Merge branch 'for-4.2/writeback' of git://git.kernel.dk/linux-block

Pull cgroup writeback support from Jens Axboe:
"This is the big pull request for adding cgroup writeback support.

This code has been in development for a long time, and it has been
simmering in for-next for a good chunk of this cycle too. This is one
of those problems that has been talked about for at least half a
decade, finally there's a solution and code to go with it.

Also see last weeks writeup on LWN:

http://lwn.net/Articles/648292/"

* 'for-4.2/writeback' of git://git.kernel.dk/linux-block: (85 commits)
writeback, blkio: add documentation for cgroup writeback support
vfs, writeback: replace FS_CGROUP_WRITEBACK with SB_I_CGROUPWB
writeback: do foreign inode detection iff cgroup writeback is enabled
v9fs: fix error handling in v9fs_session_init()
bdi: fix wrong error return value in cgwb_create()
buffer: remove unusued 'ret' variable
writeback: disassociate inodes from dying bdi_writebacks
writeback: implement foreign cgroup inode bdi_writeback switching
writeback: add lockdep annotation to inode_to_wb()
writeback: use unlocked_inode_to_wb transaction in inode_congested()
writeback: implement unlocked_inode_to_wb transaction and use it for stat updates
writeback: implement [locked_]inode_to_wb_and_lock_list()
writeback: implement foreign cgroup inode detection
writeback: make writeback_control track the inode being written back
writeback: relocate wb[_try]_get(), wb_put(), inode_{attach|detach}_wb()
mm: vmscan: disable memcg direct reclaim stalling if cgroup writeback support is in use
writeback: implement memcg writeback domain based throttling
writeback: reset wb_domain->dirty_limit[_tstmp] when memcg domain size changes
writeback: implement memcg wb_domain
writeback: update wb_over_bg_thresh() to use wb_domain aware operations
...

Documentation/cgroups/blkio-controller.txt

···
 IO to keep disk busy. In that case set group_idle=0, and CFQ will not idle
 on individual groups and throughput should improve.
 
-What works
-==========
-- Currently only sync IO queues are support. All the buffered writes are
-  still system wide and not per group. Hence we will not see service
-  differentiation between buffered writes between groups.
+Writeback
+=========
+
+Page cache is dirtied through buffered writes and shared mmaps and
+written asynchronously to the backing filesystem by the writeback
+mechanism.  Writeback sits between the memory and IO domains and
+regulates the proportion of dirty memory by balancing dirtying and
+write IOs.
+
+On traditional cgroup hierarchies, relationships between different
+controllers cannot be established making it impossible for writeback
+to operate accounting for cgroup resource restrictions and all
+writeback IOs are attributed to the root cgroup.
+
+If both the blkio and memory controllers are used on the v2 hierarchy
+and the filesystem supports cgroup writeback, writeback operations
+correctly follow the resource restrictions imposed by both memory and
+blkio controllers.
+
+Writeback examines both system-wide and per-cgroup dirty memory status
+and enforces the more restrictive of the two.  Also, writeback control
+parameters which are absolute values - vm.dirty_bytes and
+vm.dirty_background_bytes - are distributed across cgroups according
+to their current writeback bandwidth.
+
+There's a peculiarity stemming from the discrepancy in ownership
+granularity between memory controller and writeback.  While memory
+controller tracks ownership per page, writeback operates on inode
+basis.  cgroup writeback bridges the gap by tracking ownership by
+inode but migrating ownership if too many foreign pages, pages which
+don't match the current inode ownership, have been encountered while
+writing back the inode.
+
+This is a conscious design choice as writeback operations are
+inherently tied to inodes making strictly following page ownership
+complicated and inefficient.  The only use case which suffers from
+this compromise is multiple cgroups concurrently dirtying disjoint
+regions of the same inode, which is an unlikely use case and decided
+to be unsupported.  Note that as memory controller assigns page
+ownership on the first use and doesn't update it until the page is
+released, even if cgroup writeback strictly follows page ownership,
+multiple cgroups dirtying overlapping areas wouldn't work as expected.
+In general, write-sharing an inode across multiple cgroups is not well
+supported.
+
+Filesystem support for cgroup writeback
+---------------------------------------
+
+A filesystem can make writeback IOs cgroup-aware by updating
+address_space_operations->writepage[s]() to annotate bio's using the
+following two functions.
+
+* wbc_init_bio(@wbc, @bio)
+
+  Should be called for each bio carrying writeback data and associates
+  the bio with the inode's owner cgroup.  Can be called anytime
+  between bio allocation and submission.
+
+* wbc_account_io(@wbc, @page, @bytes)
+
+  Should be called for each data segment being written out.  While
+  this function doesn't care exactly when it's called during the
+  writeback session, it's the easiest and most natural to call it as
+  data segments are added to a bio.
+
+With writeback bio's annotated, cgroup support can be enabled per
+super_block by setting MS_CGROUPWB in ->s_flags.  This allows for
+selective disabling of cgroup writeback support which is helpful when
+certain filesystem features, e.g. journaled data mode, are
+incompatible.
+
+wbc_init_bio() binds the specified bio to its cgroup.  Depending on
+the configuration, the bio may be executed at a lower priority and if
+the writeback session is holding shared resources, e.g. a journal
+entry, may lead to priority inversion.  There is no one easy solution
+for the problem.  Filesystems can try to work around specific problem
+cases by skipping wbc_init_bio() or using bio_associate_blkcg()
+directly.

Documentation/cgroups/memory.txt

···
 pgpgout		- # of uncharging events to the memory cgroup. The uncharging
 		event happens each time a page is unaccounted from the cgroup.
 swap		- # of bytes of swap usage
+dirty		- # of bytes that are waiting to get written back to the disk.
 writeback	- # of bytes of file/anon cache that are queued for syncing to
 		disk.
 inactive_anon	- # of bytes of anonymous and swap cache memory on inactive

block/bio.c

···
 EXPORT_SYMBOL(bioset_create_nobvec);
 
 #ifdef CONFIG_BLK_CGROUP
+
+/**
+ * bio_associate_blkcg - associate a bio with the specified blkcg
+ * @bio: target bio
+ * @blkcg_css: css of the blkcg to associate
+ *
+ * Associate @bio with the blkcg specified by @blkcg_css.  Block layer will
+ * treat @bio as if it were issued by a task which belongs to the blkcg.
+ *
+ * This function takes an extra reference of @blkcg_css which will be put
+ * when @bio is released.  The caller must own @bio and is responsible for
+ * synchronizing calls to this function.
+ */
+int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css)
+{
+	if (unlikely(bio->bi_css))
+		return -EBUSY;
+	css_get(blkcg_css);
+	bio->bi_css = blkcg_css;
+	return 0;
+}
+
 /**
  * bio_associate_current - associate a bio with %current
  * @bio: target bio
···
 int bio_associate_current(struct bio *bio)
 {
 	struct io_context *ioc;
-	struct cgroup_subsys_state *css;
 
-	if (bio->bi_ioc)
+	if (bio->bi_css)
 		return -EBUSY;
 
 	ioc = current->io_context;
 	if (!ioc)
 		return -ENOENT;
 
-	/* acquire active ref on @ioc and associate */
 	get_io_context_active(ioc);
 	bio->bi_ioc = ioc;
-
-	/* associate blkcg if exists */
-	rcu_read_lock();
-	css = task_css(current, blkio_cgrp_id);
-	if (css && css_tryget_online(css))
-		bio->bi_css = css;
-	rcu_read_unlock();
-
+	bio->bi_css = task_get_css(current, blkio_cgrp_id);
 	return 0;
 }
 

block/blk-cgroup.c

···
 #include <linux/module.h>
 #include <linux/err.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/slab.h>
 #include <linux/genhd.h>
 #include <linux/delay.h>
 #include <linux/atomic.h>
-#include "blk-cgroup.h"
+#include <linux/blk-cgroup.h>
 #include "blk.h"
 
 #define MAX_KEY_LEN 100
···
 
 struct blkcg blkcg_root;
 EXPORT_SYMBOL_GPL(blkcg_root);
+
+struct cgroup_subsys_state * const blkcg_root_css = &blkcg_root.css;
 
 static struct blkcg_policy *blkcg_policy[BLKCG_MAX_POLS];
 
···
 				    struct blkcg_gq *new_blkg)
 {
 	struct blkcg_gq *blkg;
+	struct bdi_writeback_congested *wb_congested;
 	int i, ret;
 
 	WARN_ON_ONCE(!rcu_read_lock_held());
···
 		goto err_free_blkg;
 	}
 
+	wb_congested = wb_congested_get_create(&q->backing_dev_info,
+					       blkcg->css.id, GFP_ATOMIC);
+	if (!wb_congested) {
+		ret = -ENOMEM;
+		goto err_put_css;
+	}
+
 	/* allocate */
 	if (!new_blkg) {
 		new_blkg = blkg_alloc(blkcg, q, GFP_ATOMIC);
 		if (unlikely(!new_blkg)) {
 			ret = -ENOMEM;
-			goto err_put_css;
+			goto err_put_congested;
 		}
 	}
 	blkg = new_blkg;
+	blkg->wb_congested = wb_congested;
 
 	/* link parent */
 	if (blkcg_parent(blkcg)) {
 		blkg->parent = __blkg_lookup(blkcg_parent(blkcg), q, false);
 		if (WARN_ON_ONCE(!blkg->parent)) {
 			ret = -EINVAL;
-			goto err_put_css;
+			goto err_put_congested;
 		}
 		blkg_get(blkg->parent);
 	}
···
 	blkg->online = true;
 	spin_unlock(&blkcg->lock);
 
-	if (!ret) {
-		if (blkcg == &blkcg_root) {
-			q->root_blkg = blkg;
-			q->root_rl.blkg = blkg;
-		}
+	if (!ret)
 		return blkg;
-	}
 
 	/* @blkg failed fully initialized, use the usual release path */
 	blkg_put(blkg);
 	return ERR_PTR(ret);
 
+err_put_congested:
+	wb_congested_put(wb_congested);
 err_put_css:
 	css_put(&blkcg->css);
 err_free_blkg:
···
 		rcu_assign_pointer(blkcg->blkg_hint, NULL);
 
 	/*
-	 * If root blkg is destroyed.  Just clear the pointer since root_rl
-	 * does not take reference on root blkg.
-	 */
-	if (blkcg == &blkcg_root) {
-		blkg->q->root_blkg = NULL;
-		blkg->q->root_rl.blkg = NULL;
-	}
-
-	/*
 	 * Put the reference taken at the time of creation so that when all
 	 * queues are gone, group can be destroyed.
 	 */
···
 	css_put(&blkg->blkcg->css);
 	if (blkg->parent)
 		blkg_put(blkg->parent);
+
+	wb_congested_put(blkg->wb_congested);
 
 	blkg_free(blkg);
 }
···
 	}
 
 	spin_unlock_irq(&blkcg->lock);
+
+	wb_blkcg_offline(blkcg);
 }
 
 static void blkcg_css_free(struct cgroup_subsys_state *css)
···
 	spin_lock_init(&blkcg->lock);
 	INIT_RADIX_TREE(&blkcg->blkg_tree, GFP_ATOMIC);
 	INIT_HLIST_HEAD(&blkcg->blkg_list);
-
+#ifdef CONFIG_CGROUP_WRITEBACK
+	INIT_LIST_HEAD(&blkcg->cgwb_list);
+#endif
 	return &blkcg->css;
 
 free_pd_blkcg:
···
  */
 int blkcg_init_queue(struct request_queue *q)
 {
-	might_sleep();
+	struct blkcg_gq *new_blkg, *blkg;
+	bool preloaded;
+	int ret;
 
-	return blk_throtl_init(q);
+	new_blkg = blkg_alloc(&blkcg_root, q, GFP_KERNEL);
+	if (!new_blkg)
+		return -ENOMEM;
+
+	preloaded = !radix_tree_preload(GFP_KERNEL);
+
+	/*
+	 * Make sure the root blkg exists and count the existing blkgs.  As
+	 * @q is bypassing at this point, blkg_lookup_create() can't be
+	 * used.  Open code insertion.
+	 */
+	rcu_read_lock();
+	spin_lock_irq(q->queue_lock);
+	blkg = blkg_create(&blkcg_root, q, new_blkg);
+	spin_unlock_irq(q->queue_lock);
+	rcu_read_unlock();
+
+	if (preloaded)
+		radix_tree_preload_end();
+
+	if (IS_ERR(blkg)) {
+		kfree(new_blkg);
+		return PTR_ERR(blkg);
+	}
+
+	q->root_blkg = blkg;
+	q->root_rl.blkg = blkg;
+
+	ret = blk_throtl_init(q);
+	if (ret) {
+		spin_lock_irq(q->queue_lock);
+		blkg_destroy_all(q);
+		spin_unlock_irq(q->queue_lock);
+	}
+	return ret;
 }
 
 /**
···
 {
 	LIST_HEAD(pds);
 	LIST_HEAD(cpds);
-	struct blkcg_gq *blkg, *new_blkg;
+	struct blkcg_gq *blkg;
 	struct blkg_policy_data *pd, *nd;
 	struct blkcg_policy_data *cpd, *cnd;
 	int cnt = 0, ret;
-	bool preloaded;
 
 	if (blkcg_policy_enabled(q, pol))
 		return 0;
 
-	/* preallocations for root blkg */
-	new_blkg = blkg_alloc(&blkcg_root, q, GFP_KERNEL);
-	if (!new_blkg)
-		return -ENOMEM;
-
+	/* count and allocate policy_data for all existing blkgs */
 	blk_queue_bypass_start(q);
-
-	preloaded = !radix_tree_preload(GFP_KERNEL);
-
-	/*
-	 * Make sure the root blkg exists and count the existing blkgs.  As
-	 * @q is bypassing at this point, blkg_lookup_create() can't be
-	 * used.  Open code it.
-	 */
 	spin_lock_irq(q->queue_lock);
-
-	rcu_read_lock();
-	blkg = __blkg_lookup(&blkcg_root, q, false);
-	if (blkg)
-		blkg_free(new_blkg);
-	else
-		blkg = blkg_create(&blkcg_root, q, new_blkg);
-	rcu_read_unlock();
-
-	if (preloaded)
-		radix_tree_preload_end();
-
-	if (IS_ERR(blkg)) {
-		ret = PTR_ERR(blkg);
-		goto out_unlock;
-	}
-
 	list_for_each_entry(blkg, &q->blkg_list, q_node)
 		cnt++;
-
 	spin_unlock_irq(q->queue_lock);
 
 	/*
···
 	spin_lock_irq(q->queue_lock);
 
 	__clear_bit(pol->plid, q->blkcg_pols);
-
-	/* if no policy is left, no need for blkgs - shoot them down */
-	if (bitmap_empty(q->blkcg_pols, BLKCG_MAX_POLS))
-		blkg_destroy_all(q);
 
 	list_for_each_entry(blkg, &q->blkg_list, q_node) {
 		/* grab blkcg lock too while removing @pd from @blkg */

block/blk-cgroup.h include/linux/blk-cgroup.h

···
 	struct hlist_head		blkg_list;
 
 	struct blkcg_policy_data	*pd[BLKCG_MAX_POLS];
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct list_head		cgwb_list;
+#endif
 };
 
 struct blkg_stat {
···
 	struct hlist_node		blkcg_node;
 	struct blkcg			*blkcg;
 
+	/*
+	 * Each blkg gets congested separately and the congestion state is
+	 * propagated to the matching bdi_writeback_congested.
+	 */
+	struct bdi_writeback_congested	*wb_congested;
+
 	/* all non-root blkcg_gq's are guaranteed to have access to parent */
 	struct blkcg_gq			*parent;
 
···
 };
 
 extern struct blkcg blkcg_root;
+extern struct cgroup_subsys_state * const blkcg_root_css;
 
 struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, struct request_queue *q);
 struct blkcg_gq *blkg_lookup_create(struct blkcg *blkcg,
···
 	if (bio && bio->bi_css)
 		return css_to_blkcg(bio->bi_css);
 	return task_blkcg(current);
+}
+
+static inline struct cgroup_subsys_state *
+task_get_blkcg_css(struct task_struct *task)
+{
+	return task_get_css(task, blkio_cgrp_id);
 }
 
 /**
···
 
 #else	/* CONFIG_BLK_CGROUP */
 
-struct cgroup;
-struct blkcg;
+struct blkcg {
+};
 
 struct blkg_policy_data {
 };
···
 
 struct blkcg_policy {
 };
+
+#define blkcg_root_css	((struct cgroup_subsys_state *)ERR_PTR(-EINVAL))
+
+static inline struct cgroup_subsys_state *
+task_get_blkcg_css(struct task_struct *task)
+{
+	return NULL;
+}
+
+#ifdef CONFIG_BLOCK
 
 static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, void *key) { return NULL; }
 static inline int blkcg_init_queue(struct request_queue *q) { return 0; }
···
 #define blk_queue_for_each_rl(rl, q)	\
 	for ((rl) = &(q)->root_rl; (rl); (rl) = NULL)
 
+#endif	/* CONFIG_BLOCK */
 #endif	/* CONFIG_BLK_CGROUP */
 #endif	/* _BLK_CGROUP_H */

block/blk-core.c

···
 #include <linux/delay.h>
 #include <linux/ratelimit.h>
 #include <linux/pm_runtime.h>
+#include <linux/blk-cgroup.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/block.h>
 
 #include "blk.h"
-#include "blk-cgroup.h"
 #include "blk-mq.h"
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap);
···
  * Controlling structure to kblockd
  */
 static struct workqueue_struct *kblockd_workqueue;
+
+static void blk_clear_congested(struct request_list *rl, int sync)
+{
+#ifdef CONFIG_CGROUP_WRITEBACK
+	clear_wb_congested(rl->blkg->wb_congested, sync);
+#else
+	/*
+	 * If !CGROUP_WRITEBACK, all blkg's map to bdi->wb and we shouldn't
+	 * flip its congestion state for events on other blkcgs.
+	 */
+	if (rl == &rl->q->root_rl)
+		clear_wb_congested(rl->q->backing_dev_info.wb.congested, sync);
+#endif
+}
+
+static void blk_set_congested(struct request_list *rl, int sync)
+{
+#ifdef CONFIG_CGROUP_WRITEBACK
+	set_wb_congested(rl->blkg->wb_congested, sync);
+#else
+	/* see blk_clear_congested() */
+	if (rl == &rl->q->root_rl)
+		set_wb_congested(rl->q->backing_dev_info.wb.congested, sync);
+#endif
+}
 
 void blk_queue_congestion_threshold(struct request_queue *q)
 {
···
 
 	q->backing_dev_info.ra_pages =
 			(VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
-	q->backing_dev_info.state = 0;
-	q->backing_dev_info.capabilities = 0;
+	q->backing_dev_info.capabilities = BDI_CAP_CGROUP_WRITEBACK;
 	q->backing_dev_info.name = "block";
 	q->node = node_id;
 
···
 {
 	struct request_queue *q = rl->q;
 
