc0 [<ffffffff811cd645>] wb_exit+0x85/0xa0 [<ffffffff811cdc84>] cgwb_release_workfn+0x54/0xb0 [<ffffffff810a68d0>] process_one_work+0x150/0x410 [<ffffffff810a71fd>] worker_thread+0x6d/0x520 [<ffffffff810ad3dc>] kthread+0x12c/0x160 [<ffffffff81969019>] ret_from_fork+0x29/0x40 [<ffffffffffffffff>] 0xffffffffffffffff The events leading to the lockup are... 1. A lot of cgwb_release_workfn() is queued at the same time and all system_wq kworkers are assigned to execute them. 2. They all end up calling synchronize_rcu_expedited(). One of them wins and tries to perform the expedited synchronization. 3. However, that invovles queueing rcu_exp_work to system_wq and waiting for it. Because #1 is holding all available kworkers on system_wq, rcu_exp_work can't be executed. cgwb_release_workfn() is waiting for synchronize_rcu_expedited() which in turn is waiting for cgwb_release_workfn() to free up some of the kworkers. We shouldn't be scheduling hundreds of cgwb_release_workfn() at the same time. There's nothing to be gained from that. This patch updates cgwb release path to use a dedicated percpu workqueue with @max_active of 1. While this resolves the problem at hand, it might be a good idea to isolate rcu_exp_work to its own workqueue too as it can be used from various paths and is prone to this sort of indirect A-A deadlocks. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Cc: stable@vger.kernel.org Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-05-23 17:56:32 +00:00
static int __init cgwb_init(void)
{
/*
* There can be many concurrent release work items overwhelming
* system_wq. Put them in a separate wq and limit concurrency.
* There's no point in executing many of these in parallel.
*/
cgwb_release_wq = alloc_workqueue("cgwb_release", 0, 1);
if (!cgwb_release_wq)
return -ENOMEM;
return 0;
}
subsys_initcall(cgwb_init);
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
#else /* CONFIG_CGROUP_WRITEBACK */
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
static int cgwb_bdi_init(struct backing_dev_info *bdi)
{
return wb_init(&bdi->wb, bdi, GFP_KERNEL);
writeback: don't embed root bdi_writeback_congested in bdi_writeback 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") made bdi (backing_dev_info) host per-cgroup wb's (bdi_writeback's). As the congested state needs to be per-wb and referenced from blkcg side and multiple wbs, the patch made all non-root cong's (bdi_writeback_congested's) reference counted and indexed on bdi. When a bdi is destroyed, cgwb_bdi_destroy() tries to drain all non-root cong's; however, this can hang indefinitely because wb's can also be referenced from blkcg_gq's which are destroyed after bdi destruction is complete. To fix the bug, bdi destruction will be updated to not wait for cong's to drain, which naturally means that cong's may outlive the associated bdi. This is fine for non-root cong's but is problematic for the root cong's which are embedded in their bdi's as they may end up getting dereferenced after the containing bdi's are freed. This patch makes root cong's behave the same as non-root cong's. They are no longer embedded in their bdi's but allocated separately during bdi initialization, indexed and reference counted the same way. * As cong handling is the same for all wb's, wb->congested initialization is moved into wb_init(). * When !CONFIG_CGROUP_WRITEBACK, there was no indexing or refcnting. bdi->wb_congested is now a pointer pointing to the root cong allocated during bdi init and minimal refcnting operations are implemented. * The above makes root wb init paths diverge depending on CONFIG_CGROUP_WRITEBACK. root wb init is moved to cgwb_bdi_init(). This patch in itself shouldn't cause any consequential behavior differences but prepares for the actual fix. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jon Christopherson <jon@jons.org> Link: https://bugzilla.kernel.org/show_bug.cgi?id=100681 Tested-by: Jon Christopherson <jon@jons.org> Added <linux/slab.h> include to backing-dev.h for kfree() definition. Signed-off-by: Jens Axboe <axboe@fb.com>
2015-07-02 14:44:34 +00:00
}
static void cgwb_bdi_unregister(struct backing_dev_info *bdi) { }
static void cgwb_bdi_register(struct backing_dev_info *bdi)
{
list_add_tail_rcu(&bdi->wb.bdi_node, &bdi->wb_list);
}
static void cgwb_remove_from_bdi_list(struct bdi_writeback *wb)
{
list_del_rcu(&wb->bdi_node);
}
writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: kbuild test robot <fengguang.wu@intel.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:37 +00:00
#endif /* CONFIG_CGROUP_WRITEBACK */
writeback: bdi write bandwidth estimation The estimation value will start from 100MB/s and adapt to the real bandwidth in seconds. It tries to update the bandwidth only when disk is fully utilized. Any inactive period of more than one second will be skipped. The estimated bandwidth will be reflecting how fast the device can writeout when _fully utilized_, and won't drop to 0 when it goes idle. The value will remain constant at disk idle time. At busy write time, if not considering fluctuations, it will also remain high unless be knocked down by possible concurrent reads that compete for the disk time and bandwidth with async writes. The estimation is not done purely in the flusher because there is no guarantee for write_cache_pages() to return timely to update bandwidth. The bdi->avg_write_bandwidth smoothing is very effective for filtering out sudden spikes, however may be a little biased in long term. The overheads are low because the bdi bandwidth update only occurs at 200ms intervals. The 200ms update interval is suitable, because it's not possible to get the real bandwidth for the instance at all, due to large fluctuations. The NFS commits can be as large as seconds worth of data. One XFS completion may be as large as half second worth of data if we are going to increase the write chunk to half second worth of data. In ext4, fluctuations with time period of around 5 seconds is observed. And there is another pattern of irregular periods of up to 20 seconds on SSD tests. That's why we are not only doing the estimation at 200ms intervals, but also averaging them over a period of 3 seconds and then go further to do another level of smoothing in avg_write_bandwidth. CC: Li Shaohua <shaohua.li@intel.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-08-29 17:22:30 +00:00
static int bdi_init(struct backing_dev_info *bdi)
{
writeback: bdi_writeback iteration must not skip dying ones bdi_for_each_wb() is used in several places to wake up or issue writeback work items to all wb's (bdi_writeback's) on a given bdi. The iteration is performed by walking bdi->cgwb_tree; however, the tree only indexes wb's which are currently active. For example, when a memcg gets associated with a different blkcg, the old wb is removed from the tree so that the new one can be indexed. The old wb starts dying from then on but will linger till all its inodes are drained. As these dying wb's may still host dirty inodes, writeback operations which affect all wb's must include them. bdi_for_each_wb() skipping dying wb's led to sync(2) missing and failing to sync the inodes belonging to those wb's. This patch adds a RCU protected @bdi->wb_list which lists all wb's beloinging to that bdi. wb's are added on creation and removed on release rather than on the start of destruction. bdi_for_each_wb() usages are replaced with list_for_each[_continue]_rcu() iterations over @bdi->wb_list and bdi_for_each_wb() and its helpers are removed. v2: Updated as per Jan. last_wb ref leak in bdi_split_work_to_wbs() fixed and unnecessary list head severing in cgwb_bdi_destroy() removed. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-and-tested-by: Artem Bityutskiy <dedekind1@gmail.com> Fixes: ebe41ab0c79d ("writeback: implement bdi_for_each_wb()") Link: http://lkml.kernel.org/g/1443012552.19983.209.camel@gmail.com Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-02 18:47:05 +00:00
int ret;
bdi->dev = NULL;
kref_init(&bdi->refcnt);
bdi->min_ratio = 0;
bdi->max_ratio = 100;
bdi->max_prop_frac = FPROP_FRAC_BASE;
INIT_LIST_HEAD(&bdi->bdi_list);