-	/*
-	 * bdi isn't aware of blkcg yet.  As all async IOs end up root
-	 * blkcg anyway, just use root blkcg state.
-	 */
-	if (rl == &q->root_rl &&
-	    rl->count[sync] < queue_congestion_off_threshold(q))
-		blk_clear_queue_congested(q, sync);
+	if (rl->count[sync] < queue_congestion_off_threshold(q))
+		blk_clear_congested(rl, sync);
 
 	if (rl->count[sync] + 1 <= q->nr_requests) {
 		if (waitqueue_active(&rl->wait[sync]))
···
 int blk_update_nr_requests(struct request_queue *q, unsigned int nr)
 {
 	struct request_list *rl;
+	int on_thresh, off_thresh;
 
 	spin_lock_irq(q->queue_lock);
 	q->nr_requests = nr;
 	blk_queue_congestion_threshold(q);
-
-	/* congestion isn't cgroup aware and follows root blkcg for now */
-	rl = &q->root_rl;
-
-	if (rl->count[BLK_RW_SYNC] >= queue_congestion_on_threshold(q))
-		blk_set_queue_congested(q, BLK_RW_SYNC);
-	else if (rl->count[BLK_RW_SYNC] < queue_congestion_off_threshold(q))
-		blk_clear_queue_congested(q, BLK_RW_SYNC);
-
-	if (rl->count[BLK_RW_ASYNC] >= queue_congestion_on_threshold(q))
-		blk_set_queue_congested(q, BLK_RW_ASYNC);
-	else if (rl->count[BLK_RW_ASYNC] < queue_congestion_off_threshold(q))
-		blk_clear_queue_congested(q, BLK_RW_ASYNC);
+	on_thresh = queue_congestion_on_threshold(q);
+	off_thresh = queue_congestion_off_threshold(q);
 
 	blk_queue_for_each_rl(rl, q) {
+		if (rl->count[BLK_RW_SYNC] >= on_thresh)
+			blk_set_congested(rl, BLK_RW_SYNC);
+		else if (rl->count[BLK_RW_SYNC] < off_thresh)
+			blk_clear_congested(rl, BLK_RW_SYNC);
+
+		if (rl->count[BLK_RW_ASYNC] >= on_thresh)
+			blk_set_congested(rl, BLK_RW_ASYNC);
+		else if (rl->count[BLK_RW_ASYNC] < off_thresh)
+			blk_clear_congested(rl, BLK_RW_ASYNC);
+
 		if (rl->count[BLK_RW_SYNC] >= q->nr_requests) {
 			blk_set_rl_full(rl, BLK_RW_SYNC);
 		} else {
···
 				}
 			}
 		}
-		/*
-		 * bdi isn't aware of blkcg yet.  As all async IOs end up
-		 * root blkcg anyway, just use root blkcg state.
-		 */
-		if (rl == &q->root_rl)
-			blk_set_queue_congested(q, is_sync);
+		blk_set_congested(rl, is_sync);
 	}
 
 	/*

block/blk-integrity.c

···
  */
 
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/mempool.h>
 #include <linux/bio.h>
 #include <linux/scatterlist.h>

block/blk-sysfs.c

···
 #include <linux/module.h>
 #include <linux/bio.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/blktrace_api.h>
 #include <linux/blk-mq.h>
+#include <linux/blk-cgroup.h>
 
 #include "blk.h"
-#include "blk-cgroup.h"
 #include "blk-mq.h"
 
 struct queue_sysfs_entry {

block/blk-throttle.c

···
 #include <linux/blkdev.h>
 #include <linux/bio.h>
 #include <linux/blktrace_api.h>
-#include "blk-cgroup.h"
+#include <linux/blk-cgroup.h>
 #include "blk.h"
 
 /* Max dispatch from a group in 1 round */

block/bounce.c

···
 #include <linux/pagemap.h>
 #include <linux/mempool.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/init.h>
 #include <linux/hash.h>
 #include <linux/highmem.h>

block/cfq-iosched.c

···
 #include <linux/rbtree.h>
 #include <linux/ioprio.h>
 #include <linux/blktrace_api.h>
+#include <linux/blk-cgroup.h>
 #include "blk.h"
-#include "blk-cgroup.h"
 
 /*
  * tunables

block/elevator.c

···
 #include <linux/hash.h>
 #include <linux/uaccess.h>
 #include <linux/pm_runtime.h>
+#include <linux/blk-cgroup.h>
 
 #include <trace/events/block.h>
 
 #include "blk.h"
-#include "blk-cgroup.h"
 
 static DEFINE_SPINLOCK(elv_list_lock);
 static LIST_HEAD(elv_list);

block/genhd.c

···
 #include <linux/kdev_t.h>
 #include <linux/kernel.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include <linux/proc_fs.h>

drivers/block/drbd/drbd_int.h

···
 #include <linux/mutex.h>
 #include <linux/major.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/genhd.h>
 #include <linux/idr.h>
 #include <net/tcp.h>

drivers/block/drbd/drbd_main.c

···
  * @congested_data:	User data
  * @bdi_bits:		Bits the BDI flusher thread is currently interested in
  *
- * Returns 1<<BDI_async_congested and/or 1<<BDI_sync_congested if we are congested.
+ * Returns 1<<WB_async_congested and/or 1<<WB_sync_congested if we are congested.
  */
 static int drbd_congested(void *congested_data, int bdi_bits)
 {
···
 	}
 
 	if (test_bit(CALLBACK_PENDING, &first_peer_device(device)->connection->flags)) {
-		r |= (1 << BDI_async_congested);
+		r |= (1 << WB_async_congested);
 		/* Without good local data, we would need to read from remote,
 		 * and that would need the worker thread as well, which is
 		 * currently blocked waiting for that usermode helper to
 		 * finish.
 		 */
 		if (!get_ldev_if_state(device, D_UP_TO_DATE))
-			r |= (1 << BDI_sync_congested);
+			r |= (1 << WB_sync_congested);
 		else
 			put_ldev(device);
 		r &= bdi_bits;
···
 			reason = 'b';
 	}
 
-	if (bdi_bits & (1 << BDI_async_congested) &&
+	if (bdi_bits & (1 << WB_async_congested) &&
 	    test_bit(NET_CONGESTED, &first_peer_device(device)->connection->flags)) {
-		r |= (1 << BDI_async_congested);
+		r |= (1 << WB_async_congested);
 		reason = reason == 'b' ? 'a' : 'n';
 	}
 

drivers/block/pktcdvd.c

···
 #include <linux/freezer.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
+#include <linux/backing-dev.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_ioctl.h>
 #include <scsi/scsi.h>

drivers/char/raw.c

···
 #include <linux/fs.h>
 #include <linux/major.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/module.h>
 #include <linux/raw.h>
 #include <linux/capability.h>

drivers/md/bcache/request.c

···
 #include <linux/module.h>
 #include <linux/hash.h>
 #include <linux/random.h>
+#include <linux/backing-dev.h>
 
 #include <trace/events/bcache.h>
 

drivers/md/dm.c

···
 			 * the query about congestion status of request_queue
 			 */
 			if (dm_request_based(md))
-				r = md->queue->backing_dev_info.state &
+				r = md->queue->backing_dev_info.wb.state &
 				    bdi_bits;
 			else
 				r = dm_table_any_congested(map, bdi_bits);

drivers/md/dm.h

···
 #include <linux/device-mapper.h>
 #include <linux/list.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/hdreg.h>
 #include <linux/completion.h>
 #include <linux/kobject.h>

drivers/md/md.h

···
 #define _MD_MD_H
 
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/kobject.h>
 #include <linux/list.h>
 #include <linux/mm.h>

drivers/md/raid1.c

···
 	struct r1conf *conf = mddev->private;
 	int i, ret = 0;
 
-	if ((bits & (1 << BDI_async_congested)) &&
+	if ((bits & (1 << WB_async_congested)) &&
 	    conf->pending_count >= max_queued_requests)
 		return 1;
 
···
 			/* Note the '|| 1' - when read_balance prefers
 			 * non-congested targets, it can be removed
 			 */
-			if ((bits & (1<<BDI_async_congested)) || 1)
+			if ((bits & (1 << WB_async_congested)) || 1)
 				ret |= bdi_congested(&q->backing_dev_info, bits);
 			else
 				ret &= bdi_congested(&q->backing_dev_info, bits);

drivers/md/raid10.c

···
 	struct r10conf *conf = mddev->private;
 	int i, ret = 0;
 
-	if ((bits & (1 << BDI_async_congested)) &&
+	if ((bits & (1 << WB_async_congested)) &&
 	    conf->pending_count >= max_queued_requests)
 		return 1;
 

drivers/mtd/devices/block2mtd.c

···
 #include <linux/delay.h>
 #include <linux/fs.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/bio.h>
 #include <linux/pagemap.h>
 #include <linux/list.h>

drivers/staging/lustre/lustre/include/linux/lustre_patchless_compat.h

···
 	if (PagePrivate(page))
 		page->mapping->a_ops->invalidatepage(page, 0, PAGE_CACHE_SIZE);
 
-	if (TestClearPageDirty(page))
-		account_page_cleaned(page, mapping);
-
+	cancel_dirty_page(page);
 	ClearPageMappedToDisk(page);
 	ll_delete_from_page_cache(page);
 }

fs/9p/v9fs.c

···
 struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
 		  const char *dev_name, char *data)
 {
-	int retval = -EINVAL;
 	struct p9_fid *fid;
-	int rc;
+	int rc = -ENOMEM;
 
 	v9ses->uname = kstrdup(V9FS_DEFUSER, GFP_KERNEL);
 	if (!v9ses->uname)
-		return ERR_PTR(-ENOMEM);
+		goto err_names;
 
 	v9ses->aname = kstrdup(V9FS_DEFANAME, GFP_KERNEL);
-	if (!v9ses->aname) {
-		kfree(v9ses->uname);
-		return ERR_PTR(-ENOMEM);
-	}
+	if (!v9ses->aname)
+		goto err_names;
 	init_rwsem(&v9ses->rename_sem);
 
 	rc = bdi_setup_and_register(&v9ses->bdi, "9p");
-	if (rc) {
-		kfree(v9ses->aname);
-		kfree(v9ses->uname);
-		return ERR_PTR(rc);
-	}
-
-	spin_lock(&v9fs_sessionlist_lock);
-	list_add(&v9ses->slist, &v9fs_sessionlist);
-	spin_unlock(&v9fs_sessionlist_lock);
+	if (rc)
+		goto err_names;
 
 	v9ses->uid = INVALID_UID;
 	v9ses->dfltuid = V9FS_DEFUID;
···
 
 	v9ses->clnt = p9_client_create(dev_name, data);
 	if (IS_ERR(v9ses->clnt)) {
-		retval = PTR_ERR(v9ses->clnt);
-		v9ses->clnt = NULL;
+		rc = PTR_ERR(v9ses->clnt);
 		p9_debug(P9_DEBUG_ERROR, "problem initializing 9p client\n");
-		goto error;
+		goto err_bdi;
 	}
 
 	v9ses->flags = V9FS_ACCESS_USER;
···
 	}
 
 	rc = v9fs_parse_options(v9ses, data);
-	if (rc < 0) {
-		retval = rc;
-		goto error;
-	}
+	if (rc < 0)
+		goto err_clnt;
 
 	v9ses->maxdata = v9ses->clnt->msize - P9_IOHDRSZ;
 
···
 	fid = p9_client_attach(v9ses->clnt, NULL, v9ses->uname, INVALID_UID,
 							v9ses->aname);
 	if (IS_ERR(fid)) {
-		retval = PTR_ERR(fid);
-		fid = NULL;
+		rc = PTR_ERR(fid);
 		p9_debug(P9_DEBUG_ERROR, "cannot attach\n");
-		goto error;
+		goto err_clnt;
 	}
 
 	if ((v9ses->flags & V9FS_ACCESS_MASK) == V9FS_ACCESS_SINGLE)
···
 	/* register the session for caching */
 	v9fs_cache_session_get_cookie(v9ses);
 #endif
+	spin_lock(&v9fs_sessionlist_lock);
+	list_add(&v9ses->slist, &v9fs_sessionlist);
+	spin_unlock(&v9fs_sessionlist_lock);
 
 	return fid;
 
-error:
+err_clnt:
+	p9_client_destroy(v9ses->clnt);
+err_bdi:
 	bdi_destroy(&v9ses->bdi);
-	return ERR_PTR(retval);
+err_names:
+	kfree(v9ses->uname);
+	kfree(v9ses->aname);
+	return ERR_PTR(rc);
 }
 
 /**

fs/9p/vfs_super.c

···
 	fid = v9fs_session_init(v9ses, dev_name, data);
 	if (IS_ERR(fid)) {
 		retval = PTR_ERR(fid);
-		/*
-		 * we need to call session_close to tear down some
-		 * of the data structure setup by session_init
-		 */
-		goto close_session;
+		goto free_session;
 	}
 
 	sb = sget(fs_type, NULL, v9fs_set_super, flags, v9ses);
···
 
 clunk_fid:
 	p9_client_clunk(fid);
-close_session:
 	v9fs_session_close(v9ses);
+free_session:
 	kfree(v9ses);
 	return ERR_PTR(retval);
 

fs/block_dev.c

···
 #include <linux/device_cgroup.h>
 #include <linux/highmem.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/module.h>
 #include <linux/blkpg.h>
 #include <linux/magic.h>
···
 	.kill_sb	= kill_anon_super,
 };
 
-static struct super_block *blockdev_superblock __read_mostly;
+struct super_block *blockdev_superblock __read_mostly;
+EXPORT_SYMBOL_GPL(blockdev_superblock);
 
 void __init bdev_cache_init(void)
 {
···
 		spin_unlock(&bdev_lock);
 	}
 	return bdev;
-}
-
-int sb_is_blkdev_sb(struct super_block *sb)
-{
-	return sb == blockdev_superblock;
 }
 
 /* Call when you free inode */

fs/buffer.c

···
 #include <linux/quotaops.h>
 #include <linux/highmem.h>
 #include <linux/export.h>
+#include <linux/backing-dev.h>
 #include <linux/writeback.h>
 #include <linux/hash.h>
 #include <linux/suspend.h>
···
 #include <trace/events/block.h>
 
 static int fsync_buffers_list(spinlock_t *lock, struct list_head *list);
+static int submit_bh_wbc(int rw, struct buffer_head *bh,
+			 unsigned long bio_flags,
+			 struct writeback_control *wbc);
 
 #define BH_ENTRY(list) list_entry((list), struct buffer_head, b_assoc_buffers)
 
···
  *
  * If warn is true, then emit a warning if the page is not uptodate and has
  * not been truncated.
+ *
+ * The caller must hold mem_cgroup_begin_page_stat() lock.
  */
-static void __set_page_dirty(struct page *page,
-		struct address_space *mapping, int warn)
+static void __set_page_dirty(struct page *page, struct address_space *mapping,
+			     struct mem_cgroup *memcg, int warn)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&mapping->tree_lock, flags);
 	if (page->mapping) {	/* Race with truncate? */
 		WARN_ON_ONCE(warn && !PageUptodate(page));
-		account_page_dirtied(page, mapping);
+		account_page_dirtied(page, mapping, memcg);
 		radix_tree_tag_set(&mapping->page_tree,
 				page_index(page), PAGECACHE_TAG_DIRTY);
 	}
 	spin_unlock_irqrestore(&mapping->tree_lock, flags);
-	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 }
 
 /*
···
 int __set_page_dirty_buffers(struct page *page)
 {
 	int newly_dirty;
+	struct mem_cgroup *memcg;
 	struct address_space *mapping = page_mapping(page);
 
 	if (unlikely(!mapping))
···
 			bh = bh->b_this_page;
 		} while (bh != head);
 	}
+	/*
+	 * Use mem_group_begin_page_stat() to keep PageDirty synchronized with
+	 * per-memcg dirty page counters.
+	 */
+	memcg = mem_cgroup_begin_page_stat(page);
 	newly_dirty = !TestSetPageDirty(page);
 	spin_unlock(&mapping->private_lock);
 
 	if (newly_dirty)
-		__set_page_dirty(page, mapping, 1);
+		__set_page_dirty(page, mapping, memcg, 1);
+
+	mem_cgroup_end_page_stat(memcg);
+
+	if (newly_dirty)
+		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+
 	return newly_dirty;
 }
 EXPORT_SYMBOL(__set_page_dirty_buffers);
···
 
 	if (!test_set_buffer_dirty(bh)) {
 		struct page *page = bh->b_page;
+		struct address_space *mapping = NULL;
+		struct mem_cgroup *memcg;
+
+		memcg = mem_cgroup_begin_page_stat(page);
 		if (!TestSetPageDirty(page)) {
-			struct address_space *mapping = page_mapping(page);
+			mapping = page_mapping(page);
 			if (mapping)
-				__set_page_dirty(page, mapping, 0);
+				__set_page_dirty(page, mapping, memcg, 0);
 		}
+		mem_cgroup_end_page_stat(memcg);
+		if (mapping)
+			__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 	}
 }
 EXPORT_SYMBOL(mark_buffer_dirty);
···
 	struct buffer_head *bh, *head;
 	unsigned int blocksize, bbits;
 	int nr_underway = 0;
-	int write_op = (wbc->sync_mode == WB_SYNC_ALL ?
-			WRITE_SYNC : WRITE);
+	int write_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE);
 
 	head = create_page_buffers(page, inode,
 					(1 << BH_Dirty)|(1 << BH_Uptodate));
···
 	do {
 		struct buffer_head *next = bh->b_this_page;
 		if (buffer_async_write(bh)) {
-			submit_bh(write_op, bh);
+			submit_bh_wbc(write_op, bh, 0, wbc);
 			nr_underway++;
 		}
 		bh = next;
···
 		struct buffer_head *next = bh->b_this_page;
 		if (buffer_async_write(bh)) {
 			clear_buffer_dirty(bh);
-			submit_bh(write_op, bh);
+			submit_bh_wbc(write_op, bh, 0, wbc);
 			nr_underway++;
 		}
 		bh = next;
···
 	}
 }
 
-int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags)
+static int submit_bh_wbc(int rw, struct buffer_head *bh,
+			 unsigned long bio_flags, struct writeback_control *wbc)
 {
 	struct bio *bio;
 
···
 	 * submit_bio -> generic_make_request may further map this bio around
 	 */
 	bio = bio_alloc(GFP_NOIO, 1);
+
+	if (wbc) {
+		wbc_init_bio(wbc, bio);
+		wbc_account_io(wbc, bh->b_page, bh->b_size);
+	}
 
 	bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
 	bio->bi_bdev = bh->b_bdev;
···
 	submit_bio(rw, bio);
 	return 0;
 }
+
+int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags)
+{
+	return submit_bh_wbc(rw, bh, bio_flags, NULL);
+}
 EXPORT_SYMBOL_GPL(_submit_bh);
 
 int submit_bh(int rw, struct buffer_head *bh)
 {
-	return _submit_bh(rw, bh, 0);
+	return submit_bh_wbc(rw, bh, 0, NULL);
 }
 EXPORT_SYMBOL(submit_bh);
 
···
 	 * to synchronise against __set_page_dirty_buffers and prevent the
 	 * dirty bit from being lost.
 	 */
-	if (ret && TestClearPageDirty(page))
-		account_page_cleaned(page, mapping);
+	if (ret)
+		cancel_dirty_page(page);
 	spin_unlock(&mapping->private_lock);
 out:
 	if (buffers_to_free) {

fs/ext2/super.c

···
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
 		((EXT2_SB(sb)->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ?
 		 MS_POSIXACL : 0);
+	sb->s_iflags |= SB_I_CGROUPWB;
 
 	if (le32_to_cpu(es->s_rev_level) == EXT2_GOOD_OLD_REV &&
 	    (EXT2_HAS_COMPAT_FEATURE(sb, ~0U) ||

fs/ext4/extents.c

···
 #include <linux/slab.h>
 #include <asm/uaccess.h>
 #include <linux/fiemap.h>
+#include <linux/backing-dev.h>
 #include "ext4_jbd2.h"
 #include "ext4_extents.h"
 #include "xattr.h"

fs/ext4/mballoc.c

···
 #include <linux/log2.h>
 #include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/backing-dev.h>
 #include <trace/events/ext4.h>
 
 #ifdef CONFIG_EXT4_DEBUG

fs/ext4/super.c

···
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/parser.h>
 #include <linux/buffer_head.h>
 #include <linux/exportfs.h>

fs/f2fs/node.c

···
 							PAGE_CACHE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2);
 	} else if (type == DIRTY_DENTS) {
-		if (sbi->sb->s_bdi->dirty_exceeded)
+		if (sbi->sb->s_bdi->wb.dirty_exceeded)
 			return false;
 		mem_size = get_pages(sbi, F2FS_DIRTY_DENTS);
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
···
 				sizeof(struct extent_node)) >> PAGE_CACHE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
 	} else {
-		if (sbi->sb->s_bdi->dirty_exceeded)
+		if (sbi->sb->s_bdi->wb.dirty_exceeded)
 			return false;
 	}
 	return res;

fs/f2fs/segment.h

···
  * published by the Free Software Foundation.
  */
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 
 /* constant macro */
 #define NULL_SEGNO			((unsigned int)(~0))
···
  */
 static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type)
 {
-	if (sbi->sb->s_bdi->dirty_exceeded)
+	if (sbi->sb->s_bdi->wb.dirty_exceeded)
 		return 0;
 
 	if (type == DATA)

fs/fat/file.c

···
 #include <linux/compat.h>
 #include <linux/mount.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/fsnotify.h>
 #include <linux/security.h>
 #include "fat.h"

fs/fat/inode.c

···
 #include <linux/parser.h>
 #include <linux/uio.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <asm/unaligned.h>
 #include "fat.h"
 

fs/fs-writeback.c

···
 #include <linux/backing-dev.h>
 #include <linux/tracepoint.h>
 #include <linux/device.h>
+#include <linux/memcontrol.h>
 #include "internal.h"
 
 /*
  * 4MB minimal write chunk size
  */
 #define MIN_WRITEBACK_PAGES	(4096UL >> (PAGE_CACHE_SHIFT - 10))
+
+struct wb_completion {
+	atomic_t		cnt;
+};
 
 /*
  * Passed into wb_writeback(), essentially a subset of writeback_control
···
 	unsigned int range_cyclic:1;
 	unsigned int for_background:1;
 	unsigned int for_sync:1;	/* sync(2) WB_SYNC_ALL writeback */
+	unsigned int auto_free:1;	/* free on completion */
+	unsigned int single_wait:1;
+	unsigned int single_done:1;
 	enum wb_reason reason;		/* why was writeback initiated? */
 
 	struct list_head list;		/* pending work list */
-	struct completion *done;	/* set if the caller waits */
+	struct wb_completion *done;	/* set if the caller waits */
 };
+
+/*
+ * If one wants to wait for one or more wb_writeback_works, each work's
+ * ->done should be set to a wb_completion defined using the following
+ * macro.  Once all work items are issued with wb_queue_work(), the caller
+ * can wait for the completion of all using wb_wait_for_completion().  Work
+ * items which are waited upon aren't freed automatically on completion.
+ */
+#define DEFINE_WB_COMPLETION_ONSTACK(cmpl)				\
+	struct wb_completion cmpl = {					\
+		.cnt		= ATOMIC_INIT(1),			\
+	}
+
 
 /*
  * If an inode is constantly having its pages dirtied, but then the
···
  * few inodes might not their timestamps updated for 24 hours.
  */
 unsigned int dirtytime_expire_interval = 12 * 60 * 60;
-
-/**
- * writeback_in_progress - determine whether there is writeback in progress
- * @bdi: the device's backing_dev_info structure.
- *
- * Determine whether there is writeback waiting to be handled against a
- * backing device.
- */
-int writeback_in_progress(struct backing_dev_info *bdi)
-{
-	return test_bit(BDI_writeback_running, &bdi->state);
-}
-EXPORT_SYMBOL(writeback_in_progress);
-
-struct backing_dev_info *inode_to_bdi(struct inode *inode)
-{
-	struct super_block *sb;
-
-	if (!inode)
-		return &noop_backing_dev_info;
-
-	sb = inode->i_sb;
-#ifdef CONFIG_BLOCK
-	if (sb_is_blkdev_sb(sb))
-		return blk_get_backing_dev_info(I_BDEV(inode));
-#endif
-	return sb->s_bdi;
-}
-EXPORT_SYMBOL_GPL(inode_to_bdi);
 
 static inline struct inode *wb_inode(struct list_head *head)
 {
···
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(wbc_writepage);
 
-static void bdi_wakeup_thread(struct backing_dev_info *bdi)
+static bool wb_io_lists_populated(struct bdi_writeback *wb)
 {
-	spin_lock_bh(&bdi->wb_lock);
-	if (test_bit(BDI_registered, &bdi->state))
-		mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
-	spin_unlock_bh(&bdi->wb_lock);
+	if (wb_has_dirty_io(wb)) {
+		return false;
+	} else {
+		set_bit(WB_has_dirty_io, &wb->state);
+		WARN_ON_ONCE(!wb->avg_write_bandwidth);
+		atomic_long_add(wb->avg_write_bandwidth,
+				&wb->bdi->tot_write_bandwidth);
+		return true;
+	}
 }
 
-static void bdi_queue_work(struct backing_dev_info *bdi,
-			   struct wb_writeback_work *work)
+static void wb_io_lists_depopulated(struct bdi_writeback *wb)
 {
-	trace_writeback_queue(bdi, work);
+	if (wb_has_dirty_io(wb) && list_empty(&wb->b_dirty) &&
+	    list_empty(&wb->b_io) && list_empty(&wb->b_more_io)) {
+		clear_bit(WB_has_dirty_io, &wb->state);
+		WARN_ON_ONCE(atomic_long_sub_return(wb->avg_write_bandwidth,
+					&wb->bdi->tot_write_bandwidth) < 0);
+	}
+}
 
-	spin_lock_bh(&bdi->wb_lock);
-	if (!test_bit(BDI_registered, &bdi->state)) {
-		if (work->done)
-			complete(work->done);
+/**
+ * inode_wb_list_move_locked - move an inode onto a bdi_writeback IO list
+ * @inode: inode to be moved
+ * @wb: target bdi_writeback
+ * @head: one of @wb->b_{dirty|io|more_io}
+ *
+ * Move @inode->i_wb_list to @list of @wb and set %WB_has_dirty_io.
+ * Returns %true if @inode is the first occupant of the !dirty_time IO
+ * lists; otherwise, %false.
+ */
+static bool inode_wb_list_move_locked(struct inode *inode,
+				      struct bdi_writeback *wb,
+				      struct list_head *head)
+{
+	assert_spin_locked(&wb->list_lock);
+
+	list_move(&inode->i_wb_list, head);
+
+	/* dirty_time doesn't count as dirty_io until expiration */
+	if (head != &wb->b_dirty_time)
+		return wb_io_lists_populated(wb);
+
+	wb_io_lists_depopulated(wb);
+	return false;
+}
+
+/**
+ * inode_wb_list_del_locked - remove an inode from its bdi_writeback IO list
+ * @inode: inode to be removed
+ * @wb: bdi_writeback @inode is being removed from
+ *
+ * Remove @inode which may be on one of @wb->b_{dirty|io|more_io} lists and
+ * clear %WB_has_dirty_io if all are empty afterwards.
+ */
+static void inode_wb_list_del_locked(struct inode *inode,
+				     struct bdi_writeback *wb)
+{
+	assert_spin_locked(&wb->list_lock);
+
+	list_del_init(&inode->i_wb_list);
+	wb_io_lists_depopulated(wb);
+}
+
+static void wb_wakeup(struct bdi_writeback *wb)
+{
+	spin_lock_bh(&wb->work_lock);
+	if (test_bit(WB_registered, &wb->state))
+		mod_delayed_work(bdi_wq, &wb->dwork, 0);
+	spin_unlock_bh(&wb->work_lock);
+}
+
+static void wb_queue_work(struct bdi_writeback *wb,
+			  struct wb_writeback_work *work)
+{
+	trace_writeback_queue(wb->bdi, work);
+
+	spin_lock_bh(&wb->work_lock);
+	if (!test_bit(WB_registered, &wb->state)) {
+		if (work->single_wait)
+			work->single_done = 1;
 		goto out_unlock;
 	}
-	list_add_tail(&work->list, &bdi->work_list);
-	mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
+	if (work->done)
+		atomic_inc(&work->done->cnt);
+	list_add_tail(&work->list, &wb->work_list);
+	mod_delayed_work(bdi_wq, &wb->dwork, 0);
 out_unlock:
-	spin_unlock_bh(&bdi->wb_lock);
+	spin_unlock_bh(&wb->work_lock);
 }
 