writeback: bdi_writeback iteration must not skip dying ones bdi_for_each_wb() is used in several places to wake up or issue writeback work items to all wb's (bdi_writeback's) on a given bdi. The iteration is performed by walking bdi->cgwb_tree; however, the tree only indexes wb's which are currently active. For example, when a memcg gets associated with a different blkcg, the old wb is removed from the tree so that the new one can be indexed. The old wb starts dying from then on but will linger till all its inodes are drained. As these dying wb's may still host dirty inodes, writeback operations which affect all wb's must include them. bdi_for_each_wb() skipping dying wb's led to sync(2) missing and failing to sync the inodes belonging to those wb's. This patch adds a RCU protected @bdi->wb_list which lists all wb's beloinging to that bdi. wb's are added on creation and removed on release rather than on the start of destruction. bdi_for_each_wb() usages are replaced with list_for_each[_continue]_rcu() iterations over @bdi->wb_list and bdi_for_each_wb() and its helpers are removed. v2: Updated as per Jan. last_wb ref leak in bdi_split_work_to_wbs() fixed and unnecessary list head severing in cgwb_bdi_destroy() removed. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-and-tested-by: Artem Bityutskiy <dedekind1@gmail.com> Fixes: ebe41ab0c79d ("writeback: implement bdi_for_each_wb()") Link: http://lkml.kernel.org/g/1443012552.19983.209.camel@gmail.com Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-02 18:47:05 +00:00
INIT_LIST_HEAD(&bdi->wb_list);
init_waitqueue_head(&bdi->wb_waitq);
writeback: switch to per-bdi threads for flushing data This gets rid of pdflush for bdi writeout and kupdated style cleaning. pdflush writeout suffers from lack of locality and also requires more threads to handle the same workload, since it has to work in a non-blocking fashion against each queue. This also introduces lumpy behaviour and potential request starvation, since pdflush can be starved for queue access if others are accessing it. A sample ffsb workload that does random writes to files is about 8% faster here on a simple SATA drive during the benchmark phase. File layout also seems a LOT more smooth in vmstat: r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 608848 2652 375372 0 0 0 71024 604 24 1 10 48 42 0 1 0 549644 2712 433736 0 0 0 60692 505 27 1 8 48 44 1 0 0 476928 2784 505192 0 0 4 29540 553 24 0 9 53 37 0 1 0 457972 2808 524008 0 0 0 54876 331 16 0 4 38 58 0 1 0 366128 2928 614284 0 0 4 92168 710 58 0 13 53 34 0 1 0 295092 3000 684140 0 0 0 62924 572 23 0 9 53 37 0 1 0 236592 3064 741704 0 0 4 58256 523 17 0 8 48 44 0 1 0 165608 3132 811464 0 0 0 57460 560 21 0 8 54 38 0 1 0 102952 3200 873164 0 0 4 74748 540 29 1 10 48 41 0 1 0 48604 3252 926472 0 0 0 53248 469 29 0 7 47 45 where vanilla tends to fluctuate a lot in the creation phase: r b swpd free buff cache si so bi bo in cs us sy id wa 1 1 0 678716 5792 303380 0 0 0 74064 565 50 1 11 52 36 1 0 0 662488 5864 319396 0 0 4 352 302 329 0 2 47 51 0 1 0 599312 5924 381468 0 0 0 78164 516 55 0 9 51 40 0 1 0 519952 6008 459516 0 0 4 78156 622 56 1 11 52 37 1 1 0 436640 6092 541632 0 0 0 82244 622 54 0 11 48 41 0 1 0 436640 6092 541660 0 0 0 8 152 39 0 0 51 49 0 1 0 332224 6200 644252 0 0 4 102800 728 46 1 13 49 36 1 0 0 274492 6260 701056 0 0 4 12328 459 49 0 7 50 43 0 1 0 211220 6324 763356 0 0 0 106940 515 37 1 10 51 39 1 0 0 160412 6376 813468 0 0 0 8224 415 43 0 6 49 45 1 1 0 85980 6452 886556 0 0 4 113516 575 39 1 11 54 34 0 2 0 85968 6452 886620 0 0 0 1640 158 211 0 0 46 54 A 10 disk test with btrfs performs 26% faster with per-bdi flushing. A SSD based writeback test on XFS performs over 20% better as well, with the throughput being very stable around 1GB/sec, where pdflush only manages 750MB/sec and fluctuates wildly while doing so. Random buffered writes to many files behave a lot better as well, as does random mmap'ed writes. A separate thread is added to sync the super blocks. In the long term, adding sync_supers_bdi() functionality could get rid of this thread again. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2009-09-09 07:08:54 +00:00
writeback: bdi_writeback iteration must not skip dying ones bdi_for_each_wb() is used in several places to wake up or issue writeback work items to all wb's (bdi_writeback's) on a given bdi. The iteration is performed by walking bdi->cgwb_tree; however, the tree only indexes wb's which are currently active. For example, when a memcg gets associated with a different blkcg, the old wb is removed from the tree so that the new one can be indexed. The old wb starts dying from then on but will linger till all its inodes are drained. As these dying wb's may still host dirty inodes, writeback operations which affect all wb's must include them. bdi_for_each_wb() skipping dying wb's led to sync(2) missing and failing to sync the inodes belonging to those wb's. This patch adds a RCU protected @bdi->wb_list which lists all wb's beloinging to that bdi. wb's are added on creation and removed on release rather than on the start of destruction. bdi_for_each_wb() usages are replaced with list_for_each[_continue]_rcu() iterations over @bdi->wb_list and bdi_for_each_wb() and its helpers are removed. v2: Updated as per Jan. last_wb ref leak in bdi_split_work_to_wbs() fixed and unnecessary list head severing in cgwb_bdi_destroy() removed. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-and-tested-by: Artem Bityutskiy <dedekind1@gmail.com> Fixes: ebe41ab0c79d ("writeback: implement bdi_for_each_wb()") Link: http://lkml.kernel.org/g/1443012552.19983.209.camel@gmail.com Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-02 18:47:05 +00:00
ret = cgwb_bdi_init(bdi);
return ret;
}
writeback: bdi write bandwidth estimation The estimation value will start from 100MB/s and adapt to the real bandwidth in seconds. It tries to update the bandwidth only when disk is fully utilized. Any inactive period of more than one second will be skipped. The estimated bandwidth will be reflecting how fast the device can writeout when _fully utilized_, and won't drop to 0 when it goes idle. The value will remain constant at disk idle time. At busy write time, if not considering fluctuations, it will also remain high unless be knocked down by possible concurrent reads that compete for the disk time and bandwidth with async writes. The estimation is not done purely in the flusher because there is no guarantee for write_cache_pages() to return timely to update bandwidth. The bdi->avg_write_bandwidth smoothing is very effective for filtering out sudden spikes, however may be a little biased in long term. The overheads are low because the bdi bandwidth update only occurs at 200ms intervals. The 200ms update interval is suitable, because it's not possible to get the real bandwidth for the instance at all, due to large fluctuations. The NFS commits can be as large as seconds worth of data. One XFS completion may be as large as half second worth of data if we are going to increase the write chunk to half second worth of data. In ext4, fluctuations with time period of around 5 seconds is observed. And there is another pattern of irregular periods of up to 20 seconds on SSD tests. That's why we are not only doing the estimation at 200ms intervals, but also averaging them over a period of 3 seconds and then go further to do another level of smoothing in avg_write_bandwidth. CC: Li Shaohua <shaohua.li@intel.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-08-29 17:22:30 +00:00
struct backing_dev_info *bdi_alloc(int node_id)
{
struct backing_dev_info *bdi;
bdi = kzalloc_node(sizeof(*bdi), GFP_KERNEL, node_id);
if (!bdi)
return NULL;
if (bdi_init(bdi)) {
kfree(bdi);
return NULL;
}
bdi->capabilities = BDI_CAP_WRITEBACK | BDI_CAP_WRITEBACK_ACCT;
bdi->ra_pages = VM_READAHEAD_PAGES;
bdi->io_pages = VM_READAHEAD_PAGES;
return bdi;
}
EXPORT_SYMBOL(bdi_alloc);
static struct rb_node **bdi_lookup_rb_node(u64 id, struct rb_node **parentp)
{
struct rb_node **p = &bdi_tree.rb_node;
struct rb_node *parent = NULL;
struct backing_dev_info *bdi;
lockdep_assert_held(&bdi_lock);
while (*p) {
parent = *p;
bdi = rb_entry(parent, struct backing_dev_info, rb_node);
if (bdi->id > id)
p = &(*p)->rb_left;
else if (bdi->id < id)
p = &(*p)->rb_right;
else
break;
}
if (parentp)
*parentp = parent;
return p;
}
/**
* bdi_get_by_id - lookup and get bdi from its id
* @id: bdi id to lookup
*
* Find bdi matching @id and get it. Returns NULL if the matching bdi
* doesn't exist or is already unregistered.
*/
struct backing_dev_info *bdi_get_by_id(u64 id)
{
struct backing_dev_info *bdi = NULL;
struct rb_node **p;
spin_lock_bh(&bdi_lock);
p = bdi_lookup_rb_node(id, NULL);
if (*p) {
bdi = rb_entry(*p, struct backing_dev_info, rb_node);
bdi_get(bdi);
}
spin_unlock_bh(&bdi_lock);
return bdi;
}
int bdi_register_va(struct backing_dev_info *bdi, const char *fmt, va_list args)
{
struct device *dev;
struct rb_node *parent, **p;
writeback: bdi write bandwidth estimation The estimation value will start from 100MB/s and adapt to the real bandwidth in seconds. It tries to update the bandwidth only when disk is fully utilized. Any inactive period of more than one second will be skipped. The estimated bandwidth will be reflecting how fast the device can writeout when _fully utilized_, and won't drop to 0 when it goes idle. The value will remain constant at disk idle time. At busy write time, if not considering fluctuations, it will also remain high unless be knocked down by possible concurrent reads that compete for the disk time and bandwidth with async writes. The estimation is not done purely in the flusher because there is no guarantee for write_cache_pages() to return timely to update bandwidth. The bdi->avg_write_bandwidth smoothing is very effective for filtering out sudden spikes, however may be a little biased in long term. The overheads are low because the bdi bandwidth update only occurs at 200ms intervals. The 200ms update interval is suitable, because it's not possible to get the real bandwidth for the instance at all, due to large fluctuations. The NFS commits can be as large as seconds worth of data. One XFS completion may be as large as half second worth of data if we are going to increase the write chunk to half second worth of data. In ext4, fluctuations with time period of around 5 seconds is observed. And there is another pattern of irregular periods of up to 20 seconds on SSD tests. That's why we are not only doing the estimation at 200ms intervals, but also averaging them over a period of 3 seconds and then go further to do another level of smoothing in avg_write_bandwidth. CC: Li Shaohua <shaohua.li@intel.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-08-29 17:22:30 +00:00
if (bdi->dev) /* The driver needs to use separate queues per device */
return 0;
writeback: bdi write bandwidth estimation The estimation value will start from 100MB/s and adapt to the real bandwidth in seconds. It tries to update the bandwidth only when disk is fully utilized. Any inactive period of more than one second will be skipped. The estimated bandwidth will be reflecting how fast the device can writeout when _fully utilized_, and won't drop to 0 when it goes idle. The value will remain constant at disk idle time. At busy write time, if not considering fluctuations, it will also remain high unless be knocked down by possible concurrent reads that compete for the disk time and bandwidth with async writes. The estimation is not done purely in the flusher because there is no guarantee for write_cache_pages() to return timely to update bandwidth. The bdi->avg_write_bandwidth smoothing is very effective for filtering out sudden spikes, however may be a little biased in long term. The overheads are low because the bdi bandwidth update only occurs at 200ms intervals. The 200ms update interval is suitable, because it's not possible to get the real bandwidth for the instance at all, due to large fluctuations. The NFS commits can be as large as seconds worth of data. One XFS completion may be as large as half second worth of data if we are going to increase the write chunk to half second worth of data. In ext4, fluctuations with time period of around 5 seconds is observed. And there is another pattern of irregular periods of up to 20 seconds on SSD tests. That's why we are not only doing the estimation at 200ms intervals, but also averaging them over a period of 3 seconds and then go further to do another level of smoothing in avg_write_bandwidth. CC: Li Shaohua <shaohua.li@intel.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-08-29 17:22:30 +00:00
vsnprintf(bdi->dev_name, sizeof(bdi->dev_name), fmt, args);
dev = device_create(bdi_class, NULL, MKDEV(0, 0), bdi, bdi->dev_name);
if (IS_ERR(dev))
return PTR_ERR(dev);
mm: per device dirty threshold Scale writeback cache per backing device, proportional to its writeout speed. By decoupling the BDI dirty thresholds a number of problems we currently have will go away, namely: - mutual interference starvation (for any number of BDIs); - deadlocks with stacked BDIs (loop, FUSE and local NFS mounts). It might be that all dirty pages are for a single BDI while other BDIs are idling. By giving each BDI a 'fair' share of the dirty limit, each one can have dirty pages outstanding and make progress. A global threshold also creates a deadlock for stacked BDIs; when A writes to B, and A generates enough dirty pages to get throttled, B will never start writeback until the dirty pages go away. Again, by giving each BDI its own 'independent' dirty limit, this problem is avoided. So the problem is to determine how to distribute the total dirty limit across the BDIs fairly and efficiently. A DBI that has a large dirty limit but does not have any dirty pages outstanding is a waste. What is done is to keep a floating proportion between the DBIs based on writeback completions. This way faster/more active devices get a larger share than slower/idle devices. [akpm@linux-foundation.org: fix warnings] [hugh@veritas.com: Fix occasional hang when a task couldn't get out of balance_dirty_pages] Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 06:25:50 +00:00
cgwb_bdi_register(bdi);
bdi->dev = dev;
bdi_debug_register(bdi, dev_name(dev));
set_bit(WB_registered, &bdi->wb.state);
spin_lock_bh(&bdi_lock);
bdi->id = ++bdi_id_cursor;
p = bdi_lookup_rb_node(bdi->id, &parent);
rb_link_node(&bdi->rb_node, parent, p);
rb_insert_color(&bdi->rb_node, &bdi_tree);
list_add_tail_rcu(&bdi->bdi_list, &bdi_list);
spin_unlock_bh(&bdi_lock);
trace_writeback_bdi_register(bdi);
return 0;
}
int bdi_register(struct backing_dev_info *bdi, const char *fmt, ...)