-static void
-__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
-		      bool range_cyclic, enum wb_reason reason)
+/**
+ * wb_wait_for_completion - wait for completion of bdi_writeback_works
+ * @bdi: bdi work items were issued to
+ * @done: target wb_completion
+ *
+ * Wait for one or more work items issued to @bdi with their ->done field
+ * set to @done, which should have been defined with
+ * DEFINE_WB_COMPLETION_ONSTACK().  This function returns after all such
+ * work items are completed.  Work items which are waited upon aren't freed
+ * automatically on completion.
+ */
+static void wb_wait_for_completion(struct backing_dev_info *bdi,
+				   struct wb_completion *done)
+{
+	atomic_dec(&done->cnt);		/* put down the initial count */
+	wait_event(bdi->wb_waitq, !atomic_read(&done->cnt));
+}
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+/* parameters for foreign inode detection, see wb_detach_inode() */
+#define WB_FRN_TIME_SHIFT	13	/* 1s = 2^13, upto 8 secs w/ 16bit */
+#define WB_FRN_TIME_AVG_SHIFT	3	/* avg = avg * 7/8 + new * 1/8 */
+#define WB_FRN_TIME_CUT_DIV	2	/* ignore rounds < avg / 2 */
+#define WB_FRN_TIME_PERIOD	(2 * (1 << WB_FRN_TIME_SHIFT))	/* 2s */
+
+#define WB_FRN_HIST_SLOTS	16	/* inode->i_wb_frn_history is 16bit */
+#define WB_FRN_HIST_UNIT	(WB_FRN_TIME_PERIOD / WB_FRN_HIST_SLOTS)
+					/* each slot's duration is 2s / 16 */
+#define WB_FRN_HIST_THR_SLOTS	(WB_FRN_HIST_SLOTS / 2)
+					/* if foreign slots >= 8, switch */
+#define WB_FRN_HIST_MAX_SLOTS	(WB_FRN_HIST_THR_SLOTS / 2 + 1)
+					/* one round can affect upto 5 slots */
+
+void __inode_attach_wb(struct inode *inode, struct page *page)
+{
+	struct backing_dev_info *bdi = inode_to_bdi(inode);
+	struct bdi_writeback *wb = NULL;
+
+	if (inode_cgwb_enabled(inode)) {
+		struct cgroup_subsys_state *memcg_css;
+
+		if (page) {
+			memcg_css = mem_cgroup_css_from_page(page);
+			wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC);
+		} else {
+			/* must pin memcg_css, see wb_get_create() */
+			memcg_css = task_get_css(current, memory_cgrp_id);
+			wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC);
+			css_put(memcg_css);
+		}
+	}
+
+	if (!wb)
+		wb = &bdi->wb;
+
+	/*
+	 * There may be multiple instances of this function racing to
+	 * update the same inode.  Use cmpxchg() to tell the winner.
+	 */
+	if (unlikely(cmpxchg(&inode->i_wb, NULL, wb)))
+		wb_put(wb);
+}
+
+/**
+ * locked_inode_to_wb_and_lock_list - determine a locked inode's wb and lock it
+ * @inode: inode of interest with i_lock held
+ *
+ * Returns @inode's wb with its list_lock held.  @inode->i_lock must be
+ * held on entry and is released on return.  The returned wb is guaranteed
+ * to stay @inode's associated wb until its list_lock is released.
+ */
+static struct bdi_writeback *
+locked_inode_to_wb_and_lock_list(struct inode *inode)
+	__releases(&inode->i_lock)
+	__acquires(&wb->list_lock)
+{
+	while (true) {
+		struct bdi_writeback *wb = inode_to_wb(inode);
+
+		/*
+		 * inode_to_wb() association is protected by both
+		 * @inode->i_lock and @wb->list_lock but list_lock nests
+		 * outside i_lock.  Drop i_lock and verify that the
+		 * association hasn't changed after acquiring list_lock.
+		 */
+		wb_get(wb);
+		spin_unlock(&inode->i_lock);
+		spin_lock(&wb->list_lock);
+		wb_put(wb);		/* not gonna deref it anymore */
+
+		/* i_wb may have changed inbetween, can't use inode_to_wb() */
+		if (likely(wb == inode->i_wb))
+			return wb;	/* @inode already has ref */
+
+		spin_unlock(&wb->list_lock);
+		cpu_relax();
+		spin_lock(&inode->i_lock);
+	}
+}
+
+/**
+ * inode_to_wb_and_lock_list - determine an inode's wb and lock it
+ * @inode: inode of interest
+ *
+ * Same as locked_inode_to_wb_and_lock_list() but @inode->i_lock isn't held
+ * on entry.
+ */
+static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode)
+	__acquires(&wb->list_lock)
+{
+	spin_lock(&inode->i_lock);
+	return locked_inode_to_wb_and_lock_list(inode);
+}
+
+struct inode_switch_wbs_context {
+	struct inode		*inode;
+	struct bdi_writeback	*new_wb;
+
+	struct rcu_head		rcu_head;
+	struct work_struct	work;
+};
+
+static void inode_switch_wbs_work_fn(struct work_struct *work)
+{
+	struct inode_switch_wbs_context *isw =
+		container_of(work, struct inode_switch_wbs_context, work);
+	struct inode *inode = isw->inode;
+	struct address_space *mapping = inode->i_mapping;
+	struct bdi_writeback *old_wb = inode->i_wb;
+	struct bdi_writeback *new_wb = isw->new_wb;
+	struct radix_tree_iter iter;
+	bool switched = false;
+	void **slot;
+
+	/*
+	 * By the time control reaches here, RCU grace period has passed
+	 * since I_WB_SWITCH assertion and all wb stat update transactions
+	 * between unlocked_inode_to_wb_begin/end() are guaranteed to be
+	 * synchronizing against mapping->tree_lock.
+	 *
+	 * Grabbing old_wb->list_lock, inode->i_lock and mapping->tree_lock
+	 * gives us exclusion against all wb related operations on @inode
+	 * including IO list manipulations and stat updates.
+	 */
+	if (old_wb < new_wb) {
+		spin_lock(&old_wb->list_lock);
+		spin_lock_nested(&new_wb->list_lock, SINGLE_DEPTH_NESTING);
+	} else {
+		spin_lock(&new_wb->list_lock);
+		spin_lock_nested(&old_wb->list_lock, SINGLE_DEPTH_NESTING);
+	}
+	spin_lock(&inode->i_lock);
+	spin_lock_irq(&mapping->tree_lock);
+
+	/*
+	 * Once I_FREEING is visible under i_lock, the eviction path owns
+	 * the inode and we shouldn't modify ->i_wb_list.
+	 */
+	if (unlikely(inode->i_state & I_FREEING))
+		goto skip_switch;
+
+	/*
+	 * Count and transfer stats.  Note that PAGECACHE_TAG_DIRTY points
+	 * to possibly dirty pages while PAGECACHE_TAG_WRITEBACK points to
+	 * pages actually under underwriteback.
+	 */
+	radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, 0,
+				   PAGECACHE_TAG_DIRTY) {
+		struct page *page = radix_tree_deref_slot_protected(slot,
+							&mapping->tree_lock);
+		if (likely(page) && PageDirty(page)) {
+			__dec_wb_stat(old_wb, WB_RECLAIMABLE);
+			__inc_wb_stat(new_wb, WB_RECLAIMABLE);
+		}
+	}
+
+	radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, 0,
+				   PAGECACHE_TAG_WRITEBACK) {
+		struct page *page = radix_tree_deref_slot_protected(slot,
+							&mapping->tree_lock);
+		if (likely(page)) {
+			WARN_ON_ONCE(!PageWriteback(page));
+			__dec_wb_stat(old_wb, WB_WRITEBACK);
+			__inc_wb_stat(new_wb, WB_WRITEBACK);
+		}
+	}
+
+	wb_get(new_wb);
+
+	/*
+	 * Transfer to @new_wb's IO list if necessary.  The specific list
+	 * @inode was on is ignored and the inode is put on ->b_dirty which
+	 * is always correct including from ->b_dirty_time.  The transfer
+	 * preserves @inode->dirtied_when ordering.
+	 */
+	if (!list_empty(&inode->i_wb_list)) {
+		struct inode *pos;
+
+		inode_wb_list_del_locked(inode, old_wb);
+		inode->i_wb = new_wb;
+		list_for_each_entry(pos, &new_wb->b_dirty, i_wb_list)
+			if (time_after_eq(inode->dirtied_when,
+					  pos->dirtied_when))
+				break;
+		inode_wb_list_move_locked(inode, new_wb, pos->i_wb_list.prev);
+	} else {
+		inode->i_wb = new_wb;
+	}
+
+	/* ->i_wb_frn updates may race wbc_detach_inode() but doesn't matter */
+	inode->i_wb_frn_winner = 0;
+	inode->i_wb_frn_avg_time = 0;
+	inode->i_wb_frn_history = 0;
+	switched = true;
+skip_switch:
+	/*
+	 * Paired with load_acquire in unlocked_inode_to_wb_begin() and
+	 * ensures that the new wb is visible if they see !I_WB_SWITCH.
+	 */
+	smp_store_release(&inode->i_state, inode->i_state & ~I_WB_SWITCH);
+
+	spin_unlock_irq(&mapping->tree_lock);
+	spin_unlock(&inode->i_lock);
+	spin_unlock(&new_wb->list_lock);
+	spin_unlock(&old_wb->list_lock);
+
+	if (switched) {
+		wb_wakeup(new_wb);
+		wb_put(old_wb);
+	}
+	wb_put(new_wb);
+
+	iput(inode);
+	kfree(isw);
+}
+
+static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head)
+{
+	struct inode_switch_wbs_context *isw = container_of(rcu_head,
+				struct inode_switch_wbs_context, rcu_head);
+
+	/* needs to grab bh-unsafe locks, bounce to work item */
+	INIT_WORK(&isw->work, inode_switch_wbs_work_fn);
+	schedule_work(&isw->work);
+}
+
+/**
+ * inode_switch_wbs - change the wb association of an inode
+ * @inode: target inode
+ * @new_wb_id: ID of the new wb
+ *
+ * Switch @inode's wb association to the wb identified by @new_wb_id.  The
+ * switching is performed asynchronously and may fail silently.
+ */
+static void inode_switch_wbs(struct inode *inode, int new_wb_id)
+{
+	struct backing_dev_info *bdi = inode_to_bdi(inode);
+	struct cgroup_subsys_state *memcg_css;
+	struct inode_switch_wbs_context *isw;
+
+	/* noop if seems to be already in progress */
+	if (inode->i_state & I_WB_SWITCH)
+		return;
+
+	isw = kzalloc(sizeof(*isw), GFP_ATOMIC);
+	if (!isw)
+		return;
+
+	/* find and pin the new wb */
+	rcu_read_lock();
+	memcg_css = css_from_id(new_wb_id, &memory_cgrp_subsys);
+	if (memcg_css)
+		isw->new_wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC);
+	rcu_read_unlock();
+	if (!isw->new_wb)
+		goto out_free;
+
+	/* while holding I_WB_SWITCH, no one else can update the association */
+	spin_lock(&inode->i_lock);
+	if (inode->i_state & (I_WB_SWITCH | I_FREEING) ||
+	    inode_to_wb(inode) == isw->new_wb) {
+		spin_unlock(&inode->i_lock);
+		goto out_free;
+	}
+	inode->i_state |= I_WB_SWITCH;
+	spin_unlock(&inode->i_lock);
+
+	ihold(inode);
+	isw->inode = inode;
+
+	/*
+	 * In addition to synchronizing among switchers, I_WB_SWITCH tells
+	 * the RCU protected stat update paths to grab the mapping's
+	 * tree_lock so that stat transfer can synchronize against them.
+	 * Let's continue after I_WB_SWITCH is guaranteed to be visible.
+	 */
+	call_rcu(&isw->rcu_head, inode_switch_wbs_rcu_fn);
+	return;
+
+out_free:
+	if (isw->new_wb)
+		wb_put(isw->new_wb);
+	kfree(isw);
+}
+
+/**
+ * wbc_attach_and_unlock_inode - associate wbc with target inode and unlock it
+ * @wbc: writeback_control of interest
+ * @inode: target inode
+ *
+ * @inode is locked and about to be written back under the control of @wbc.
+ * Record @inode's writeback context into @wbc and unlock the i_lock.  On
+ * writeback completion, wbc_detach_inode() should be called.  This is used
+ * to track the cgroup writeback context.
+ */
+void wbc_attach_and_unlock_inode(struct writeback_control *wbc,
+				 struct inode *inode)
+{
+	if (!inode_cgwb_enabled(inode)) {
+		spin_unlock(&inode->i_lock);
+		return;
+	}
+
+	wbc->wb = inode_to_wb(inode);
+	wbc->inode = inode;
+
+	wbc->wb_id = wbc->wb->memcg_css->id;
+	wbc->wb_lcand_id = inode->i_wb_frn_winner;
+	wbc->wb_tcand_id = 0;
+	wbc->wb_bytes = 0;
+	wbc->wb_lcand_bytes = 0;
+	wbc->wb_tcand_bytes = 0;
+
+	wb_get(wbc->wb);
+	spin_unlock(&inode->i_lock);
+
+	/*
+	 * A dying wb indicates that the memcg-blkcg mapping has changed
+	 * and a new wb is already serving the memcg.  Switch immediately.
+	 */
+	if (unlikely(wb_dying(wbc->wb)))
+		inode_switch_wbs(inode, wbc->wb_id);
+}
+
+/**
+ * wbc_detach_inode - disassociate wbc from inode and perform foreign detection
+ * @wbc: writeback_control of the just finished writeback
+ *
+ * To be called after a writeback attempt of an inode finishes and undoes
+ * wbc_attach_and_unlock_inode().  Can be called under any context.
+ *
+ * As concurrent write sharing of an inode is expected to be very rare and
+ * memcg only tracks page ownership on first-use basis severely confining
+ * the usefulness of such sharing, cgroup writeback tracks ownership
+ * per-inode.  While the support for concurrent write sharing of an inode
+ * is deemed unnecessary, an inode being written to by different cgroups at
+ * different points in time is a lot more common, and, more importantly,
+ * charging only by first-use can too readily lead to grossly incorrect
+ * behaviors (single foreign page can lead to gigabytes of writeback to be
+ * incorrectly attributed).
+ *
+ * To resolve this issue, cgroup writeback detects the majority dirtier of
+ * an inode and transfers the ownership to it.  To avoid unnnecessary
+ * oscillation, the detection mechanism keeps track of history and gives
+ * out the switch verdict only if the foreign usage pattern is stable over
+ * a certain amount of time and/or writeback attempts.
+ *
+ * On each writeback attempt, @wbc tries to detect the majority writer
+ * using Boyer-Moore majority vote algorithm.  In addition to the byte
+ * count from the majority voting, it also counts the bytes written for the
+ * current wb and the last round's winner wb (max of last round's current
+ * wb, the winner from two rounds ago, and the last round's majority
+ * candidate).  Keeping track of the historical winner helps the algorithm
+ * to semi-reliably detect the most active writer even when it's not the
+ * absolute majority.
+ *
+ * Once the winner of the round is determined, whether the winner is
+ * foreign or not and how much IO time the round consumed is recorded in
+ * inode->i_wb_frn_history.  If the amount of recorded foreign IO time is
+ * over a certain threshold, the switch verdict is given.
+ */
+void wbc_detach_inode(struct writeback_control *wbc)
+{
+	struct bdi_writeback *wb = wbc->wb;
+	struct inode *inode = wbc->inode;
+	unsigned long avg_time, max_bytes, max_time;
+	u16 history;
+	int max_id;
+
+	if (!wb)
+		return;
+
+	history = inode->i_wb_frn_history;
+	avg_time = inode->i_wb_frn_avg_time;
+
+	/* pick the winner of this round */
+	if (wbc->wb_bytes >= wbc->wb_lcand_bytes &&
+	    wbc->wb_bytes >= wbc->wb_tcand_bytes) {
+		max_id = wbc->wb_id;
+		max_bytes = wbc->wb_bytes;
+	} else if (wbc->wb_lcand_bytes >= wbc->wb_tcand_bytes) {
+		max_id = wbc->wb_lcand_id;
+		max_bytes = wbc->wb_lcand_bytes;
+	} else {
+		max_id = wbc->wb_tcand_id;
+		max_bytes = wbc->wb_tcand_bytes;
+	}
+
+	/*
+	 * Calculate the amount of IO time the winner consumed and fold it
+	 * into the running average kept per inode.  If the consumed IO
+	 * time is lower than avag / WB_FRN_TIME_CUT_DIV, ignore it for
+	 * deciding whether to switch or not.  This is to prevent one-off
+	 * small dirtiers from skewing the verdict.
+	 */
+	max_time = DIV_ROUND_UP((max_bytes >> PAGE_SHIFT) << WB_FRN_TIME_SHIFT,
+				wb->avg_write_bandwidth);
+	if (avg_time)
+		avg_time += (max_time >> WB_FRN_TIME_AVG_SHIFT) -
+			    (avg_time >> WB_FRN_TIME_AVG_SHIFT);
+	else
+		avg_time = max_time;	/* immediate catch up on first run */
+
+	if (max_time >= avg_time / WB_FRN_TIME_CUT_DIV) {
+		int slots;
+
+		/*
+		 * The switch verdict is reached if foreign wb's consume
+		 * more than a certain proportion of IO time in a
+		 * WB_FRN_TIME_PERIOD.  This is loosely tracked by 16 slot
+		 * history mask where each bit represents one sixteenth of
+		 * the period.  Determine the number of slots to shift into
+		 * history from @max_time.
+		 */
+		slots = min(DIV_ROUND_UP(max_time, WB_FRN_HIST_UNIT),
+			    (unsigned long)WB_FRN_HIST_MAX_SLOTS);
+		history <<= slots;
+		if (wbc->wb_id != max_id)
+			history |= (1U << slots) - 1;
+
+		/*
+		 * Switch if the current wb isn't the consistent winner.
+		 * If there are multiple closely competing dirtiers, the
+		 * inode may switch across them repeatedly over time, which
+		 * is okay.  The main goal is avoiding keeping an inode on
+		 * the wrong wb for an extended period of time.
+		 */
+		if (hweight32(history) > WB_FRN_HIST_THR_SLOTS)
+			inode_switch_wbs(inode, max_id);
+	}
+
+	/*
+	 * Multiple instances of this function may race to update the
+	 * following fields but we don't mind occassional inaccuracies.
+	 */
+	inode->i_wb_frn_winner = max_id;
+	inode->i_wb_frn_avg_time = min(avg_time, (unsigned long)U16_MAX);
+	inode->i_wb_frn_history = history;
+
+	wb_put(wbc->wb);
+	wbc->wb = NULL;
+}
+
+/**
+ * wbc_account_io - account IO issued during writeback
+ * @wbc: writeback_control of the writeback in progress
+ * @page: page being written out
+ * @bytes: number of bytes being written out
+ *
+ * @bytes from @page are about to written out during the writeback
+ * controlled by @wbc.  Keep the book for foreign inode detection.  See
+ * wbc_detach_inode().
+ */
+void wbc_account_io(struct writeback_control *wbc, struct page *page,
+		    size_t bytes)
+{
+	int id;
+
+	/*
+	 * pageout() path doesn't attach @wbc to the inode being written
+	 * out.  This is intentional as we don't want the function to block
+	 * behind a slow cgroup.  Ultimately, we want pageout() to kick off
+	 * regular writeback instead of writing things out itself.
+	 */
+	if (!wbc->wb)
+		return;
+
+	rcu_read_lock();
+	id = mem_cgroup_css_from_page(page)->id;
+	rcu_read_unlock();
+
+	if (id == wbc->wb_id) {
+		wbc->wb_bytes += bytes;
+		return;
+	}
+
+	if (id == wbc->wb_lcand_id)
+		wbc->wb_lcand_bytes += bytes;
+
+	/* Boyer-Moore majority vote algorithm */
+	if (!wbc->wb_tcand_bytes)
+		wbc->wb_tcand_id = id;
+	if (id == wbc->wb_tcand_id)
+		wbc->wb_tcand_bytes += bytes;
+	else
+		wbc->wb_tcand_bytes -= min(bytes, wbc->wb_tcand_bytes);
+}
+
+/**
+ * inode_congested - test whether an inode is congested
+ * @inode: inode to test for congestion
+ * @cong_bits: mask of WB_[a]sync_congested bits to test
+ *
+ * Tests whether @inode is congested.  @cong_bits is the mask of congestion
+ * bits to test and the return value is the mask of set bits.
+ *
+ * If cgroup writeback is enabled for @inode, the congestion state is
+ * determined by whether the cgwb (cgroup bdi_writeback) for the blkcg
+ * associated with @inode is congested; otherwise, the root wb's congestion
+ * state is used.
+ */
+int inode_congested(struct inode *inode, int cong_bits)
+{
+	/*
+	 * Once set, ->i_wb never becomes NULL while the inode is alive.
+	 * Start transaction iff ->i_wb is visible.
+	 */
+	if (inode && inode_to_wb_is_valid(inode)) {
+		struct bdi_writeback *wb;
+		bool locked, congested;
+
+		wb = unlocked_inode_to_wb_begin(inode, &locked);
+		congested = wb_congested(wb, cong_bits);
+		unlocked_inode_to_wb_end(inode, locked);
+		return congested;
+	}
+
+	return wb_congested(&inode_to_bdi(inode)->wb, cong_bits);
+}
+EXPORT_SYMBOL_GPL(inode_congested);
+
+/**
+ * wb_wait_for_single_work - wait for completion of a single bdi_writeback_work
+ * @bdi: bdi the work item was issued to
+ * @work: work item to wait for
+ *
+ * Wait for the completion of @work which was issued to one of @bdi's
+ * bdi_writeback's.  The caller must have set @work->single_wait before
+ * issuing it.  This wait operates independently fo
+ * wb_wait_for_completion() and also disables automatic freeing of @work.
+ */
+static void wb_wait_for_single_work(struct backing_dev_info *bdi,
+				    struct wb_writeback_work *work)
+{
+	if (WARN_ON_ONCE(!work->single_wait))
+		return;
+
+	wait_event(bdi->wb_waitq, work->single_done);
+
+	/*
+	 * Paired with smp_wmb() in wb_do_writeback() and ensures that all
+	 * modifications to @work prior to assertion of ->single_done is
+	 * visible to the caller once this function returns.
+	 */
+	smp_rmb();
+}
+
+/**
+ * wb_split_bdi_pages - split nr_pages to write according to bandwidth
+ * @wb: target bdi_writeback to split @nr_pages to
+ * @nr_pages: number of pages to write for the whole bdi
+ *
+ * Split @wb's portion of @nr_pages according to @wb's write bandwidth in
+ * relation to the total write bandwidth of all wb's w/ dirty inodes on
+ * @wb->bdi.
+ */
+static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages)
+{
+	unsigned long this_bw = wb->avg_write_bandwidth;
+	unsigned long tot_bw = atomic_long_read(&wb->bdi->tot_write_bandwidth);
+
+	if (nr_pages == LONG_MAX)
+		return LONG_MAX;
+
+	/*
+	 * This may be called on clean wb's and proportional distribution
+	 * may not make sense, just use the original @nr_pages in those
+	 * cases.  In general, we wanna err on the side of writing more.
+	 */
+	if (!tot_bw || this_bw >= tot_bw)
+		return nr_pages;
+	else
+		return DIV_ROUND_UP_ULL((u64)nr_pages * this_bw, tot_bw);
+}
+
+/**
+ * wb_clone_and_queue_work - clone a wb_writeback_work and issue it to a wb
+ * @wb: target bdi_writeback
+ * @base_work: source wb_writeback_work
+ *
+ * Try to make a clone of @base_work and issue it to @wb.  If cloning
+ * succeeds, %true is returned; otherwise, @base_work is issued directly
+ * and %false is returned.  In the latter case, the caller is required to
+ * wait for @base_work's completion using wb_wait_for_single_work().
+ *
+ * A clone is auto-freed on completion.  @base_work never is.
+ */
+static bool wb_clone_and_queue_work(struct bdi_writeback *wb,
+				    struct wb_writeback_work *base_work)
 {
 	struct wb_writeback_work *work;
+
+	work = kmalloc(sizeof(*work), GFP_ATOMIC);
+	if (work) {
+		*work = *base_work;
+		work->auto_free = 1;
+		work->single_wait = 0;
+	} else {
+		work = base_work;
+		work->auto_free = 0;
+		work->single_wait = 1;
+	}
+	work->single_done = 0;
+	wb_queue_work(wb, work);
+	return work != base_work;
+}
+
+/**
+ * bdi_split_work_to_wbs - split a wb_writeback_work to all wb's of a bdi
+ * @bdi: target backing_dev_info
+ * @base_work: wb_writeback_work to issue
+ * @skip_if_busy: skip wb's which already have writeback in progress
+ *
+ * Split and issue @base_work to all wb's (bdi_writeback's) of @bdi which
+ * have dirty inodes.  If @base_work->nr_page isn't %LONG_MAX, it's
+ * distributed to the busy wbs according to each wb's proportion in the
+ * total active write bandwidth of @bdi.
+ */
+static void bdi_split_work_to_wbs(struct backing_dev_info *bdi,
+				  struct wb_writeback_work *base_work,
+				  bool skip_if_busy)
+{
+	long nr_pages = base_work->nr_pages;
+	int next_blkcg_id = 0;
+	struct bdi_writeback *wb;
+	struct wb_iter iter;
+
+	might_sleep();
+
+	if (!bdi_has_dirty_io(bdi))
+		return;
+restart:
+	rcu_read_lock();
+	bdi_for_each_wb(wb, bdi, &iter, next_blkcg_id) {
+		if (!wb_has_dirty_io(wb) ||
+		    (skip_if_busy && writeback_in_progress(wb)))
+			continue;
+
+		base_work->nr_pages = wb_split_bdi_pages(wb, nr_pages);
+		if (!wb_clone_and_queue_work(wb, base_work)) {
+			next_blkcg_id = wb->blkcg_css->id + 1;
+			rcu_read_unlock();
+			wb_wait_for_single_work(bdi, base_work);
+			goto restart;
+		}
+	}
+	rcu_read_unlock();
+}
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static struct bdi_writeback *
+locked_inode_to_wb_and_lock_list(struct inode *inode)
+	__releases(&inode->i_lock)
+	__acquires(&wb->list_lock)
+{
+	struct bdi_writeback *wb = inode_to_wb(inode);
+
+	spin_unlock(&inode->i_lock);
+	spin_lock(&wb->list_lock);
+	return wb;
+}
+
+static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode)
+	__acquires(&wb->list_lock)
+{
+	struct bdi_writeback *wb = inode_to_wb(inode);
+
+	spin_lock(&wb->list_lock);
+	return wb;
+}
+
+static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages)
+{
+	return nr_pages;
+}
+
+static void bdi_split_work_to_wbs(struct backing_dev_info *bdi,
+				  struct wb_writeback_work *base_work,
+				  bool skip_if_busy)
+{
+	might_sleep();
+
+	if (bdi_has_dirty_io(bdi) &&
+	    (!skip_if_busy || !writeback_in_progress(&bdi->wb))) {
+		base_work->auto_free = 0;
+		base_work->single_wait = 0;
+		base_work->single_done = 0;
+		wb_queue_work(&bdi->wb, base_work);
+	}
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
+
+void wb_start_writeback(struct bdi_writeback *wb, long nr_pages,
+			bool range_cyclic, enum wb_reason reason)
+{
+	struct wb_writeback_work *work;
+
+	if (!wb_has_dirty_io(wb))
+		return;
 
 	/*
 	 * This is WB_SYNC_NONE writeback, so if allocation fails just
···
 	 */
 	work = kzalloc(sizeof(*work), GFP_ATOMIC);
 	if (!work) {
-		trace_writeback_nowork(bdi);
-		bdi_wakeup_thread(bdi);
+		trace_writeback_nowork(wb->bdi);
+		wb_wakeup(wb);
 		return;
 	}
 
···
 	work->nr_pages	= nr_pages;
 	work->range_cyclic = range_cyclic;
 	work->reason	= reason;
+	work->auto_free	= 1;
 
-	bdi_queue_work(bdi, work);
+	wb_queue_work(wb, work);
 }
 
 /**
- * bdi_start_writeback - start writeback
- * @bdi: the backing device to write from
- * @nr_pages: the number of pages to write
- * @reason: reason why some writeback work was initiated
- *
- * Description:
- *   This does WB_SYNC_NONE opportunistic writeback. The IO is only
- *   started when this function returns, we make no guarantees on
- *   completion. Caller need not hold sb s_umount semaphore.
- *
- */
-void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
-			enum wb_reason reason)
-{
-	__bdi_start_writeback(bdi, nr_pages, true, reason);
-}
-
-/**
- * bdi_start_background_writeback - start background writeback
- * @bdi: the backing device to write from
+ * wb_start_background_writeback - start background writeback
+ * @wb: bdi_writback to write from
  *
  * Description:
  *   This makes sure WB_SYNC_NONE background writeback happens. When
- *   this function returns, it is only guaranteed that for given BDI
+ *   this function returns, it is only guaranteed that for given wb
  *   some IO is happening if we are over background dirty threshold.
  *   Caller need not hold sb s_umount semaphore.
  */
-void bdi_start_background_writeback(struct backing_dev_info *bdi)
+void wb_start_background_writeback(struct bdi_writeback *wb)
 {
 	/*
 	 * We just wake up the flusher thread. It will perform background
 	 * writeback as soon as there is no other work to do.
 	 */
-	trace_writeback_wake_background(bdi);
-	bdi_wakeup_thread(bdi);
+	trace_writeback_wake_background(wb->bdi);
+	wb_wakeup(wb);
 }
 
 /*
···
  */
 void inode_wb_list_del(struct inode *inode)
 {
-	struct backing_dev_info *bdi = inode_to_bdi(inode);
+	struct bdi_writeback *wb;
 