{
va_list args;
int ret;
va_start(args, fmt);
ret = bdi_register_va(bdi, fmt, args);
va_end(args);
return ret;
}
EXPORT_SYMBOL(bdi_register);
void bdi_set_owner(struct backing_dev_info *bdi, struct device *owner)
block: fix bdi vs gendisk lifetime mismatch The name for a bdi of a gendisk is derived from the gendisk's devt. However, since the gendisk is destroyed before the bdi it leaves a window where a new gendisk could dynamically reuse the same devt while a bdi with the same name is still live. Arrange for the bdi to hold a reference against its "owner" disk device while it is registered. Otherwise we can hit sysfs duplicate name collisions like the following: WARNING: CPU: 10 PID: 2078 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x64/0x80 sysfs: cannot create duplicate filename '/devices/virtual/bdi/259:1' Hardware name: HP ProLiant DL580 Gen8, BIOS P79 05/06/2015 0000000000000286 0000000002c04ad5 ffff88006f24f970 ffffffff8134caec ffff88006f24f9c0 0000000000000000 ffff88006f24f9b0 ffffffff8108c351 0000001f0000000c ffff88105d236000 ffff88105d1031e0 ffff8800357427f8 Call Trace: [<ffffffff8134caec>] dump_stack+0x63/0x87 [<ffffffff8108c351>] __warn+0xd1/0xf0 [<ffffffff8108c3cf>] warn_slowpath_fmt+0x5f/0x80 [<ffffffff812a0d34>] sysfs_warn_dup+0x64/0x80 [<ffffffff812a0e1e>] sysfs_create_dir_ns+0x7e/0x90 [<ffffffff8134faaa>] kobject_add_internal+0xaa/0x320 [<ffffffff81358d4e>] ? vsnprintf+0x34e/0x4d0 [<ffffffff8134ff55>] kobject_add+0x75/0xd0 [<ffffffff816e66b2>] ? mutex_lock+0x12/0x2f [<ffffffff8148b0a5>] device_add+0x125/0x610 [<ffffffff8148b788>] device_create_groups_vargs+0xd8/0x100 [<ffffffff8148b7cc>] device_create_vargs+0x1c/0x20 [<ffffffff811b775c>] bdi_register+0x8c/0x180 [<ffffffff811b7877>] bdi_register_dev+0x27/0x30 [<ffffffff813317f5>] add_disk+0x175/0x4a0 Cc: <stable@vger.kernel.org> Reported-by: Yi Zhang <yizhan@redhat.com> Tested-by: Yi Zhang <yizhan@redhat.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Fixed up missing 0 return in bdi_register_owner(). Signed-off-by: Jens Axboe <axboe@fb.com>
2016-07-31 18:15:13 +00:00
{
WARN_ON_ONCE(bdi->owner);
block: fix bdi vs gendisk lifetime mismatch The name for a bdi of a gendisk is derived from the gendisk's devt. However, since the gendisk is destroyed before the bdi it leaves a window where a new gendisk could dynamically reuse the same devt while a bdi with the same name is still live. Arrange for the bdi to hold a reference against its "owner" disk device while it is registered. Otherwise we can hit sysfs duplicate name collisions like the following: WARNING: CPU: 10 PID: 2078 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x64/0x80 sysfs: cannot create duplicate filename '/devices/virtual/bdi/259:1' Hardware name: HP ProLiant DL580 Gen8, BIOS P79 05/06/2015 0000000000000286 0000000002c04ad5 ffff88006f24f970 ffffffff8134caec ffff88006f24f9c0 0000000000000000 ffff88006f24f9b0 ffffffff8108c351 0000001f0000000c ffff88105d236000 ffff88105d1031e0 ffff8800357427f8 Call Trace: [<ffffffff8134caec>] dump_stack+0x63/0x87 [<ffffffff8108c351>] __warn+0xd1/0xf0 [<ffffffff8108c3cf>] warn_slowpath_fmt+0x5f/0x80 [<ffffffff812a0d34>] sysfs_warn_dup+0x64/0x80 [<ffffffff812a0e1e>] sysfs_create_dir_ns+0x7e/0x90 [<ffffffff8134faaa>] kobject_add_internal+0xaa/0x320 [<ffffffff81358d4e>] ? vsnprintf+0x34e/0x4d0 [<ffffffff8134ff55>] kobject_add+0x75/0xd0 [<ffffffff816e66b2>] ? mutex_lock+0x12/0x2f [<ffffffff8148b0a5>] device_add+0x125/0x610 [<ffffffff8148b788>] device_create_groups_vargs+0xd8/0x100 [<ffffffff8148b7cc>] device_create_vargs+0x1c/0x20 [<ffffffff811b775c>] bdi_register+0x8c/0x180 [<ffffffff811b7877>] bdi_register_dev+0x27/0x30 [<ffffffff813317f5>] add_disk+0x175/0x4a0 Cc: <stable@vger.kernel.org> Reported-by: Yi Zhang <yizhan@redhat.com> Tested-by: Yi Zhang <yizhan@redhat.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Fixed up missing 0 return in bdi_register_owner(). Signed-off-by: Jens Axboe <axboe@fb.com>
2016-07-31 18:15:13 +00:00
bdi->owner = owner;
get_device(owner);
}
/*
* Remove bdi from bdi_list, and ensure that it is no longer visible
*/
static void bdi_remove_from_list(struct backing_dev_info *bdi)
{
spin_lock_bh(&bdi_lock);
rb_erase(&bdi->rb_node, &bdi_tree);
list_del_rcu(&bdi->bdi_list);
spin_unlock_bh(&bdi_lock);
synchronize_rcu_expedited();
}
block: don't release bdi while request_queue has live references bdi's are initialized in two steps, bdi_init() and bdi_register(), but destroyed in a single step by bdi_destroy() which, for a bdi embedded in a request_queue, is called during blk_cleanup_queue() which makes the queue invisible and starts the draining of remaining usages. A request_queue's user can access the congestion state of the embedded bdi as long as it holds a reference to the queue. As such, it may access the congested state of a queue which finished blk_cleanup_queue() but hasn't reached blk_release_queue() yet. Because the congested state was embedded in backing_dev_info which in turn is embedded in request_queue, accessing the congested state after bdi_destroy() was called was fine. The bdi was destroyed but the memory region for the congested state remained accessible till the queue got released. a13f35e87140 ("writeback: don't embed root bdi_writeback_congested in bdi_writeback") changed the situation. Now, the root congested state which is expected to be pinned while request_queue remains accessible is separately reference counted and the base ref is put during bdi_destroy(). This means that the root congested state may go away prematurely while the queue is between bdi_dstroy() and blk_cleanup_queue(), which was detected by Andrey's KASAN tests. The root cause of this problem is that bdi doesn't distinguish the two steps of destruction, unregistration and release, and now the root congested state actually requires a separate release step. To fix the issue, this patch separates out bdi_unregister() and bdi_exit() from bdi_destroy(). bdi_unregister() is called from blk_cleanup_queue() and bdi_exit() from blk_release_queue(). bdi_destroy() is now just a simple wrapper calling the two steps back-to-back. While at it, the prototype of bdi_destroy() is moved right below bdi_setup_and_register() so that the counterpart operations are located together. Signed-off-by: Tejun Heo <tj@kernel.org> Fixes: a13f35e87140 ("writeback: don't embed root bdi_writeback_congested in bdi_writeback") Cc: stable@vger.kernel.org # v4.2+ Reported-and-tested-by: Andrey Konovalov <andreyknvl@google.com> Link: http://lkml.kernel.org/g/CAAeHK+zUJ74Zn17=rOyxacHU18SgCfC6bsYW=6kCY5GXJBwGfQ@mail.gmail.com Reviewed-by: Jan Kara <jack@suse.com> Reviewed-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-09-08 16:20:22 +00:00
void bdi_unregister(struct backing_dev_info *bdi)
{
writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-22 21:13:30 +00:00
/* make sure nobody finds us on the bdi_list anymore */
bdi_remove_from_list(bdi);
wb_shutdown(&bdi->wb);
cgwb_bdi_unregister(bdi);
backing-dev: ensure wakeup_timer is deleted bdi_prune_sb() in bdi_unregister() attempts to removes the bdi links from all super_blocks and then del_timer_sync() the writeback timer. However, this can race with __mark_inode_dirty(), leading to bdi_wakeup_thread_delayed() rearming the writeback timer on the bdi we're unregistering, after we've called del_timer_sync(). This can end up with the bdi being freed with an active timer inside it, as in the case of the following dump after the removal of an SD card. Fix this by redoing the del_timer_sync() in bdi_destory(). ------------[ cut here ]------------ WARNING: at /home/rabin/kernel/arm/lib/debugobjects.c:262 debug_print_object+0x9c/0xc8() ODEBUG: free active (active state 0) object type: timer_list hint: wakeup_timer_fn+0x0/0x180 Modules linked in: Backtrace: [<c00109dc>] (dump_backtrace+0x0/0x110) from [<c0236e4c>] (dump_stack+0x18/0x1c) r6:c02bc638 r5:00000106 r4:c79f5d18 r3:00000000 [<c0236e34>] (dump_stack+0x0/0x1c) from [<c0025e6c>] (warn_slowpath_common+0x54/0x6c) [<c0025e18>] (warn_slowpath_common+0x0/0x6c) from [<c0025f28>] (warn_slowpath_fmt+0x38/0x40) r8:20000013 r7:c780c6f0 r6:c031613c r5:c780c6f0 r4:c02b1b29 r3:00000009 [<c0025ef0>] (warn_slowpath_fmt+0x0/0x40) from [<c015eb4c>] (debug_print_object+0x9c/0xc8) r3:c02b1b29 r2:c02bc662 [<c015eab0>] (debug_print_object+0x0/0xc8) from [<c015f574>] (debug_check_no_obj_freed+0xac/0x1dc) r6:c7964000 r5:00000001 r4:c7964000 [<c015f4c8>] (debug_check_no_obj_freed+0x0/0x1dc) from [<c00a9e38>] (kmem_cache_free+0x88/0x1f8) [<c00a9db0>] (kmem_cache_free+0x0/0x1f8) from [<c014286c>] (blk_release_queue+0x70/0x78) [<c01427fc>] (blk_release_queue+0x0/0x78) from [<c015290c>] (kobject_release+0x70/0x84) r5:c79641f0 r4:c796420c [<c015289c>] (kobject_release+0x0/0x84) from [<c0153ce4>] (kref_put+0x68/0x80) r7:00000083 r6:c74083d0 r5:c015289c r4:c796420c [<c0153c7c>] (kref_put+0x0/0x80) from [<c01527d0>] (kobject_put+0x48/0x5c) r5:c79643b4 r4:c79641f0 [<c0152788>] (kobject_put+0x0/0x5c) from [<c013ddd8>] (blk_cleanup_queue+0x68/0x74) r4:c7964000 [<c013dd70>] (blk_cleanup_queue+0x0/0x74) from [<c01a6370>] (mmc_blk_put+0x78/0xe8) r5:00000000 r4:c794c400 [<c01a62f8>] (mmc_blk_put+0x0/0xe8) from [<c01a64b4>] (mmc_blk_release+0x24/0x38) r5:c794c400 r4:c0322824 [<c01a6490>] (mmc_blk_release+0x0/0x38) from [<c00de11c>] (__blkdev_put+0xe8/0x170) r5:c78d5e00 r4:c74083c0 [<c00de034>] (__blkdev_put+0x0/0x170) from [<c00de2c0>] (blkdev_put+0x11c/0x12c) r8:c79f5f70 r7:00000001 r6:c74083d0 r5:00000083 r4:c74083c0 r3:00000000 [<c00de1a4>] (blkdev_put+0x0/0x12c) from [<c00b0724>] (kill_block_super+0x60/0x6c) r7:c7942300 r6:c79f4000 r5:00000083 r4:c74083c0 [<c00b06c4>] (kill_block_super+0x0/0x6c) from [<c00b0a94>] (deactivate_locked_super+0x44/0x70) r6:c79f4000 r5:c031af64 r4:c794dc00 r3:c00b06c4 [<c00b0a50>] (deactivate_locked_super+0x0/0x70) from [<c00b1358>] (deactivate_super+0x6c/0x70) r5:c794dc00 r4:c794dc00 [<c00b12ec>] (deactivate_super+0x0/0x70) from [<c00c88b0>] (mntput_no_expire+0x188/0x194) r5:c794dc00 r4:c7942300 [<c00c8728>] (mntput_no_expire+0x0/0x194) from [<c00c95e0>] (sys_umount+0x2e4/0x310) r6:c7942300 r5:00000000 r4:00000000 r3:00000000 [<c00c92fc>] (sys_umount+0x0/0x310) from [<c000d940>] (ret_fast_syscall+0x0/0x30) ---[ end trace e5c83c92ada51c76 ]--- Cc: stable@kernel.org Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2011-11-11 12:29:04 +00:00
if (bdi->dev) {
bdi_debug_unregister(bdi);
device_unregister(bdi->dev);
bdi->dev = NULL;
}
block: fix bdi vs gendisk lifetime mismatch The name for a bdi of a gendisk is derived from the gendisk's devt. However, since the gendisk is destroyed before the bdi it leaves a window where a new gendisk could dynamically reuse the same devt while a bdi with the same name is still live. Arrange for the bdi to hold a reference against its "owner" disk device while it is registered. Otherwise we can hit sysfs duplicate name collisions like the following: WARNING: CPU: 10 PID: 2078 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x64/0x80 sysfs: cannot create duplicate filename '/devices/virtual/bdi/259:1' Hardware name: HP ProLiant DL580 Gen8, BIOS P79 05/06/2015 0000000000000286 0000000002c04ad5 ffff88006f24f970 ffffffff8134caec ffff88006f24f9c0 0000000000000000 ffff88006f24f9b0 ffffffff8108c351 0000001f0000000c ffff88105d236000 ffff88105d1031e0 ffff8800357427f8 Call Trace: [<ffffffff8134caec>] dump_stack+0x63/0x87 [<ffffffff8108c351>] __warn+0xd1/0xf0 [<ffffffff8108c3cf>] warn_slowpath_fmt+0x5f/0x80 [<ffffffff812a0d34>] sysfs_warn_dup+0x64/0x80 [<ffffffff812a0e1e>] sysfs_create_dir_ns+0x7e/0x90 [<ffffffff8134faaa>] kobject_add_internal+0xaa/0x320 [<ffffffff81358d4e>] ? vsnprintf+0x34e/0x4d0 [<ffffffff8134ff55>] kobject_add+0x75/0xd0 [<ffffffff816e66b2>] ? mutex_lock+0x12/0x2f [<ffffffff8148b0a5>] device_add+0x125/0x610 [<ffffffff8148b788>] device_create_groups_vargs+0xd8/0x100 [<ffffffff8148b7cc>] device_create_vargs+0x1c/0x20 [<ffffffff811b775c>] bdi_register+0x8c/0x180 [<ffffffff811b7877>] bdi_register_dev+0x27/0x30 [<ffffffff813317f5>] add_disk+0x175/0x4a0 Cc: <stable@vger.kernel.org> Reported-by: Yi Zhang <yizhan@redhat.com> Tested-by: Yi Zhang <yizhan@redhat.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Fixed up missing 0 return in bdi_register_owner(). Signed-off-by: Jens Axboe <axboe@fb.com>
2016-07-31 18:15:13 +00:00
if (bdi->owner) {
put_device(bdi->owner);
bdi->owner = NULL;
}
block: don't release bdi while request_queue has live references bdi's are initialized in two steps, bdi_init() and bdi_register(), but destroyed in a single step by bdi_destroy() which, for a bdi embedded in a request_queue, is called during blk_cleanup_queue() which makes the queue invisible and starts the draining of remaining usages. A request_queue's user can access the congestion state of the embedded bdi as long as it holds a reference to the queue. As such, it may access the congested state of a queue which finished blk_cleanup_queue() but hasn't reached blk_release_queue() yet. Because the congested state was embedded in backing_dev_info which in turn is embedded in request_queue, accessing the congested state after bdi_destroy() was called was fine. The bdi was destroyed but the memory region for the congested state remained accessible till the queue got released. a13f35e87140 ("writeback: don't embed root bdi_writeback_congested in bdi_writeback") changed the situation. Now, the root congested state which is expected to be pinned while request_queue remains accessible is separately reference counted and the base ref is put during bdi_destroy(). This means that the root congested state may go away prematurely while the queue is between bdi_dstroy() and blk_cleanup_queue(), which was detected by Andrey's KASAN tests. The root cause of this problem is that bdi doesn't distinguish the two steps of destruction, unregistration and release, and now the root congested state actually requires a separate release step. To fix the issue, this patch separates out bdi_unregister() and bdi_exit() from bdi_destroy(). bdi_unregister() is called from blk_cleanup_queue() and bdi_exit() from blk_release_queue(). bdi_destroy() is now just a simple wrapper calling the two steps back-to-back. While at it, the prototype of bdi_destroy() is moved right below bdi_setup_and_register() so that the counterpart operations are located together. Signed-off-by: Tejun Heo <tj@kernel.org> Fixes: a13f35e87140 ("writeback: don't embed root bdi_writeback_congested in bdi_writeback") Cc: stable@vger.kernel.org # v4.2+ Reported-and-tested-by: Andrey Konovalov <andreyknvl@google.com> Link: http://lkml.kernel.org/g/CAAeHK+zUJ74Zn17=rOyxacHU18SgCfC6bsYW=6kCY5GXJBwGfQ@mail.gmail.com Reviewed-by: Jan Kara <jack@suse.com> Reviewed-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-09-08 16:20:22 +00:00
}
static void release_bdi(struct kref *ref)
{
struct backing_dev_info *bdi =
container_of(ref, struct backing_dev_info, refcnt);
if (test_bit(WB_registered, &bdi->wb.state))
bdi_unregister(bdi);
WARN_ON_ONCE(bdi->dev);
wb_exit(&bdi->wb);
kfree(bdi);
}
void bdi_put(struct backing_dev_info *bdi)
{
kref_put(&bdi->refcnt, release_bdi);
}
EXPORT_SYMBOL(bdi_put);
const char *bdi_dev_name(struct backing_dev_info *bdi)
{
if (!bdi || !bdi->dev)
return bdi_unknown_name;
return bdi->dev_name;
}
EXPORT_SYMBOL_GPL(bdi_dev_name);
static wait_queue_head_t congestion_wqh[2] = {
__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
};
static atomic_t nr_wb_congested[2];
void clear_bdi_congested(struct backing_dev_info *bdi, int sync)
{
wait_queue_head_t *wqh = &congestion_wqh[sync];
enum wb_congested_state bit;
bit = sync ? WB_sync_congested : WB_async_congested;
if (test_and_clear_bit(bit, &bdi->wb.congested))
atomic_dec(&nr_wb_congested[sync]);
smp_mb__after_atomic();
if (waitqueue_active(wqh))
wake_up(wqh);
}