-	spin_lock(&bdi->wb.list_lock);
-	list_del_init(&inode->i_wb_list);
-	spin_unlock(&bdi->wb.list_lock);
+	wb = inode_to_wb_and_lock_list(inode);
+	inode_wb_list_del_locked(inode, wb);
+	spin_unlock(&wb->list_lock);
 }
 
 /*
···
  */
 static void redirty_tail(struct inode *inode, struct bdi_writeback *wb)
 {
-	assert_spin_locked(&wb->list_lock);
 	if (!list_empty(&wb->b_dirty)) {
 		struct inode *tail;
 
···
 		if (time_before(inode->dirtied_when, tail->dirtied_when))
 			inode->dirtied_when = jiffies;
 	}
-	list_move(&inode->i_wb_list, &wb->b_dirty);
+	inode_wb_list_move_locked(inode, wb, &wb->b_dirty);
 }
 
 /*
···
  */
 static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
 {
-	assert_spin_locked(&wb->list_lock);
-	list_move(&inode->i_wb_list, &wb->b_more_io);
+	inode_wb_list_move_locked(inode, wb, &wb->b_more_io);
 }
 
 static void inode_sync_complete(struct inode *inode)
···
 	moved = move_expired_inodes(&wb->b_dirty, &wb->b_io, 0, work);
 	moved += move_expired_inodes(&wb->b_dirty_time, &wb->b_io,
 				     EXPIRE_DIRTY_ATIME, work);
+	if (moved)
+		wb_io_lists_populated(wb);
 	trace_writeback_queue_io(wb, work, moved);
 }
 
···
 		redirty_tail(inode, wb);
 	} else if (inode->i_state & I_DIRTY_TIME) {
 		inode->dirtied_when = jiffies;
-		list_move(&inode->i_wb_list, &wb->b_dirty_time);
+		inode_wb_list_move_locked(inode, wb, &wb->b_dirty_time);
 	} else {
 		/* The inode is clean. Remove from writeback lists. */
-		list_del_init(&inode->i_wb_list);
+		inode_wb_list_del_locked(inode, wb);
 	}
 }
 
···
 	     !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_WRITEBACK)))
 		goto out;
 	inode->i_state |= I_SYNC;
-	spin_unlock(&inode->i_lock);
+	wbc_attach_and_unlock_inode(wbc, inode);
 
 	ret = __writeback_single_inode(inode, wbc);
 
+	wbc_detach_inode(wbc);
 	spin_lock(&wb->list_lock);
 	spin_lock(&inode->i_lock);
 	/*
···
 	 * touch it. See comment above for explanation.
 	 */
 	if (!(inode->i_state & I_DIRTY_ALL))
-		list_del_init(&inode->i_wb_list);
+		inode_wb_list_del_locked(inode, wb);
 	spin_unlock(&wb->list_lock);
 	inode_sync_complete(inode);
 out:
···
 	return ret;
 }
 
-static long writeback_chunk_size(struct backing_dev_info *bdi,
+static long writeback_chunk_size(struct bdi_writeback *wb,
 				 struct wb_writeback_work *work)
 {
 	long pages;
···
 	if (work->sync_mode == WB_SYNC_ALL || work->tagged_writepages)
 		pages = LONG_MAX;
 	else {
-		pages = min(bdi->avg_write_bandwidth / 2,
-			    global_dirty_limit / DIRTY_SCOPE);
+		pages = min(wb->avg_write_bandwidth / 2,
+			    global_wb_domain.dirty_limit / DIRTY_SCOPE);
 		pages = min(pages, work->nr_pages);
 		pages = round_down(pages + MIN_WRITEBACK_PAGES,
 				   MIN_WRITEBACK_PAGES);
···
 			continue;
 		}
 		inode->i_state |= I_SYNC;
-		spin_unlock(&inode->i_lock);
+		wbc_attach_and_unlock_inode(&wbc, inode);
 
-		write_chunk = writeback_chunk_size(wb->bdi, work);
+		write_chunk = writeback_chunk_size(wb, work);
 		wbc.nr_to_write = write_chunk;
 		wbc.pages_skipped = 0;
 
···
 		 */
 		__writeback_single_inode(inode, &wbc);
 
+		wbc_detach_inode(&wbc);
 		work->nr_pages -= write_chunk - wbc.nr_to_write;
 		wrote += write_chunk - wbc.nr_to_write;
 		spin_lock(&wb->list_lock);
···
 	return nr_pages - work.nr_pages;
 }
 
-static bool over_bground_thresh(struct backing_dev_info *bdi)
-{
-	unsigned long background_thresh, dirty_thresh;
-
-	global_dirty_limits(&background_thresh, &dirty_thresh);
-
-	if (global_page_state(NR_FILE_DIRTY) +
-	    global_page_state(NR_UNSTABLE_NFS) > background_thresh)
-		return true;
-
-	if (bdi_stat(bdi, BDI_RECLAIMABLE) >
-				bdi_dirty_limit(bdi, background_thresh))
-		return true;
-
-	return false;
-}
-
-/*
- * Called under wb->list_lock. If there are multiple wb per bdi,
- * only the flusher working on the first wb should do it.
- */
-static void wb_update_bandwidth(struct bdi_writeback *wb,
-				unsigned long start_time)
-{
-	__bdi_update_bandwidth(wb->bdi, 0, 0, 0, 0, 0, start_time);
-}
-
 /*
  * Explicit flushing or periodic writeback of "old" data.
  *
···
 		 * after the other works are all done.
 		 */
 		if ((work->for_background || work->for_kupdate) &&
-		    !list_empty(&wb->bdi->work_list))
+		    !list_empty(&wb->work_list))
 			break;
 
 		/*
 		 * For background writeout, stop when we are below the
 		 * background dirty threshold
 		 */
-		if (work->for_background && !over_bground_thresh(wb->bdi))
+		if (work->for_background && !wb_over_bg_thresh(wb))
 			break;
 
 		/*
···
 /*
  * Return the next wb_writeback_work struct that hasn't been processed yet.
  */
-static struct wb_writeback_work *
-get_next_work_item(struct backing_dev_info *bdi)
+static struct wb_writeback_work *get_next_work_item(struct bdi_writeback *wb)
 {
 	struct wb_writeback_work *work = NULL;
 
-	spin_lock_bh(&bdi->wb_lock);
-	if (!list_empty(&bdi->work_list)) {
-		work = list_entry(bdi->work_list.next,
+	spin_lock_bh(&wb->work_lock);
+	if (!list_empty(&wb->work_list)) {
+		work = list_entry(wb->work_list.next,
 				  struct wb_writeback_work, list);
 		list_del_init(&work->list);
 	}
-	spin_unlock_bh(&bdi->wb_lock);
+	spin_unlock_bh(&wb->work_lock);
 	return work;
 }
 
···
 
 static long wb_check_background_flush(struct bdi_writeback *wb)
 {
-	if (over_bground_thresh(wb->bdi)) {
+	if (wb_over_bg_thresh(wb)) {
 
 		struct wb_writeback_work work = {
 			.nr_pages	= LONG_MAX,
···
  */
 static long wb_do_writeback(struct bdi_writeback *wb)
 {
-	struct backing_dev_info *bdi = wb->bdi;
 	struct wb_writeback_work *work;
 	long wrote = 0;
 
-	set_bit(BDI_writeback_running, &wb->bdi->state);
-	while ((work = get_next_work_item(bdi)) != NULL) {
+	set_bit(WB_writeback_running, &wb->state);
+	while ((work = get_next_work_item(wb)) != NULL) {
+		struct wb_completion *done = work->done;
+		bool need_wake_up = false;
 
-		trace_writeback_exec(bdi, work);
+		trace_writeback_exec(wb->bdi, work);
 
 		wrote += wb_writeback(wb, work);
 
-		/*
-		 * Notify the caller of completion if this is a synchronous
-		 * work item, otherwise just free it.
-		 */
-		if (work->done)
-			complete(work->done);
-		else
+		if (work->single_wait) {
+			WARN_ON_ONCE(work->auto_free);
+			/* paired w/ rmb in wb_wait_for_single_work() */
+			smp_wmb();
+			work->single_done = 1;
+			need_wake_up = true;
+		} else if (work->auto_free) {
 			kfree(work);
+		}
+
+		if (done && atomic_dec_and_test(&done->cnt))
+			need_wake_up = true;
+
+		if (need_wake_up)
+			wake_up_all(&wb->bdi->wb_waitq);
 	}
 
 	/*
···
 	 */
 	wrote += wb_check_old_data_flush(wb);
 	wrote += wb_check_background_flush(wb);
-	clear_bit(BDI_writeback_running, &wb->bdi->state);
+	clear_bit(WB_writeback_running, &wb->state);
 
 	return wrote;
 }
···
  * Handle writeback of dirty data for the device backed by this bdi. Also
  * reschedules periodically and does kupdated style flushing.
  */
-void bdi_writeback_workfn(struct work_struct *work)
+void wb_workfn(struct work_struct *work)
 {
 	struct bdi_writeback *wb = container_of(to_delayed_work(work),
 						struct bdi_writeback, dwork);
-	struct backing_dev_info *bdi = wb->bdi;
 	long pages_written;
 
-	set_worker_desc("flush-%s", dev_name(bdi->dev));
+	set_worker_desc("flush-%s", dev_name(wb->bdi->dev));
 	current->flags |= PF_SWAPWRITE;
 
 	if (likely(!current_is_workqueue_rescuer() ||
-		   !test_bit(BDI_registered, &bdi->state))) {
+		   !test_bit(WB_registered, &wb->state))) {
 		/*
-		 * The normal path.  Keep writing back @bdi until its
+		 * The normal path.  Keep writing back @wb until its
 		 * work_list is empty.  Note that this path is also taken
-		 * if @bdi is shutting down even when we're running off the
+		 * if @wb is shutting down even when we're running off the
 		 * rescuer as work_list needs to be drained.
 		 */
 		do {
 			pages_written = wb_do_writeback(wb);
 			trace_writeback_pages_written(pages_written);
-		} while (!list_empty(&bdi->work_list));
+		} while (!list_empty(&wb->work_list));
 	} else {
 		/*
 		 * bdi_wq can't get enough workers and we're running off
 		 * the emergency worker.  Don't hog it.  Hopefully, 1024 is
 		 * enough for efficient IO.
 		 */
-		pages_written = writeback_inodes_wb(&bdi->wb, 1024,
+		pages_written = writeback_inodes_wb(wb, 1024,
 						    WB_REASON_FORKER_THREAD);
 		trace_writeback_pages_written(pages_written);
 	}
 
-	if (!list_empty(&bdi->work_list))
+	if (!list_empty(&wb->work_list))
 		mod_delayed_work(bdi_wq, &wb->dwork, 0);
 	else if (wb_has_dirty_io(wb) && dirty_writeback_interval)
-		bdi_wakeup_thread_delayed(bdi);
+		wb_wakeup_delayed(wb);
 
 	current->flags &= ~PF_SWAPWRITE;
 }
···
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
+		struct bdi_writeback *wb;
+		struct wb_iter iter;
+
 		if (!bdi_has_dirty_io(bdi))
 			continue;
-		__bdi_start_writeback(bdi, nr_pages, false, reason);
+
+		bdi_for_each_wb(wb, bdi, &iter, 0)
+			wb_start_writeback(wb, wb_split_bdi_pages(wb, nr_pages),
+					   false, reason);
 	}
 	rcu_read_unlock();
 }
···
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
-		if (list_empty(&bdi->wb.b_dirty_time))
-			continue;
-		bdi_wakeup_thread(bdi);
+		struct bdi_writeback *wb;
+		struct wb_iter iter;
+
+		bdi_for_each_wb(wb, bdi, &iter, 0)
+			if (!list_empty(&bdi->wb.b_dirty_time))
+				wb_wakeup(&bdi->wb);
 	}
 	rcu_read_unlock();
 	schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ);
···
 void __mark_inode_dirty(struct inode *inode, int flags)
 {
 	struct super_block *sb = inode->i_sb;
-	struct backing_dev_info *bdi = NULL;
 	int dirtytime;
 
 	trace_writeback_mark_inode_dirty(inode, flags);
···
 	if ((inode->i_state & flags) != flags) {
 		const int was_dirty = inode->i_state & I_DIRTY;
 
+		inode_attach_wb(inode, NULL);
+
 		if (flags & I_DIRTY_INODE)
 			inode->i_state &= ~I_DIRTY_TIME;
 		inode->i_state |= flags;
···
 		 * reposition it (that would break b_dirty time-ordering).
 		 */
 		if (!was_dirty) {
+			struct bdi_writeback *wb;
+			struct list_head *dirty_list;
 			bool wakeup_bdi = false;
-			bdi = inode_to_bdi(inode);
 
-			spin_unlock(&inode->i_lock);
-			spin_lock(&bdi->wb.list_lock);
-			if (bdi_cap_writeback_dirty(bdi)) {
-				WARN(!test_bit(BDI_registered, &bdi->state),
-				     "bdi-%s not registered\n", bdi->name);
+			wb = locked_inode_to_wb_and_lock_list(inode);
 
-				/*
-				 * If this is the first dirty inode for this
-				 * bdi, we have to wake-up the corresponding
-				 * bdi thread to make sure background
-				 * write-back happens later.
-				 */
-				if (!wb_has_dirty_io(&bdi->wb))
-					wakeup_bdi = true;
-			}
+			WARN(bdi_cap_writeback_dirty(wb->bdi) &&
+			     !test_bit(WB_registered, &wb->state),
+			     "bdi-%s not registered\n", wb->bdi->name);
 
 			inode->dirtied_when = jiffies;
 			if (dirtytime)
 				inode->dirtied_time_when = jiffies;
+
 			if (inode->i_state & (I_DIRTY_INODE | I_DIRTY_PAGES))
-				list_move(&inode->i_wb_list, &bdi->wb.b_dirty);
+				dirty_list = &wb->b_dirty;
 			else
-				list_move(&inode->i_wb_list,
-					  &bdi->wb.b_dirty_time);
-			spin_unlock(&bdi->wb.list_lock);
+				dirty_list = &wb->b_dirty_time;
+
+			wakeup_bdi = inode_wb_list_move_locked(inode, wb,
+							       dirty_list);
+
+			spin_unlock(&wb->list_lock);
 			trace_writeback_dirty_inode_enqueue(inode);
 
-			if (wakeup_bdi)
-				bdi_wakeup_thread_delayed(bdi);
+			/*
+			 * If this is the first dirty inode for this bdi,
+			 * we have to wake-up the corresponding bdi thread
+			 * to make sure background write-back happens
+			 * later.
+			 */
+			if (bdi_cap_writeback_dirty(wb->bdi) && wakeup_bdi)
+				wb_wakeup_delayed(wb);
 			return;
 		}
 	}
···
 	iput(old_inode);
 }
 
+static void __writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr,
+				     enum wb_reason reason, bool skip_if_busy)
+{
+	DEFINE_WB_COMPLETION_ONSTACK(done);
+	struct wb_writeback_work work = {
+		.sb			= sb,
+		.sync_mode		= WB_SYNC_NONE,
+		.tagged_writepages	= 1,
+		.done			= &done,
+		.nr_pages		= nr,
+		.reason			= reason,
+	};
+	struct backing_dev_info *bdi = sb->s_bdi;
+
+	if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info)
+		return;
+	WARN_ON(!rwsem_is_locked(&sb->s_umount));
+
+	bdi_split_work_to_wbs(sb->s_bdi, &work, skip_if_busy);
+	wb_wait_for_completion(bdi, &done);
+}
+
 /**
  * writeback_inodes_sb_nr -	writeback dirty inodes from given super_block
  * @sb: the superblock
···
 			    unsigned long nr,
 			    enum wb_reason reason)
 {
-	DECLARE_COMPLETION_ONSTACK(done);
-	struct wb_writeback_work work = {
-		.sb			= sb,
-		.sync_mode		= WB_SYNC_NONE,
-		.tagged_writepages	= 1,
-		.done			= &done,
-		.nr_pages		= nr,
-		.reason			= reason,
-	};
-
-	if (sb->s_bdi == &noop_backing_dev_info)
-		return;
-	WARN_ON(!rwsem_is_locked(&sb->s_umount));
-	bdi_queue_work(sb->s_bdi, &work);
-	wait_for_completion(&done);
+	__writeback_inodes_sb_nr(sb, nr, reason, false);
 }
 EXPORT_SYMBOL(writeback_inodes_sb_nr);
 
···
  * Invoke writeback_inodes_sb_nr if no writeback is currently underway.
  * Returns 1 if writeback was started, 0 if not.
  */
-int try_to_writeback_inodes_sb_nr(struct super_block *sb,
-				  unsigned long nr,
-				  enum wb_reason reason)
+bool try_to_writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr,
+				   enum wb_reason reason)
 {
-	if (writeback_in_progress(sb->s_bdi))
-		return 1;
-
 	if (!down_read_trylock(&sb->s_umount))
-		return 0;
+		return false;
 
-	writeback_inodes_sb_nr(sb, nr, reason);
+	__writeback_inodes_sb_nr(sb, nr, reason, true);
 	up_read(&sb->s_umount);
-	return 1;
+	return true;
 }
 EXPORT_SYMBOL(try_to_writeback_inodes_sb_nr);
 
···
  * Implement by try_to_writeback_inodes_sb_nr()
  * Returns 1 if writeback was started, 0 if not.
  */
-int try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason)
+bool try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason)
 {
 	return try_to_writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason);
 }
···
  */
 void sync_inodes_sb(struct super_block *sb)
 {
-	DECLARE_COMPLETION_ONSTACK(done);
+	DEFINE_WB_COMPLETION_ONSTACK(done);
 	struct wb_writeback_work work = {
 		.sb		= sb,
 		.sync_mode	= WB_SYNC_ALL,
···
 		.reason		= WB_REASON_SYNC,
 		.for_sync	= 1,
 	};
+	struct backing_dev_info *bdi = sb->s_bdi;
 
 	/* Nothing to do? */
-	if (sb->s_bdi == &noop_backing_dev_info)
+	if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info)
 		return;
 	WARN_ON(!rwsem_is_locked(&sb->s_umount));
 
-	bdi_queue_work(sb->s_bdi, &work);
-	wait_for_completion(&done);
+	bdi_split_work_to_wbs(bdi, &work, false);
+	wb_wait_for_completion(bdi, &done);
 
 	wait_sb_inodes(sb);
 }

fs/fuse/file.c

···
 
 	list_del(&req->writepages_entry);
 	for (i = 0; i < req->num_pages; i++) {
-		dec_bdi_stat(bdi, BDI_WRITEBACK);
+		dec_wb_stat(&bdi->wb, WB_WRITEBACK);
 		dec_zone_page_state(req->pages[i], NR_WRITEBACK_TEMP);
-		bdi_writeout_inc(bdi);
+		wb_writeout_inc(&bdi->wb);
 	}
 	wake_up(&fi->page_waitq);
 }
···
 	req->end = fuse_writepage_end;
 	req->inode = inode;
 
-	inc_bdi_stat(inode_to_bdi(inode), BDI_WRITEBACK);
+	inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK);
 	inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP);
 
 	spin_lock(&fc->lock);
···
 		copy_highpage(old_req->pages[0], page);
 		spin_unlock(&fc->lock);
 
-		dec_bdi_stat(bdi, BDI_WRITEBACK);
+		dec_wb_stat(&bdi->wb, WB_WRITEBACK);
 		dec_zone_page_state(page, NR_WRITEBACK_TEMP);
-		bdi_writeout_inc(bdi);
+		wb_writeout_inc(&bdi->wb);
 		fuse_writepage_free(fc, new_req);
 		fuse_request_free(new_req);
 		goto out;
···
 	req->page_descs[req->num_pages].offset = 0;
 	req->page_descs[req->num_pages].length = PAGE_SIZE;
 
-	inc_bdi_stat(inode_to_bdi(inode), BDI_WRITEBACK);
+	inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK);
 	inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP);
 
 	err = 0;

fs/gfs2/super.c

···
 
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		gfs2_log_flush(GFS2_SB(inode), ip->i_gl, NORMAL_FLUSH);
-	if (bdi->dirty_exceeded)
+	if (bdi->wb.dirty_exceeded)
 		gfs2_ail1_flush(sdp, wbc);
 	else
 		filemap_fdatawrite(metamapping);

fs/hfs/super.c

···
 
 #include <linux/module.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/mount.h>
 #include <linux/init.h>
 #include <linux/nls.h>

fs/hfsplus/super.c

···
 #include <linux/init.h>
 #include <linux/pagemap.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/fs.h>
 #include <linux/slab.h>
 #include <linux/vfs.h>

fs/inode.c

···
 void __destroy_inode(struct inode *inode)
 {
 	BUG_ON(inode_has_buffers(inode));
+	inode_detach_wb(inode);
 	security_inode_free(inode);
 	fsnotify_inode_delete(inode);
 	locks_free_lock_context(inode->i_flctx);

fs/mpage.c

···
 				bio_get_nr_vecs(bdev), GFP_NOFS|__GFP_HIGH);
 		if (bio == NULL)
 			goto confused;
+
+		wbc_init_bio(wbc, bio);
 	}
 
 	/*
···
 	 * the confused fail path above (OOM) will be very confused when
 	 * it finds all bh marked clean (i.e. it will not write anything)
 	 */
+	wbc_account_io(wbc, page, PAGE_SIZE);
 	length = first_unmapped << blkbits;
 	if (bio_add_page(bio, page, length, 0) < length) {
 		bio = mpage_bio_submit(WRITE, bio);

fs/nfs/filelayout/filelayout.c

···
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
 #include <linux/module.h>
+#include <linux/backing-dev.h>
 
 #include <linux/sunrpc/metrics.h>
 

fs/nfs/internal.h

···
 	struct inode *inode = page_file_mapping(page)->host;
 
 	inc_zone_page_state(page, NR_UNSTABLE_NFS);
-	inc_bdi_stat(inode_to_bdi(inode), BDI_RECLAIMABLE);
+	inc_wb_stat(&inode_to_bdi(inode)->wb, WB_RECLAIMABLE);
 	 __mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 }
 

fs/nfs/write.c

···
 nfs_clear_page_commit(struct page *page)
 {
 	dec_zone_page_state(page, NR_UNSTABLE_NFS);
-	dec_bdi_stat(inode_to_bdi(page_file_mapping(page)->host), BDI_RECLAIMABLE);
+	dec_wb_stat(&inode_to_bdi(page_file_mapping(page)->host)->wb,
+		    WB_RECLAIMABLE);
 }
 
 /* Called holding inode (/cinfo) lock */

fs/ocfs2/file.c

···
 #include <linux/falloc.h>
 #include <linux/quotaops.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 
 #include <cluster/masklog.h>
 

fs/reiserfs/super.c

···
 #include "xattr.h"
 #include <linux/init.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/buffer_head.h>
 #include <linux/exportfs.h>
 #include <linux/quotaops.h>

fs/ufs/super.c

···
 #include <linux/stat.h>
 #include <linux/string.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/init.h>
 #include <linux/parser.h>
 #include <linux/buffer_head.h>

fs/xfs/xfs_aops.c

···
 	loff_t			end_offset;
 	loff_t			offset;
 	int			newly_dirty;
+	struct mem_cgroup	*memcg;
 
 	if (unlikely(!mapping))
 		return !TestSetPageDirty(page);
···
 			offset += 1 << inode->i_blkbits;
 		} while (bh != head);
 	}
+	/*
+	 * Use mem_group_begin_page_stat() to keep PageDirty synchronized with
+	 * per-memcg dirty page counters.
+	 */
+	memcg = mem_cgroup_begin_page_stat(page);
 	newly_dirty = !TestSetPageDirty(page);
 	spin_unlock(&mapping->private_lock);
 
···
 		spin_lock_irqsave(&mapping->tree_lock, flags);
 		if (page->mapping) {	/* Race with truncate? */
 			WARN_ON_ONCE(!PageUptodate(page));
-			account_page_dirtied(page, mapping);
+			account_page_dirtied(page, mapping, memcg);
 			radix_tree_tag_set(&mapping->page_tree,
 					page_index(page), PAGECACHE_TAG_DIRTY);
 		}
 		spin_unlock_irqrestore(&mapping->tree_lock, flags);
-		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 	}
+	mem_cgroup_end_page_stat(memcg);
+	if (newly_dirty)
+		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 	return newly_dirty;
 }
 

fs/xfs/xfs_file.c

···
 #include <linux/dcache.h>
 #include <linux/falloc.h>
 #include <linux/pagevec.h>
+#include <linux/backing-dev.h>
 
 static const struct vm_operations_struct xfs_file_vm_ops;
 

include/linux/backing-dev-defs.h

···
+#ifndef __LINUX_BACKING_DEV_DEFS_H
+#define __LINUX_BACKING_DEV_DEFS_H
+
+#include <linux/list.h>
+#include <linux/radix-tree.h>
+#include <linux/rbtree.h>
+#include <linux/spinlock.h>
+#include <linux/percpu_counter.h>
+#include <linux/percpu-refcount.h>
+#include <linux/flex_proportions.h>
+#include <linux/timer.h>
+#include <linux/workqueue.h>
+
+struct page;
+struct device;
+struct dentry;
+
+/*
+ * Bits in bdi_writeback.state
+ */
+enum wb_state {
+	WB_registered,		/* bdi_register() was done */
+	WB_writeback_running,	/* Writeback is in progress */
+	WB_has_dirty_io,	/* Dirty inodes on ->b_{dirty|io|more_io} */
+};
+
+enum wb_congested_state {
+	WB_async_congested,	/* The async (write) queue is getting full */
+	WB_sync_congested,	/* The sync queue is getting full */
+};
+
+typedef int (congested_fn)(void *, int);
+
+enum wb_stat_item {
+	WB_RECLAIMABLE,
+	WB_WRITEBACK,
+	WB_DIRTIED,
+	WB_WRITTEN,
+	NR_WB_STAT_ITEMS
+};
+
+#define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids)))
+
+/*
+ * For cgroup writeback, multiple wb's may map to the same blkcg.  Those
+ * wb's can operate mostly independently but should share the congested
+ * state.  To facilitate such sharing, the congested state is tracked using
+ * the following struct which is created on demand, indexed by blkcg ID on
+ * its bdi, and refcounted.
+ */
+struct bdi_writeback_congested {
+	unsigned long state;		/* WB_[a]sync_congested flags */
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct backing_dev_info *bdi;	/* the associated bdi */
+	atomic_t refcnt;		/* nr of attached wb's and blkg */
+	int blkcg_id;			/* ID of the associated blkcg */
+	struct rb_node rb_node;		/* on bdi->cgwb_congestion_tree */
+#endif
+};
+
+/*
+ * Each wb (bdi_writeback) can perform writeback operations, is measured
+ * and throttled, independently.  Without cgroup writeback, each bdi
+ * (bdi_writeback) is served by its embedded bdi->wb.
+ *
+ * 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, which enables isolation across different
+ * cgroups and propagation of IO back pressure down from the IO layer upto
+ * the tasks which are generating the dirty pages to be written back.
+ *
+ * A cgroup wb is indexed on its bdi by the ID of the associated memcg,
+ * refcounted with the number of inodes attached to it, and pins the memcg
+ * and the corresponding blkcg.  As the corresponding blkcg for a memcg may
+ * change as blkcg is disabled and enabled higher up in the hierarchy, a wb
+ * is tested for blkcg after lookup and removed from index on mismatch so
+ * that a new wb for the combination can be created.
+ */
+struct bdi_writeback {
+	struct backing_dev_info *bdi;	/* our parent bdi */
+
+	unsigned long state;		/* Always use atomic bitops on this */
+	unsigned long last_old_flush;	/* last old data flush */
+
+	struct list_head b_dirty;	/* dirty inodes */
+	struct list_head b_io;		/* parked for writeback */
+	struct list_head b_more_io;	/* parked for more writeback */
+	struct list_head b_dirty_time;	/* time stamps are dirty */
+	spinlock_t list_lock;		/* protects the b_* lists */
+
+	struct percpu_counter stat[NR_WB_STAT_ITEMS];
+
+	struct bdi_writeback_congested *congested;
+
+	unsigned long bw_time_stamp;	/* last time write bw is updated */
+	unsigned long dirtied_stamp;
+	unsigned long written_stamp;	/* pages written at bw_time_stamp */
+	unsigned long write_bandwidth;	/* the estimated write bandwidth */
+	unsigned long avg_write_bandwidth; /* further smoothed write bw, > 0 */
+
+	/*
+	 * The base dirty throttle rate, re-calculated on every 200ms.
+	 * All the bdi tasks' dirty rate will be curbed under it.
+	 * @dirty_ratelimit tracks the estimated @balanced_dirty_ratelimit
+	 * in small steps and is much more smooth/stable than the latter.
+	 */
+	unsigned long dirty_ratelimit;
+	unsigned long balanced_dirty_ratelimit;
+
+	struct fprop_local_percpu completions;
+	int dirty_exceeded;
+
+	spinlock_t work_lock;		/* protects work_list & dwork scheduling */
+	struct list_head work_list;
+	struct delayed_work dwork;	/* work item used for writeback */
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct percpu_ref refcnt;	/* used only for !root wb's */
+	struct fprop_local_percpu memcg_completions;
+	struct cgroup_subsys_state *memcg_css; /* the associated memcg */
+	struct cgroup_subsys_state *blkcg_css; /* and blkcg */
+	struct list_head memcg_node;	/* anchored at memcg->cgwb_list */
+	struct list_head blkcg_node;	/* anchored at blkcg->cgwb_list */
+
+	union {
+		struct work_struct release_work;
+		struct rcu_head rcu;
+	};
+#endif
+};
+
+struct backing_dev_info {
+	struct list_head bdi_list;
+	unsigned long ra_pages;	/* max readahead in PAGE_CACHE_SIZE units */
+	unsigned int capabilities; /* Device capabilities */
+	congested_fn *congested_fn; /* Function pointer if device is md/dm */
+	void *congested_data;	/* Pointer to aux data for congested func */
+
+	char *name;
+
+	unsigned int min_ratio;
+	unsigned int max_ratio, max_prop_frac;
+
+	/*
+	 * Sum of avg_write_bw of wbs with dirty inodes.  > 0 if there are
+	 * any dirty wbs, which is depended upon by bdi_has_dirty().
+	 */
+	atomic_long_t tot_write_bandwidth;
+
+	struct bdi_writeback wb;  /* the root writeback info for this bdi */
+	struct bdi_writeback_congested wb_congested; /* its congested state */
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct radix_tree_root cgwb_tree; /* radix tree of active cgroup wbs */
+	struct rb_root cgwb_congested_tree; /* their congested states */
+	atomic_t usage_cnt; /* counts both cgwbs and cgwb_contested's */
+#endif
+	wait_queue_head_t wb_waitq;
+
+	struct device *dev;
+
+	struct timer_list laptop_mode_wb_timer;
+
+#ifdef CONFIG_DEBUG_FS
+	struct dentry *debug_dir;
+	struct dentry *debug_stats;
+#endif
+};
+
+enum {
+	BLK_RW_ASYNC	= 0,
+	BLK_RW_SYNC	= 1,
+};
+
+void clear_wb_congested(struct bdi_writeback_congested *congested, int sync);
+void set_wb_congested(struct bdi_writeback_congested *congested, int sync);
+
+static inline void clear_bdi_congested(struct backing_dev_info *bdi, int sync)
+{
+	clear_wb_congested(bdi->wb.congested, sync);
+}
+
+static inline void set_bdi_congested(struct backing_dev_info *bdi, int sync)
+{
+	set_wb_congested(bdi->wb.congested, sync);
+}
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+/**
+ * wb_tryget - try to increment a wb's refcount
+ * @wb: bdi_writeback to get
+ */
+static inline bool wb_tryget(struct bdi_writeback *wb)
+{
+	if (wb != &wb->bdi->wb)
+		return percpu_ref_tryget(&wb->refcnt);
+	return true;
+}
+
+/**
+ * wb_get - increment a wb's refcount
+ * @wb: bdi_writeback to get
+ */
+static inline void wb_get(struct bdi_writeback *wb)
+{
+	if (wb != &wb->bdi->wb)
+		percpu_ref_get(&wb->refcnt);
+}
+
+/**
+ * wb_put - decrement a wb's refcount
+ * @wb: bdi_writeback to put
+ */
+static inline void wb_put(struct bdi_writeback *wb)
+{
+	if (wb != &wb->bdi->wb)
+		percpu_ref_put(&wb->refcnt);
+}
+
+/**
+ * wb_dying - is a wb dying?
+ * @wb: bdi_writeback of interest
+ *
+ * Returns whether @wb is unlinked and being drained.
+ */
+static inline bool wb_dying(struct bdi_writeback *wb)
+{
+	return percpu_ref_is_dying(&wb->refcnt);
+}
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static inline bool wb_tryget(struct bdi_writeback *wb)
+{
+	return true;
+}
+
+static inline void wb_get(struct bdi_writeback *wb)
+{
+}
+
+static inline void wb_put(struct bdi_writeback *wb)
+{
+}
+
+static inline bool wb_dying(struct bdi_writeback *wb)
+{
+	return false;
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
+
+#endif	/* __LINUX_BACKING_DEV_DEFS_H */

include/linux/backing-dev.h

···
 #ifndef _LINUX_BACKING_DEV_H
 #define _LINUX_BACKING_DEV_H
 
-#include <linux/percpu_counter.h>
-#include <linux/log2.h>
-#include <linux/flex_proportions.h>
 #include <linux/kernel.h>
 #include <linux/fs.h>
 #include <linux/sched.h>
-#include <linux/timer.h>
+#include <linux/blkdev.h>
 #include <linux/writeback.h>
-#include <linux/atomic.h>
-#include <linux/sysctl.h>
-#include <linux/workqueue.h>
-
-struct page;
-struct device;
-struct dentry;
-
-/*
- * Bits in backing_dev_info.state
- */
-enum bdi_state {
-	BDI_async_congested,	/* The async (write) queue is getting full */
-	BDI_sync_congested,	/* The sync queue is getting full */
-	BDI_registered,		/* bdi_register() was done */
-	BDI_writeback_running,	/* Writeback is in progress */
-};
-
-typedef int (congested_fn)(void *, int);
-
-enum bdi_stat_item {
-	BDI_RECLAIMABLE,
-	BDI_WRITEBACK,
-	BDI_DIRTIED,
-	BDI_WRITTEN,
-	NR_BDI_STAT_ITEMS
-};
-
-#define BDI_STAT_BATCH (8*(1+ilog2(nr_cpu_ids)))
-
-struct bdi_writeback {
-	struct backing_dev_info *bdi;	/* our parent bdi */
-
-	unsigned long last_old_flush;	/* last old data flush */
-
-	struct delayed_work dwork;	/* work item used for writeback */
-	struct list_head b_dirty;	/* dirty inodes */
-	struct list_head b_io;		/* parked for writeback */
-	struct list_head b_more_io;	/* parked for more writeback */
-	struct list_head b_dirty_time;	/* time stamps are dirty */
-	spinlock_t list_lock;		/* protects the b_* lists */
-};
-
-struct backing_dev_info {
-	struct list_head bdi_list;
-	unsigned long ra_pages;	/* max readahead in PAGE_CACHE_SIZE units */
-	unsigned long state;	/* Always use atomic bitops on this */
-	unsigned int capabilities; /* Device capabilities */
-	congested_fn *congested_fn; /* Function pointer if device is md/dm */
-	void *congested_data;	/* Pointer to aux data for congested func */
-
-	char *name;
-
-	struct percpu_counter bdi_stat[NR_BDI_STAT_ITEMS];
-
-	unsigned long bw_time_stamp;	/* last time write bw is updated */
-	unsigned long dirtied_stamp;
-	unsigned long written_stamp;	/* pages written at bw_time_stamp */
-	unsigned long write_bandwidth;	/* the estimated write bandwidth */
-	unsigned long avg_write_bandwidth; /* further smoothed write bw */
-
-	/*
-	 * The base dirty throttle rate, re-calculated on every 200ms.
-	 * All the bdi tasks' dirty rate will be curbed under it.
-	 * @dirty_ratelimit tracks the estimated @balanced_dirty_ratelimit
-	 * in small steps and is much more smooth/stable than the latter.
-	 */
-	unsigned long dirty_ratelimit;
-	unsigned long balanced_dirty_ratelimit;
-
-	struct fprop_local_percpu completions;
-	int dirty_exceeded;
-
-	unsigned int min_ratio;
-	unsigned int max_ratio, max_prop_frac;
-
-	struct bdi_writeback wb;  /* default writeback info for this bdi */
-	spinlock_t wb_lock;	  /* protects work_list & wb.dwork scheduling */
-
-	struct list_head work_list;
-
-	struct device *dev;
-
-	struct timer_list laptop_mode_wb_timer;
-
-#ifdef CONFIG_DEBUG_FS
-	struct dentry *debug_dir;
-	struct dentry *debug_stats;
-#endif
-};
-
-struct backing_dev_info *inode_to_bdi(struct inode *inode);
+#include <linux/blk-cgroup.h>
+#include <linux/backing-dev-defs.h>
 
 int __must_check bdi_init(struct backing_dev_info *bdi);
 void bdi_destroy(struct backing_dev_info *bdi);
···
 		const char *fmt, ...);
 int bdi_register_dev(struct backing_dev_info *bdi, dev_t dev);
 int __must_check bdi_setup_and_register(struct backing_dev_info *, char *);
-void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
-			enum wb_reason reason);
-void bdi_start_background_writeback(struct backing_dev_info *bdi);
-void bdi_writeback_workfn(struct work_struct *work);
-int bdi_has_dirty_io(struct backing_dev_info *bdi);
-void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi);
+void wb_start_writeback(struct bdi_writeback *wb, long nr_pages,
+			bool range_cyclic, enum wb_reason reason);
+void wb_start_background_writeback(struct bdi_writeback *wb);
+void wb_workfn(struct work_struct *work);
+void wb_wakeup_delayed(struct bdi_writeback *wb);
 
 extern spinlock_t bdi_lock;
 extern struct list_head bdi_list;
 
 extern struct workqueue_struct *bdi_wq;
 
-static inline int wb_has_dirty_io(struct bdi_writeback *wb)
+static inline bool wb_has_dirty_io(struct bdi_writeback *wb)
 {
-	return !list_empty(&wb->b_dirty) ||
-	       !list_empty(&wb->b_io) ||
-	       !list_empty(&wb->b_more_io);
+	return test_bit(WB_has_dirty_io, &wb->state);
 }
 