EXPORT_SYMBOL(clear_bdi_congested);
void set_bdi_congested(struct backing_dev_info *bdi, int sync)
{
enum wb_congested_state bit;
bit = sync ? WB_sync_congested : WB_async_congested;
if (!test_and_set_bit(bit, &bdi->wb.congested))
atomic_inc(&nr_wb_congested[sync]);
}
EXPORT_SYMBOL(set_bdi_congested);
/**
* congestion_wait - wait for a backing_dev to become uncongested
* @sync: SYNC or ASYNC IO
* @timeout: timeout in jiffies
*
* Waits for up to @timeout jiffies for a backing_dev (any backing_dev) to exit
* write congestion. If no backing_devs are congested then just wait for the
* next write to be completed.
*/
long congestion_wait(int sync, long timeout)
{
long ret;
unsigned long start = jiffies;
DEFINE_WAIT(wait);
wait_queue_head_t *wqh = &congestion_wqh[sync];
prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
ret = io_schedule_timeout(timeout);
finish_wait(wqh, &wait);
trace_writeback_congestion_wait(jiffies_to_usecs(timeout),
jiffies_to_usecs(jiffies - start));
return ret;
}
EXPORT_SYMBOL(congestion_wait);
mm: per device dirty threshold Scale writeback cache per backing device, proportional to its writeout speed. By decoupling the BDI dirty thresholds a number of problems we currently have will go away, namely: - mutual interference starvation (for any number of BDIs); - deadlocks with stacked BDIs (loop, FUSE and local NFS mounts). It might be that all dirty pages are for a single BDI while other BDIs are idling. By giving each BDI a 'fair' share of the dirty limit, each one can have dirty pages outstanding and make progress. A global threshold also creates a deadlock for stacked BDIs; when A writes to B, and A generates enough dirty pages to get throttled, B will never start writeback until the dirty pages go away. Again, by giving each BDI its own 'independent' dirty limit, this problem is avoided. So the problem is to determine how to distribute the total dirty limit across the BDIs fairly and efficiently. A DBI that has a large dirty limit but does not have any dirty pages outstanding is a waste. What is done is to keep a floating proportion between the DBIs based on writeback completions. This way faster/more active devices get a larger share than slower/idle devices. [akpm@linux-foundation.org: fix warnings] [hugh@veritas.com: Fix occasional hang when a task couldn't get out of balance_dirty_pages] Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 06:25:50 +00:00
/**
mm, vmscan: move LRU lists to node This moves the LRU lists from the zone to the node and related data such as counters, tracing, congestion tracking and writeback tracking. Unfortunately, due to reclaim and compaction retry logic, it is necessary to account for the number of LRU pages on both zone and node logic. Most reclaim logic is based on the node counters but the retry logic uses the zone counters which do not distinguish inactive and active sizes. It would be possible to leave the LRU counters on a per-zone basis but it's a heavier calculation across multiple cache lines that is much more frequent than the retry checks. Other than the LRU counters, this is mostly a mechanical patch but note that it introduces a number of anomalies. For example, the scans are per-zone but using per-node counters. We also mark a node as congested when a zone is congested. This causes weird problems that are fixed later but is easier to review. In the event that there is excessive overhead on 32-bit systems due to the nodes being on LRU then there are two potential solutions 1. Long-term isolation of highmem pages when reclaim is lowmem When pages are skipped, they are immediately added back onto the LRU list. If lowmem reclaim persisted for long periods of time, the same highmem pages get continually scanned. The idea would be that lowmem keeps those pages on a separate list until a reclaim for highmem pages arrives that splices the highmem pages back onto the LRU. It potentially could be implemented similar to the UNEVICTABLE list. That would reduce the skip rate with the potential corner case is that highmem pages have to be scanned and reclaimed to free lowmem slab pages. 2. Linear scan lowmem pages if the initial LRU shrink fails This will break LRU ordering but may be preferable and faster during memory pressure than skipping LRU pages. Link: http://lkml.kernel.org/r/1467970510-21195-4-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Rik van Riel <riel@surriel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-28 22:45:31 +00:00
* wait_iff_congested - Conditionally wait for a backing_dev to become uncongested or a pgdat to complete writes
* @sync: SYNC or ASYNC IO
* @timeout: timeout in jiffies
*
mm/vmscan: don't mess with pgdat->flags in memcg reclaim memcg reclaim may alter pgdat->flags based on the state of LRU lists in cgroup and its children. PGDAT_WRITEBACK may force kswapd to sleep congested_wait(), PGDAT_DIRTY may force kswapd to writeback filesystem pages. But the worst here is PGDAT_CONGESTED, since it may force all direct reclaims to stall in wait_iff_congested(). Note that only kswapd have powers to clear any of these bits. This might just never happen if cgroup limits configured that way. So all direct reclaims will stall as long as we have some congested bdi in the system. Leave all pgdat->flags manipulations to kswapd. kswapd scans the whole pgdat, only kswapd can clear pgdat->flags once node is balanced, thus it's reasonable to leave all decisions about node state to kswapd. Why only kswapd? Why not allow to global direct reclaim change these flags? It is because currently only kswapd can clear these flags. I'm less worried about the case when PGDAT_CONGESTED falsely not set, and more worried about the case when it falsely set. If direct reclaimer sets PGDAT_CONGESTED, do we have guarantee that after the congestion problem is sorted out, kswapd will be woken up and clear the flag? It seems like there is no such guarantee. E.g. direct reclaimers may eventually balance pgdat and kswapd simply won't wake up (see wakeup_kswapd()). Moving pgdat->flags manipulation to kswapd, means that cgroup2 recalim now loses its congestion throttling mechanism. Add per-cgroup congestion state and throttle cgroup2 reclaimers if memcg is in congestion state. Currently there is no need in per-cgroup PGDAT_WRITEBACK and PGDAT_DIRTY bits since they alter only kswapd behavior. The problem could be easily demonstrated by creating heavy congestion in one cgroup: echo "+memory" > /sys/fs/cgroup/cgroup.subtree_control mkdir -p /sys/fs/cgroup/congester echo 512M > /sys/fs/cgroup/congester/memory.max echo $$ > /sys/fs/cgroup/congester/cgroup.procs /* generate a lot of diry data on slow HDD */ while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & .... while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & and some job in another cgroup: mkdir /sys/fs/cgroup/victim echo 128M > /sys/fs/cgroup/victim/memory.max # time cat /dev/sda > /dev/null real 10m15.054s user 0m0.487s sys 1m8.505s According to the tracepoint in wait_iff_congested(), the 'cat' spent 50% of the time sleeping there. With the patch, cat don't waste time anymore: # time cat /dev/sda > /dev/null real 5m32.911s user 0m0.411s sys 0m56.664s [aryabinin@virtuozzo.com: congestion state should be per-node] Link: http://lkml.kernel.org/r/20180406135215.10057-1-aryabinin@virtuozzo.com [ayabinin@virtuozzo.com: make congestion state per-cgroup-per-node instead of just per-cgroup[ Link: http://lkml.kernel.org/r/20180406180254.8970-2-aryabinin@virtuozzo.com Link: http://lkml.kernel.org/r/20180323152029.11084-5-aryabinin@virtuozzo.com Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Tejun Heo <tj@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-04-10 23:28:03 +00:00
* In the event of a congested backing_dev (any backing_dev) this waits
* for up to @timeout jiffies for either a BDI to exit congestion of the
* given @sync queue or a write to complete.
*
* The return value is 0 if the sleep is for the full timeout. Otherwise,
* it is the number of jiffies that were still remaining when the function
* returned. return_value == timeout implies the function did not sleep.
*/
mm/vmscan: don't mess with pgdat->flags in memcg reclaim memcg reclaim may alter pgdat->flags based on the state of LRU lists in cgroup and its children. PGDAT_WRITEBACK may force kswapd to sleep congested_wait(), PGDAT_DIRTY may force kswapd to writeback filesystem pages. But the worst here is PGDAT_CONGESTED, since it may force all direct reclaims to stall in wait_iff_congested(). Note that only kswapd have powers to clear any of these bits. This might just never happen if cgroup limits configured that way. So all direct reclaims will stall as long as we have some congested bdi in the system. Leave all pgdat->flags manipulations to kswapd. kswapd scans the whole pgdat, only kswapd can clear pgdat->flags once node is balanced, thus it's reasonable to leave all decisions about node state to kswapd. Why only kswapd? Why not allow to global direct reclaim change these flags? It is because currently only kswapd can clear these flags. I'm less worried about the case when PGDAT_CONGESTED falsely not set, and more worried about the case when it falsely set. If direct reclaimer sets PGDAT_CONGESTED, do we have guarantee that after the congestion problem is sorted out, kswapd will be woken up and clear the flag? It seems like there is no such guarantee. E.g. direct reclaimers may eventually balance pgdat and kswapd simply won't wake up (see wakeup_kswapd()). Moving pgdat->flags manipulation to kswapd, means that cgroup2 recalim now loses its congestion throttling mechanism. Add per-cgroup congestion state and throttle cgroup2 reclaimers if memcg is in congestion state. Currently there is no need in per-cgroup PGDAT_WRITEBACK and PGDAT_DIRTY bits since they alter only kswapd behavior. The problem could be easily demonstrated by creating heavy congestion in one cgroup: echo "+memory" > /sys/fs/cgroup/cgroup.subtree_control mkdir -p /sys/fs/cgroup/congester echo 512M > /sys/fs/cgroup/congester/memory.max echo $$ > /sys/fs/cgroup/congester/cgroup.procs /* generate a lot of diry data on slow HDD */ while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & .... while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & and some job in another cgroup: mkdir /sys/fs/cgroup/victim echo 128M > /sys/fs/cgroup/victim/memory.max # time cat /dev/sda > /dev/null real 10m15.054s user 0m0.487s sys 1m8.505s According to the tracepoint in wait_iff_congested(), the 'cat' spent 50% of the time sleeping there. With the patch, cat don't waste time anymore: # time cat /dev/sda > /dev/null real 5m32.911s user 0m0.411s sys 0m56.664s [aryabinin@virtuozzo.com: congestion state should be per-node] Link: http://lkml.kernel.org/r/20180406135215.10057-1-aryabinin@virtuozzo.com [ayabinin@virtuozzo.com: make congestion state per-cgroup-per-node instead of just per-cgroup[ Link: http://lkml.kernel.org/r/20180406180254.8970-2-aryabinin@virtuozzo.com Link: http://lkml.kernel.org/r/20180323152029.11084-5-aryabinin@virtuozzo.com Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Tejun Heo <tj@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-04-10 23:28:03 +00:00
long wait_iff_congested(int sync, long timeout)
{
long ret;
unsigned long start = jiffies;
DEFINE_WAIT(wait);
wait_queue_head_t *wqh = &congestion_wqh[sync];
/*
mm/vmscan: don't mess with pgdat->flags in memcg reclaim memcg reclaim may alter pgdat->flags based on the state of LRU lists in cgroup and its children. PGDAT_WRITEBACK may force kswapd to sleep congested_wait(), PGDAT_DIRTY may force kswapd to writeback filesystem pages. But the worst here is PGDAT_CONGESTED, since it may force all direct reclaims to stall in wait_iff_congested(). Note that only kswapd have powers to clear any of these bits. This might just never happen if cgroup limits configured that way. So all direct reclaims will stall as long as we have some congested bdi in the system. Leave all pgdat->flags manipulations to kswapd. kswapd scans the whole pgdat, only kswapd can clear pgdat->flags once node is balanced, thus it's reasonable to leave all decisions about node state to kswapd. Why only kswapd? Why not allow to global direct reclaim change these flags? It is because currently only kswapd can clear these flags. I'm less worried about the case when PGDAT_CONGESTED falsely not set, and more worried about the case when it falsely set. If direct reclaimer sets PGDAT_CONGESTED, do we have guarantee that after the congestion problem is sorted out, kswapd will be woken up and clear the flag? It seems like there is no such guarantee. E.g. direct reclaimers may eventually balance pgdat and kswapd simply won't wake up (see wakeup_kswapd()). Moving pgdat->flags manipulation to kswapd, means that cgroup2 recalim now loses its congestion throttling mechanism. Add per-cgroup congestion state and throttle cgroup2 reclaimers if memcg is in congestion state. Currently there is no need in per-cgroup PGDAT_WRITEBACK and PGDAT_DIRTY bits since they alter only kswapd behavior. The problem could be easily demonstrated by creating heavy congestion in one cgroup: echo "+memory" > /sys/fs/cgroup/cgroup.subtree_control mkdir -p /sys/fs/cgroup/congester echo 512M > /sys/fs/cgroup/congester/memory.max echo $$ > /sys/fs/cgroup/congester/cgroup.procs /* generate a lot of diry data on slow HDD */ while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & .... while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & and some job in another cgroup: mkdir /sys/fs/cgroup/victim echo 128M > /sys/fs/cgroup/victim/memory.max # time cat /dev/sda > /dev/null real 10m15.054s user 0m0.487s sys 1m8.505s According to the tracepoint in wait_iff_congested(), the 'cat' spent 50% of the time sleeping there. With the patch, cat don't waste time anymore: # time cat /dev/sda > /dev/null real 5m32.911s user 0m0.411s sys 0m56.664s [aryabinin@virtuozzo.com: congestion state should be per-node] Link: http://lkml.kernel.org/r/20180406135215.10057-1-aryabinin@virtuozzo.com [ayabinin@virtuozzo.com: make congestion state per-cgroup-per-node instead of just per-cgroup[ Link: http://lkml.kernel.org/r/20180406180254.8970-2-aryabinin@virtuozzo.com Link: http://lkml.kernel.org/r/20180323152029.11084-5-aryabinin@virtuozzo.com Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Tejun Heo <tj@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-04-10 23:28:03 +00:00
* If there is no congestion, yield if necessary instead
* of sleeping on the congestion queue
*/