-static inline void __add_bdi_stat(struct backing_dev_info *bdi,
-		enum bdi_stat_item item, s64 amount)
+static inline bool bdi_has_dirty_io(struct backing_dev_info *bdi)
 {
-	__percpu_counter_add(&bdi->bdi_stat[item], amount, BDI_STAT_BATCH);
+	/*
+	 * @bdi->tot_write_bandwidth is guaranteed to be > 0 if there are
+	 * any dirty wbs.  See wb_update_write_bandwidth().
+	 */
+	return atomic_long_read(&bdi->tot_write_bandwidth);
 }
 
-static inline void __inc_bdi_stat(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
+static inline void __add_wb_stat(struct bdi_writeback *wb,
+				 enum wb_stat_item item, s64 amount)
 {
-	__add_bdi_stat(bdi, item, 1);
+	__percpu_counter_add(&wb->stat[item], amount, WB_STAT_BATCH);
 }
 
-static inline void inc_bdi_stat(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
+static inline void __inc_wb_stat(struct bdi_writeback *wb,
+				 enum wb_stat_item item)
 {
-	unsigned long flags;
-
-	local_irq_save(flags);
-	__inc_bdi_stat(bdi, item);
-	local_irq_restore(flags);
+	__add_wb_stat(wb, item, 1);
 }
 
-static inline void __dec_bdi_stat(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
-{
-	__add_bdi_stat(bdi, item, -1);
-}
-
-static inline void dec_bdi_stat(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
+static inline void inc_wb_stat(struct bdi_writeback *wb, enum wb_stat_item item)
 {
 	unsigned long flags;
 
 	local_irq_save(flags);
-	__dec_bdi_stat(bdi, item);
+	__inc_wb_stat(wb, item);
 	local_irq_restore(flags);
 }
 
-static inline s64 bdi_stat(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
+static inline void __dec_wb_stat(struct bdi_writeback *wb,
+				 enum wb_stat_item item)
 {
-	return percpu_counter_read_positive(&bdi->bdi_stat[item]);
+	__add_wb_stat(wb, item, -1);
 }
 
-static inline s64 __bdi_stat_sum(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
+static inline void dec_wb_stat(struct bdi_writeback *wb, enum wb_stat_item item)
 {
-	return percpu_counter_sum_positive(&bdi->bdi_stat[item]);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	__dec_wb_stat(wb, item);
+	local_irq_restore(flags);
 }
 
-static inline s64 bdi_stat_sum(struct backing_dev_info *bdi,
-		enum bdi_stat_item item)
+static inline s64 wb_stat(struct bdi_writeback *wb, enum wb_stat_item item)
+{
+	return percpu_counter_read_positive(&wb->stat[item]);
+}
+
+static inline s64 __wb_stat_sum(struct bdi_writeback *wb,
+				enum wb_stat_item item)
+{
+	return percpu_counter_sum_positive(&wb->stat[item]);
+}
+
+static inline s64 wb_stat_sum(struct bdi_writeback *wb, enum wb_stat_item item)
 {
 	s64 sum;
 	unsigned long flags;
 
 	local_irq_save(flags);
-	sum = __bdi_stat_sum(bdi, item);
+	sum = __wb_stat_sum(wb, item);
 	local_irq_restore(flags);
 
 	return sum;
 }
 
-extern void bdi_writeout_inc(struct backing_dev_info *bdi);
+extern void wb_writeout_inc(struct bdi_writeback *wb);
 
 /*
  * maximal error of a stat counter.
  */
-static inline unsigned long bdi_stat_error(struct backing_dev_info *bdi)
+static inline unsigned long wb_stat_error(struct bdi_writeback *wb)
 {
 #ifdef CONFIG_SMP
-	return nr_cpu_ids * BDI_STAT_BATCH;
+	return nr_cpu_ids * WB_STAT_BATCH;
 #else
 	return 1;
 #endif
···
  * BDI_CAP_NO_WRITEBACK:   Don't write pages back
  * BDI_CAP_NO_ACCT_WB:     Don't automatically account writeback pages
  * BDI_CAP_STRICTLIMIT:    Keep number of dirty pages below bdi threshold.
+ *
+ * BDI_CAP_CGROUP_WRITEBACK: Supports cgroup-aware writeback.
  */
 #define BDI_CAP_NO_ACCT_DIRTY	0x00000001
 #define BDI_CAP_NO_WRITEBACK	0x00000002
 #define BDI_CAP_NO_ACCT_WB	0x00000004
 #define BDI_CAP_STABLE_WRITES	0x00000008
 #define BDI_CAP_STRICTLIMIT	0x00000010
+#define BDI_CAP_CGROUP_WRITEBACK 0x00000020
 
 #define BDI_CAP_NO_ACCT_AND_WRITEBACK \
 	(BDI_CAP_NO_WRITEBACK | BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_ACCT_WB)
 
 extern struct backing_dev_info noop_backing_dev_info;
 
-int writeback_in_progress(struct backing_dev_info *bdi);
-
-static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits)
+/**
+ * writeback_in_progress - determine whether there is writeback in progress
+ * @wb: bdi_writeback of interest
+ *
+ * Determine whether there is writeback waiting to be handled against a
+ * bdi_writeback.
+ */
+static inline bool writeback_in_progress(struct bdi_writeback *wb)
 {
+	return test_bit(WB_writeback_running, &wb->state);
+}
+
+static inline struct backing_dev_info *inode_to_bdi(struct inode *inode)
+{
+	struct super_block *sb;
+
+	if (!inode)
+		return &noop_backing_dev_info;
+
+	sb = inode->i_sb;
+#ifdef CONFIG_BLOCK
+	if (sb_is_blkdev_sb(sb))
+		return blk_get_backing_dev_info(I_BDEV(inode));
+#endif
+	return sb->s_bdi;
+}
+
+static inline int wb_congested(struct bdi_writeback *wb, int cong_bits)
+{
+	struct backing_dev_info *bdi = wb->bdi;
+
 	if (bdi->congested_fn)
-		return bdi->congested_fn(bdi->congested_data, bdi_bits);
-	return (bdi->state & bdi_bits);
+		return bdi->congested_fn(bdi->congested_data, cong_bits);
+	return wb->congested->state & cong_bits;
 }
 
-static inline int bdi_read_congested(struct backing_dev_info *bdi)
-{
-	return bdi_congested(bdi, 1 << BDI_sync_congested);
-}
-
-static inline int bdi_write_congested(struct backing_dev_info *bdi)
-{
-	return bdi_congested(bdi, 1 << BDI_async_congested);
-}
-
-static inline int bdi_rw_congested(struct backing_dev_info *bdi)
-{
-	return bdi_congested(bdi, (1 << BDI_sync_congested) |
-				  (1 << BDI_async_congested));
-}
-
-enum {
-	BLK_RW_ASYNC	= 0,
-	BLK_RW_SYNC	= 1,
-};
-
-void clear_bdi_congested(struct backing_dev_info *bdi, int sync);
-void set_bdi_congested(struct backing_dev_info *bdi, int sync);
 long congestion_wait(int sync, long timeout);
 long wait_iff_congested(struct zone *zone, int sync, long timeout);
 int pdflush_proc_obsolete(struct ctl_table *table, int write,
···
 	return 0;
 }
 
-#endif		/* _LINUX_BACKING_DEV_H */
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+struct bdi_writeback_congested *
+wb_congested_get_create(struct backing_dev_info *bdi, int blkcg_id, gfp_t gfp);
+void wb_congested_put(struct bdi_writeback_congested *congested);
+struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi,
+				    struct cgroup_subsys_state *memcg_css,
+				    gfp_t gfp);
+void wb_memcg_offline(struct mem_cgroup *memcg);
+void wb_blkcg_offline(struct blkcg *blkcg);
+int inode_congested(struct inode *inode, int cong_bits);
+
+/**
+ * inode_cgwb_enabled - test whether cgroup writeback is enabled on an inode
+ * @inode: inode of interest
+ *
+ * cgroup writeback requires support from both the bdi and filesystem.
+ * Test whether @inode has both.
+ */
+static inline bool inode_cgwb_enabled(struct inode *inode)
+{
+	struct backing_dev_info *bdi = inode_to_bdi(inode);
+
+	return bdi_cap_account_dirty(bdi) &&
+		(bdi->capabilities & BDI_CAP_CGROUP_WRITEBACK) &&
+		(inode->i_sb->s_iflags & SB_I_CGROUPWB);
+}
+
+/**
+ * wb_find_current - find wb for %current on a bdi
+ * @bdi: bdi of interest
+ *
+ * Find the wb of @bdi which matches both the memcg and blkcg of %current.
+ * Must be called under rcu_read_lock() which protects the returend wb.
+ * NULL if not found.
+ */
+static inline struct bdi_writeback *wb_find_current(struct backing_dev_info *bdi)
+{
+	struct cgroup_subsys_state *memcg_css;
+	struct bdi_writeback *wb;
+
+	memcg_css = task_css(current, memory_cgrp_id);
+	if (!memcg_css->parent)
+		return &bdi->wb;
+
+	wb = radix_tree_lookup(&bdi->cgwb_tree, memcg_css->id);
+
+	/*
+	 * %current's blkcg equals the effective blkcg of its memcg.  No
+	 * need to use the relatively expensive cgroup_get_e_css().
+	 */
+	if (likely(wb && wb->blkcg_css == task_css(current, blkio_cgrp_id)))
+		return wb;
+	return NULL;
+}
+
+/**
+ * wb_get_create_current - get or create wb for %current on a bdi
+ * @bdi: bdi of interest
+ * @gfp: allocation mask
+ *
+ * Equivalent to wb_get_create() on %current's memcg.  This function is
+ * called from a relatively hot path and optimizes the common cases using
+ * wb_find_current().
+ */
+static inline struct bdi_writeback *
+wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp)
+{
+	struct bdi_writeback *wb;
+
+	rcu_read_lock();
+	wb = wb_find_current(bdi);
+	if (wb && unlikely(!wb_tryget(wb)))
+		wb = NULL;
+	rcu_read_unlock();
+
+	if (unlikely(!wb)) {
+		struct cgroup_subsys_state *memcg_css;
+
+		memcg_css = task_get_css(current, memory_cgrp_id);
+		wb = wb_get_create(bdi, memcg_css, gfp);
+		css_put(memcg_css);
+	}
+	return wb;
+}
+
+/**
+ * inode_to_wb_is_valid - test whether an inode has a wb associated
+ * @inode: inode of interest
+ *
+ * Returns %true if @inode has a wb associated.  May be called without any
+ * locking.
+ */
+static inline bool inode_to_wb_is_valid(struct inode *inode)
+{
+	return inode->i_wb;
+}
+
+/**
+ * inode_to_wb - determine the wb of an inode
+ * @inode: inode of interest
+ *
+ * Returns the wb @inode is currently associated with.  The caller must be
+ * holding either @inode->i_lock, @inode->i_mapping->tree_lock, or the
+ * associated wb's list_lock.
+ */
+static inline struct bdi_writeback *inode_to_wb(struct inode *inode)
+{
+#ifdef CONFIG_LOCKDEP
+	WARN_ON_ONCE(debug_locks &&
+		     (!lockdep_is_held(&inode->i_lock) &&
+		      !lockdep_is_held(&inode->i_mapping->tree_lock) &&
+		      !lockdep_is_held(&inode->i_wb->list_lock)));
+#endif
+	return inode->i_wb;
+}
+
+/**
+ * unlocked_inode_to_wb_begin - begin unlocked inode wb access transaction
+ * @inode: target inode
+ * @lockedp: temp bool output param, to be passed to the end function
+ *
+ * The caller wants to access the wb associated with @inode but isn't
+ * holding inode->i_lock, mapping->tree_lock or wb->list_lock.  This
+ * function determines the wb associated with @inode and ensures that the
+ * association doesn't change until the transaction is finished with
+ * unlocked_inode_to_wb_end().
+ *
+ * The caller must call unlocked_inode_to_wb_end() with *@lockdep
+ * afterwards and can't sleep during transaction.  IRQ may or may not be
+ * disabled on return.
+ */
+static inline struct bdi_writeback *
+unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp)
+{
+	rcu_read_lock();
+
+	/*
+	 * Paired with store_release in inode_switch_wb_work_fn() and
+	 * ensures that we see the new wb if we see cleared I_WB_SWITCH.
+	 */
+	*lockedp = smp_load_acquire(&inode->i_state) & I_WB_SWITCH;
+
+	if (unlikely(*lockedp))
+		spin_lock_irq(&inode->i_mapping->tree_lock);
+
+	/*
+	 * Protected by either !I_WB_SWITCH + rcu_read_lock() or tree_lock.
+	 * inode_to_wb() will bark.  Deref directly.
+	 */
+	return inode->i_wb;
+}
+
+/**
+ * unlocked_inode_to_wb_end - end inode wb access transaction
+ * @inode: target inode
+ * @locked: *@lockedp from unlocked_inode_to_wb_begin()
+ */
+static inline void unlocked_inode_to_wb_end(struct inode *inode, bool locked)
+{
+	if (unlikely(locked))
+		spin_unlock_irq(&inode->i_mapping->tree_lock);
+
+	rcu_read_unlock();
+}
+
+struct wb_iter {
+	int			start_blkcg_id;
+	struct radix_tree_iter	tree_iter;
+	void			**slot;
+};
+
+static inline struct bdi_writeback *__wb_iter_next(struct wb_iter *iter,
+						   struct backing_dev_info *bdi)
+{
+	struct radix_tree_iter *titer = &iter->tree_iter;
+
+	WARN_ON_ONCE(!rcu_read_lock_held());
+
+	if (iter->start_blkcg_id >= 0) {
+		iter->slot = radix_tree_iter_init(titer, iter->start_blkcg_id);
+		iter->start_blkcg_id = -1;
+	} else {
+		iter->slot = radix_tree_next_slot(iter->slot, titer, 0);
+	}
+
+	if (!iter->slot)
+		iter->slot = radix_tree_next_chunk(&bdi->cgwb_tree, titer, 0);
+	if (iter->slot)
+		return *iter->slot;
+	return NULL;
+}
+
+static inline struct bdi_writeback *__wb_iter_init(struct wb_iter *iter,
+						   struct backing_dev_info *bdi,
+						   int start_blkcg_id)
+{
+	iter->start_blkcg_id = start_blkcg_id;
+
+	if (start_blkcg_id)
+		return __wb_iter_next(iter, bdi);
+	else
+		return &bdi->wb;
+}
+
+/**
+ * bdi_for_each_wb - walk all wb's of a bdi in ascending blkcg ID order
+ * @wb_cur: cursor struct bdi_writeback pointer
+ * @bdi: bdi to walk wb's of
+ * @iter: pointer to struct wb_iter to be used as iteration buffer
+ * @start_blkcg_id: blkcg ID to start iteration from
+ *
+ * Iterate @wb_cur through the wb's (bdi_writeback's) of @bdi in ascending
+ * blkcg ID order starting from @start_blkcg_id.  @iter is struct wb_iter
+ * to be used as temp storage during iteration.  rcu_read_lock() must be
+ * held throughout iteration.
+ */
+#define bdi_for_each_wb(wb_cur, bdi, iter, start_blkcg_id)		\
+	for ((wb_cur) = __wb_iter_init(iter, bdi, start_blkcg_id);	\
+	     (wb_cur); (wb_cur) = __wb_iter_next(iter, bdi))
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static inline bool inode_cgwb_enabled(struct inode *inode)
+{
+	return false;
+}
+
+static inline struct bdi_writeback_congested *
+wb_congested_get_create(struct backing_dev_info *bdi, int blkcg_id, gfp_t gfp)
+{
+	return bdi->wb.congested;
+}
+
+static inline void wb_congested_put(struct bdi_writeback_congested *congested)
+{
+}
+
+static inline struct bdi_writeback *wb_find_current(struct backing_dev_info *bdi)
+{
+	return &bdi->wb;
+}
+
+static inline struct bdi_writeback *
+wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp)
+{
+	return &bdi->wb;
+}
+
+static inline bool inode_to_wb_is_valid(struct inode *inode)
+{
+	return true;
+}
+
+static inline struct bdi_writeback *inode_to_wb(struct inode *inode)
+{
+	return &inode_to_bdi(inode)->wb;
+}
+
+static inline struct bdi_writeback *
+unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp)
+{
+	return inode_to_wb(inode);
+}
+
+static inline void unlocked_inode_to_wb_end(struct inode *inode, bool locked)
+{
+}
+
+static inline void wb_memcg_offline(struct mem_cgroup *memcg)
+{
+}
+
+static inline void wb_blkcg_offline(struct blkcg *blkcg)
+{
+}
+
+struct wb_iter {
+	int		next_id;
+};
+
+#define bdi_for_each_wb(wb_cur, bdi, iter, start_blkcg_id)		\
+	for ((iter)->next_id = (start_blkcg_id);			\
+	     ({	(wb_cur) = !(iter)->next_id++ ? &(bdi)->wb : NULL; }); )
+
+static inline int inode_congested(struct inode *inode, int cong_bits)
+{
+	return wb_congested(&inode_to_bdi(inode)->wb, cong_bits);
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
+
+static inline int inode_read_congested(struct inode *inode)
+{
+	return inode_congested(inode, 1 << WB_sync_congested);
+}
+
+static inline int inode_write_congested(struct inode *inode)
+{
+	return inode_congested(inode, 1 << WB_async_congested);
+}
+
+static inline int inode_rw_congested(struct inode *inode)
+{
+	return inode_congested(inode, (1 << WB_sync_congested) |
+				      (1 << WB_async_congested));
+}
+
+static inline int bdi_congested(struct backing_dev_info *bdi, int cong_bits)
+{
+	return wb_congested(&bdi->wb, cong_bits);
+}
+
+static inline int bdi_read_congested(struct backing_dev_info *bdi)
+{
+	return bdi_congested(bdi, 1 << WB_sync_congested);
+}
+
+static inline int bdi_write_congested(struct backing_dev_info *bdi)
+{
+	return bdi_congested(bdi, 1 << WB_async_congested);
+}
+
+static inline int bdi_rw_congested(struct backing_dev_info *bdi)
+{
+	return bdi_congested(bdi, (1 << WB_sync_congested) |
+				  (1 << WB_async_congested));
+}
+
+#endif	/* _LINUX_BACKING_DEV_H */

include/linux/bio.h

···
 extern unsigned int bvec_nr_vecs(unsigned short idx);
 