mm/vmscan: don't mess with pgdat->flags in memcg reclaim memcg reclaim may alter pgdat->flags based on the state of LRU lists in cgroup and its children. PGDAT_WRITEBACK may force kswapd to sleep congested_wait(), PGDAT_DIRTY may force kswapd to writeback filesystem pages. But the worst here is PGDAT_CONGESTED, since it may force all direct reclaims to stall in wait_iff_congested(). Note that only kswapd have powers to clear any of these bits. This might just never happen if cgroup limits configured that way. So all direct reclaims will stall as long as we have some congested bdi in the system. Leave all pgdat->flags manipulations to kswapd. kswapd scans the whole pgdat, only kswapd can clear pgdat->flags once node is balanced, thus it's reasonable to leave all decisions about node state to kswapd. Why only kswapd? Why not allow to global direct reclaim change these flags? It is because currently only kswapd can clear these flags. I'm less worried about the case when PGDAT_CONGESTED falsely not set, and more worried about the case when it falsely set. If direct reclaimer sets PGDAT_CONGESTED, do we have guarantee that after the congestion problem is sorted out, kswapd will be woken up and clear the flag? It seems like there is no such guarantee. E.g. direct reclaimers may eventually balance pgdat and kswapd simply won't wake up (see wakeup_kswapd()). Moving pgdat->flags manipulation to kswapd, means that cgroup2 recalim now loses its congestion throttling mechanism. Add per-cgroup congestion state and throttle cgroup2 reclaimers if memcg is in congestion state. Currently there is no need in per-cgroup PGDAT_WRITEBACK and PGDAT_DIRTY bits since they alter only kswapd behavior. The problem could be easily demonstrated by creating heavy congestion in one cgroup: echo "+memory" > /sys/fs/cgroup/cgroup.subtree_control mkdir -p /sys/fs/cgroup/congester echo 512M > /sys/fs/cgroup/congester/memory.max echo $$ > /sys/fs/cgroup/congester/cgroup.procs /* generate a lot of diry data on slow HDD */ while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & .... while true; do dd if=/dev/zero of=/mnt/sdb/zeroes bs=1M count=1024; done & and some job in another cgroup: mkdir /sys/fs/cgroup/victim echo 128M > /sys/fs/cgroup/victim/memory.max # time cat /dev/sda > /dev/null real 10m15.054s user 0m0.487s sys 1m8.505s According to the tracepoint in wait_iff_congested(), the 'cat' spent 50% of the time sleeping there. With the patch, cat don't waste time anymore: # time cat /dev/sda > /dev/null real 5m32.911s user 0m0.411s sys 0m56.664s [aryabinin@virtuozzo.com: congestion state should be per-node] Link: http://lkml.kernel.org/r/20180406135215.10057-1-aryabinin@virtuozzo.com [ayabinin@virtuozzo.com: make congestion state per-cgroup-per-node instead of just per-cgroup[ Link: http://lkml.kernel.org/r/20180406180254.8970-2-aryabinin@virtuozzo.com Link: http://lkml.kernel.org/r/20180323152029.11084-5-aryabinin@virtuozzo.com Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Tejun Heo <tj@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-04-10 23:28:03 +00:00
if (atomic_read(&nr_wb_congested[sync]) == 0) {
mm: throttle on IO only when there are too many dirty and writeback pages wait_iff_congested has been used to throttle allocator before it retried another round of direct reclaim to allow the writeback to make some progress and prevent reclaim from looping over dirty/writeback pages without making any progress. We used to do congestion_wait before commit 0e093d99763e ("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") but that led to undesirable stalls and sleeping for the full timeout even when the BDI wasn't congested. Hence wait_iff_congested was used instead. But it seems that even wait_iff_congested doesn't work as expected. We might have a small file LRU list with all pages dirty/writeback and yet the bdi is not congested so this is just a cond_resched in the end and can end up triggering pre mature OOM. This patch replaces the unconditional wait_iff_congested by congestion_wait which is executed only if we _know_ that the last round of direct reclaim didn't make any progress and dirty+writeback pages are more than a half of the reclaimable pages on the zone which might be usable for our target allocation. This shouldn't reintroduce stalls fixed by 0e093d99763e because congestion_wait is called only when we are getting hopeless when sleeping is a better choice than OOM with many pages under IO. We have to preserve logic introduced by commit 373ccbe59270 ("mm, vmstat: allow WQ concurrency to discover memory reclaim doesn't make any progress") into the __alloc_pages_slowpath now that wait_iff_congested is not used anymore. As the only remaining user of wait_iff_congested is shrink_inactive_list we can remove the WQ specific short sleep from wait_iff_congested because the sleep is needed to be done only once in the allocation retry cycle. [mhocko@suse.com: high_zoneidx->ac_classzone_idx to evaluate memory reserves properly] Link: http://lkml.kernel.org/r/1463051677-29418-2-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Joonsoo Kim <js1304@gmail.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Vladimir Davydov <vdavydov@virtuozzo.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-20 23:57:03 +00:00
cond_resched();
mm, vmscan: move LRU lists to node This moves the LRU lists from the zone to the node and related data such as counters, tracing, congestion tracking and writeback tracking. Unfortunately, due to reclaim and compaction retry logic, it is necessary to account for the number of LRU pages on both zone and node logic. Most reclaim logic is based on the node counters but the retry logic uses the zone counters which do not distinguish inactive and active sizes. It would be possible to leave the LRU counters on a per-zone basis but it's a heavier calculation across multiple cache lines that is much more frequent than the retry checks. Other than the LRU counters, this is mostly a mechanical patch but note that it introduces a number of anomalies. For example, the scans are per-zone but using per-node counters. We also mark a node as congested when a zone is congested. This causes weird problems that are fixed later but is easier to review. In the event that there is excessive overhead on 32-bit systems due to the nodes being on LRU then there are two potential solutions 1. Long-term isolation of highmem pages when reclaim is lowmem When pages are skipped, they are immediately added back onto the LRU list. If lowmem reclaim persisted for long periods of time, the same highmem pages get continually scanned. The idea would be that lowmem keeps those pages on a separate list until a reclaim for highmem pages arrives that splices the highmem pages back onto the LRU. It potentially could be implemented similar to the UNEVICTABLE list. That would reduce the skip rate with the potential corner case is that highmem pages have to be scanned and reclaimed to free lowmem slab pages. 2. Linear scan lowmem pages if the initial LRU shrink fails This will break LRU ordering but may be preferable and faster during memory pressure than skipping LRU pages. Link: http://lkml.kernel.org/r/1467970510-21195-4-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Rik van Riel <riel@surriel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-28 22:45:31 +00:00
/* In case we scheduled, work out time remaining */
ret = timeout - (jiffies - start);
if (ret < 0)
ret = 0;
goto out;
}
/* Sleep until uncongested or a write happens */
prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
ret = io_schedule_timeout(timeout);
finish_wait(wqh, &wait);
out:
trace_writeback_wait_iff_congested(jiffies_to_usecs(timeout),
jiffies_to_usecs(jiffies - start));
return ret;
}
EXPORT_SYMBOL(wait_iff_congested);