 #ifdef CONFIG_BLK_CGROUP
+int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css);
 int bio_associate_current(struct bio *bio);
 void bio_disassociate_task(struct bio *bio);
 #else	/* CONFIG_BLK_CGROUP */
+static inline int bio_associate_blkcg(struct bio *bio,
+			struct cgroup_subsys_state *blkcg_css) { return 0; }
 static inline int bio_associate_current(struct bio *bio) { return -ENOENT; }
 static inline void bio_disassociate_task(struct bio *bio) { }
 #endif	/* CONFIG_BLK_CGROUP */

include/linux/blkdev.h

···
 #include <linux/timer.h>
 #include <linux/workqueue.h>
 #include <linux/pagemap.h>
-#include <linux/backing-dev.h>
+#include <linux/backing-dev-defs.h>
 #include <linux/wait.h>
 #include <linux/mempool.h>
 #include <linux/bio.h>
···
 			  unsigned int, void __user *);
 extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
 			 struct scsi_ioctl_command __user *);
-
-/*
- * A queue has just exitted congestion.  Note this in the global counter of
- * congested queues, and wake up anyone who was waiting for requests to be
- * put back.
- */
-static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
-{
-	clear_bdi_congested(&q->backing_dev_info, sync);
-}
-
-/*
- * A queue has just entered congestion.  Flag that in the queue's VM-visible
- * state flags and increment the global gounter of congested queues.
- */
-static inline void blk_set_queue_congested(struct request_queue *q, int sync)
-{
-	set_bdi_congested(&q->backing_dev_info, sync);
-}
 
 extern void blk_start_queue(struct request_queue *q);
 extern void blk_stop_queue(struct request_queue *q);

include/linux/cgroup.h

···
 }
 
 /**
+ * task_get_css - find and get the css for (task, subsys)
+ * @task: the target task
+ * @subsys_id: the target subsystem ID
+ *
+ * Find the css for the (@task, @subsys_id) combination, increment a
+ * reference on and return it.  This function is guaranteed to return a
+ * valid css.
+ */
+static inline struct cgroup_subsys_state *
+task_get_css(struct task_struct *task, int subsys_id)
+{
+	struct cgroup_subsys_state *css;
+
+	rcu_read_lock();
+	while (true) {
+		css = task_css(task, subsys_id);
+		if (likely(css_tryget_online(css)))
+			break;
+		cpu_relax();
+	}
+	rcu_read_unlock();
+	return css;
+}
+
+/**
  * task_css_is_root - test whether a task belongs to the root css
  * @task: the target task
  * @subsys_id: the target subsystem ID

include/linux/fs.h

···
 #include <uapi/linux/fs.h>
 
 struct backing_dev_info;
+struct bdi_writeback;
 struct export_operations;
 struct hd_geometry;
 struct iovec;
···
 
 	struct hlist_node	i_hash;
 	struct list_head	i_wb_list;	/* backing dev IO list */
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct bdi_writeback	*i_wb;		/* the associated cgroup wb */
+
+	/* foreign inode detection, see wbc_detach_inode() */
+	int			i_wb_frn_winner;
+	u16			i_wb_frn_avg_time;
+	u16			i_wb_frn_history;
+#endif
 	struct list_head	i_lru;		/* inode LRU list */
 	struct list_head	i_sb_list;
 	union {
···
 #define UMOUNT_NOFOLLOW	0x00000008	/* Don't follow symlink on umount */
 #define UMOUNT_UNUSED	0x80000000	/* Flag guaranteed to be unused */
 
+/* sb->s_iflags */
+#define SB_I_CGROUPWB	0x00000001	/* cgroup-aware writeback enabled */
 
 /* Possible states of 'frozen' field */
 enum {
···
 	const struct quotactl_ops	*s_qcop;
 	const struct export_operations *s_export_op;
 	unsigned long		s_flags;
+	unsigned long		s_iflags;	/* internal SB_I_* flags */
 	unsigned long		s_magic;
 	struct dentry		*s_root;
 	struct rw_semaphore	s_umount;
···
  *
  * I_DIO_WAKEUP		Never set.  Only used as a key for wait_on_bit().
  *
+ * I_WB_SWITCH		Cgroup bdi_writeback switching in progress.  Used to
+ *			synchronize competing switching instances and to tell
+ *			wb stat updates to grab mapping->tree_lock.  See
+ *			inode_switch_wb_work_fn() for details.
+ *
  * Q: What is the difference between I_WILL_FREE and I_FREEING?
  */
 #define I_DIRTY_SYNC		(1 << 0)
···
 #define I_DIRTY_TIME		(1 << 11)
 #define __I_DIRTY_TIME_EXPIRED	12
 #define I_DIRTY_TIME_EXPIRED	(1 << __I_DIRTY_TIME_EXPIRED)
+#define I_WB_SWITCH		(1 << 13)
 
 #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES)
 #define I_DIRTY_ALL (I_DIRTY | I_DIRTY_TIME)
···
 extern void emergency_thaw_all(void);
 extern int thaw_bdev(struct block_device *bdev, struct super_block *sb);
 extern int fsync_bdev(struct block_device *);
-extern int sb_is_blkdev_sb(struct super_block *sb);
+
+extern struct super_block *blockdev_superblock;
+
+static inline bool sb_is_blkdev_sb(struct super_block *sb)
+{
+	return sb == blockdev_superblock;
+}
 #else
 static inline void bd_forget(struct inode *inode) {}
 static inline int sync_blockdev(struct block_device *bdev) { return 0; }

include/linux/memcontrol.h

···
 	MEM_CGROUP_STAT_RSS,		/* # of pages charged as anon rss */
 	MEM_CGROUP_STAT_RSS_HUGE,	/* # of pages charged as anon huge */
 	MEM_CGROUP_STAT_FILE_MAPPED,	/* # of pages charged as file rss */
+	MEM_CGROUP_STAT_DIRTY,          /* # of dirty pages in page cache */
 	MEM_CGROUP_STAT_WRITEBACK,	/* # of pages under writeback */
 	MEM_CGROUP_STAT_SWAP,		/* # of pages, swapped out */
 	MEM_CGROUP_STAT_NSTATS,
···
 };
 
 #ifdef CONFIG_MEMCG
+extern struct cgroup_subsys_state *mem_cgroup_root_css;
+
 void mem_cgroup_events(struct mem_cgroup *memcg,
 		       enum mem_cgroup_events_index idx,
 		       unsigned int nr);
···
 }
 
 extern struct cgroup_subsys_state *mem_cgroup_css(struct mem_cgroup *memcg);
+extern struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page);
 
 struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *,
 				   struct mem_cgroup *,
···
 
 #else /* CONFIG_MEMCG */
 struct mem_cgroup;
+
+#define mem_cgroup_root_css ((struct cgroup_subsys_state *)ERR_PTR(-EINVAL))
 
 static inline void mem_cgroup_events(struct mem_cgroup *memcg,
 				     enum mem_cgroup_events_index idx,
···
 	SOFT_LIMIT,
 	OVER_LIMIT,
 };
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg);
+struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb);
+void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pavail,
+			 unsigned long *pdirty, unsigned long *pwriteback);
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static inline struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb)
+{
+	return NULL;
+}
+
+static inline void mem_cgroup_wb_stats(struct bdi_writeback *wb,
+				       unsigned long *pavail,
+				       unsigned long *pdirty,
+				       unsigned long *pwriteback)
+{
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
 
 struct sock;
 #if defined(CONFIG_INET) && defined(CONFIG_MEMCG_KMEM)

include/linux/mm.h

···
 struct file_ra_state;
 struct user_struct;
 struct writeback_control;
+struct bdi_writeback;
 
 #ifndef CONFIG_NEED_MULTIPLE_NODES	/* Don't use mapnrs, do it properly */
 extern unsigned long max_mapnr;
···
 int __set_page_dirty_no_writeback(struct page *page);
 int redirty_page_for_writepage(struct writeback_control *wbc,
 				struct page *page);
-void account_page_dirtied(struct page *page, struct address_space *mapping);
-void account_page_cleaned(struct page *page, struct address_space *mapping);
+void account_page_dirtied(struct page *page, struct address_space *mapping,
+			  struct mem_cgroup *memcg);
+void account_page_cleaned(struct page *page, struct address_space *mapping,
+			  struct mem_cgroup *memcg, struct bdi_writeback *wb);
 int set_page_dirty(struct page *page);
 int set_page_dirty_lock(struct page *page);
+void cancel_dirty_page(struct page *page);
 int clear_page_dirty_for_io(struct page *page);
 
 int get_cmdline(struct task_struct *task, char *buffer, int buflen);

include/linux/pagemap.h

···
 int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
 extern void delete_from_page_cache(struct page *page);
-extern void __delete_from_page_cache(struct page *page, void *shadow);
+extern void __delete_from_page_cache(struct page *page, void *shadow,
+				     struct mem_cgroup *memcg);
 int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask);
 
 /*

include/linux/writeback.h

···
 #include <linux/sched.h>
 #include <linux/workqueue.h>
 #include <linux/fs.h>
+#include <linux/flex_proportions.h>
+#include <linux/backing-dev-defs.h>
 
 DECLARE_PER_CPU(int, dirty_throttle_leaks);
 
···
 	unsigned for_reclaim:1;		/* Invoked from the page allocator */
 	unsigned range_cyclic:1;	/* range_start is cyclic */
 	unsigned for_sync:1;		/* sync(2) WB_SYNC_ALL writeback */
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct bdi_writeback *wb;	/* wb this writeback is issued under */
+	struct inode *inode;		/* inode being written out */
+
+	/* foreign inode detection, see wbc_detach_inode() */
+	int wb_id;			/* current wb id */
+	int wb_lcand_id;		/* last foreign candidate wb id */
+	int wb_tcand_id;		/* this foreign candidate wb id */
+	size_t wb_bytes;		/* bytes written by current wb */
+	size_t wb_lcand_bytes;		/* bytes written by last candidate */
+	size_t wb_tcand_bytes;		/* bytes written by this candidate */
+#endif
 };
+
+/*
+ * A wb_domain represents a domain that wb's (bdi_writeback's) belong to
+ * and are measured against each other in.  There always is one global
+ * domain, global_wb_domain, that every wb in the system is a member of.
+ * This allows measuring the relative bandwidth of each wb to distribute
+ * dirtyable memory accordingly.
+ */
+struct wb_domain {
+	spinlock_t lock;
+
+	/*
+	 * Scale the writeback cache size proportional to the relative
+	 * writeout speed.
+	 *
+	 * We do this by keeping a floating proportion between BDIs, based
+	 * on page writeback completions [end_page_writeback()]. Those
+	 * devices that write out pages fastest will get the larger share,
+	 * while the slower will get a smaller share.
+	 *
+	 * We use page writeout completions because we are interested in
+	 * getting rid of dirty pages. Having them written out is the
+	 * primary goal.
+	 *
+	 * We introduce a concept of time, a period over which we measure
+	 * these events, because demand can/will vary over time. The length
+	 * of this period itself is measured in page writeback completions.
+	 */
+	struct fprop_global completions;
+	struct timer_list period_timer;	/* timer for aging of completions */
+	unsigned long period_time;
+
+	/*
+	 * The dirtyable memory and dirty threshold could be suddenly
+	 * knocked down by a large amount (eg. on the startup of KVM in a
+	 * swapless system). This may throw the system into deep dirty
+	 * exceeded state and throttle heavy/light dirtiers alike. To
+	 * retain good responsiveness, maintain global_dirty_limit for
+	 * tracking slowly down to the knocked down dirty threshold.
+	 *
+	 * Both fields are protected by ->lock.
+	 */
+	unsigned long dirty_limit_tstamp;
+	unsigned long dirty_limit;
+};
+
+/**
+ * wb_domain_size_changed - memory available to a wb_domain has changed
+ * @dom: wb_domain of interest
+ *
+ * This function should be called when the amount of memory available to
+ * @dom has changed.  It resets @dom's dirty limit parameters to prevent
+ * the past values which don't match the current configuration from skewing
+ * dirty throttling.  Without this, when memory size of a wb_domain is
+ * greatly reduced, the dirty throttling logic may allow too many pages to
+ * be dirtied leading to consecutive unnecessary OOMs and may get stuck in
+ * that situation.
+ */
+static inline void wb_domain_size_changed(struct wb_domain *dom)
+{
+	spin_lock(&dom->lock);
+	dom->dirty_limit_tstamp = jiffies;
+	dom->dirty_limit = 0;
+	spin_unlock(&dom->lock);
+}
 
 /*
  * fs/fs-writeback.c
···
 void writeback_inodes_sb(struct super_block *, enum wb_reason reason);
 void writeback_inodes_sb_nr(struct super_block *, unsigned long nr,
 							enum wb_reason reason);
-int try_to_writeback_inodes_sb(struct super_block *, enum wb_reason reason);
-int try_to_writeback_inodes_sb_nr(struct super_block *, unsigned long nr,
-				  enum wb_reason reason);
+bool try_to_writeback_inodes_sb(struct super_block *, enum wb_reason reason);
+bool try_to_writeback_inodes_sb_nr(struct super_block *, unsigned long nr,
+				   enum wb_reason reason);
 void sync_inodes_sb(struct super_block *);
 void wakeup_flusher_threads(long nr_pages, enum wb_reason reason);
 void inode_wait_for_writeback(struct inode *inode);
···
 	might_sleep();
 	wait_on_bit(&inode->i_state, __I_NEW, TASK_UNINTERRUPTIBLE);
 }
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+#include <linux/cgroup.h>
+#include <linux/bio.h>
+
+void __inode_attach_wb(struct inode *inode, struct page *page);
+void wbc_attach_and_unlock_inode(struct writeback_control *wbc,
+				 struct inode *inode)
+	__releases(&inode->i_lock);
+void wbc_detach_inode(struct writeback_control *wbc);
+void wbc_account_io(struct writeback_control *wbc, struct page *page,
+		    size_t bytes);
+
+/**
+ * inode_attach_wb - associate an inode with its wb
+ * @inode: inode of interest
+ * @page: page being dirtied (may be NULL)
+ *
+ * If @inode doesn't have its wb, associate it with the wb matching the
+ * memcg of @page or, if @page is NULL, %current.  May be called w/ or w/o
+ * @inode->i_lock.
+ */
+static inline void inode_attach_wb(struct inode *inode, struct page *page)
+{
+	if (!inode->i_wb)
+		__inode_attach_wb(inode, page);
+}
+
+/**
+ * inode_detach_wb - disassociate an inode from its wb
+ * @inode: inode of interest
+ *
+ * @inode is being freed.  Detach from its wb.
+ */
+static inline void inode_detach_wb(struct inode *inode)
+{
+	if (inode->i_wb) {
+		wb_put(inode->i_wb);
+		inode->i_wb = NULL;
+	}
+}
+
+/**
+ * wbc_attach_fdatawrite_inode - associate wbc and inode for fdatawrite
+ * @wbc: writeback_control of interest
+ * @inode: target inode
+ *
+ * This function is to be used by __filemap_fdatawrite_range(), which is an
+ * alternative entry point into writeback code, and first ensures @inode is
+ * associated with a bdi_writeback and attaches it to @wbc.
+ */
+static inline void wbc_attach_fdatawrite_inode(struct writeback_control *wbc,
+					       struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	inode_attach_wb(inode, NULL);
+	wbc_attach_and_unlock_inode(wbc, inode);
+}
+
+/**
+ * wbc_init_bio - writeback specific initializtion of bio
+ * @wbc: writeback_control for the writeback in progress
+ * @bio: bio to be initialized
+ *
+ * @bio is a part of the writeback in progress controlled by @wbc.  Perform
+ * writeback specific initialization.  This is used to apply the cgroup
+ * writeback context.
+ */
+static inline void wbc_init_bio(struct writeback_control *wbc, struct bio *bio)
+{
+	/*
+	 * pageout() path doesn't attach @wbc to the inode being written
+	 * out.  This is intentional as we don't want the function to block
+	 * behind a slow cgroup.  Ultimately, we want pageout() to kick off
+	 * regular writeback instead of writing things out itself.
+	 */
+	if (wbc->wb)
+		bio_associate_blkcg(bio, wbc->wb->blkcg_css);
+}
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static inline void inode_attach_wb(struct inode *inode, struct page *page)
+{
+}
+
+static inline void inode_detach_wb(struct inode *inode)
+{
+}
+
+static inline void wbc_attach_and_unlock_inode(struct writeback_control *wbc,
+					       struct inode *inode)
+	__releases(&inode->i_lock)
+{
+	spin_unlock(&inode->i_lock);
+}
+
+static inline void wbc_attach_fdatawrite_inode(struct writeback_control *wbc,
+					       struct inode *inode)
+{
+}
+
+static inline void wbc_detach_inode(struct writeback_control *wbc)
+{
+}
+
+static inline void wbc_init_bio(struct writeback_control *wbc, struct bio *bio)
+{
+}
+
+static inline void wbc_account_io(struct writeback_control *wbc,
+				  struct page *page, size_t bytes)
+{
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
 
 /*
  * mm/page-writeback.c
···
 #endif
 void throttle_vm_writeout(gfp_t gfp_mask);
 bool zone_dirty_ok(struct zone *zone);
+int wb_domain_init(struct wb_domain *dom, gfp_t gfp);
+#ifdef CONFIG_CGROUP_WRITEBACK
+void wb_domain_exit(struct wb_domain *dom);
+#endif
 
-extern unsigned long global_dirty_limit;
+extern struct wb_domain global_wb_domain;
 
 /* These are exported to sysctl. */
 extern int dirty_background_ratio;
···
 				      void __user *, size_t *, loff_t *);
 
 void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty);
-unsigned long bdi_dirty_limit(struct backing_dev_info *bdi,
-			       unsigned long dirty);
+unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh);
 
-void __bdi_update_bandwidth(struct backing_dev_info *bdi,
-			    unsigned long thresh,
-			    unsigned long bg_thresh,
-			    unsigned long dirty,
-			    unsigned long bdi_thresh,
-			    unsigned long bdi_dirty,
-			    unsigned long start_time);
-
+void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time);
 void page_writeback_init(void);
 void balance_dirty_pages_ratelimited(struct address_space *mapping);
+bool wb_over_bg_thresh(struct bdi_writeback *wb);
 
 typedef int (*writepage_t)(struct page *page, struct writeback_control *wbc,
 				void *data);

include/trace/events/writeback.h

···
 		__entry->nr_written	= global_page_state(NR_WRITTEN);
 		__entry->background_thresh = background_thresh;
 		__entry->dirty_thresh	= dirty_thresh;
-		__entry->dirty_limit = global_dirty_limit;
+		__entry->dirty_limit	= global_wb_domain.dirty_limit;
 	),
 
 	TP_printk("dirty=%lu writeback=%lu unstable=%lu "
···
 
 	TP_fast_assign(
 		strlcpy(__entry->bdi, dev_name(bdi->dev), 32);
-		__entry->write_bw	= KBps(bdi->write_bandwidth);
-		__entry->avg_write_bw	= KBps(bdi->avg_write_bandwidth);
+		__entry->write_bw	= KBps(bdi->wb.write_bandwidth);
+		__entry->avg_write_bw	= KBps(bdi->wb.avg_write_bandwidth);
 		__entry->dirty_rate	= KBps(dirty_rate);
-		__entry->dirty_ratelimit = KBps(bdi->dirty_ratelimit);
+		__entry->dirty_ratelimit = KBps(bdi->wb.dirty_ratelimit);
 		__entry->task_ratelimit	= KBps(task_ratelimit);
 		__entry->balanced_dirty_ratelimit =
-					  KBps(bdi->balanced_dirty_ratelimit);
+					KBps(bdi->wb.balanced_dirty_ratelimit);
 	),
 
 	TP_printk("bdi %s: "
···
 		unsigned long freerun = (thresh + bg_thresh) / 2;
 		strlcpy(__entry->bdi, dev_name(bdi->dev), 32);
 
-		__entry->limit		= global_dirty_limit;
-		__entry->setpoint	= (global_dirty_limit + freerun) / 2;
+		__entry->limit		= global_wb_domain.dirty_limit;
+		__entry->setpoint	= (global_wb_domain.dirty_limit +
+						freerun) / 2;
 		__entry->dirty		= dirty;
 		__entry->bdi_setpoint	= __entry->setpoint *
 						bdi_thresh / (thresh + 1);

init/Kconfig

···
 	Enable some debugging help. Currently it exports additional stat
 	files in a cgroup which can be useful for debugging.
 
+config CGROUP_WRITEBACK
+	bool
+	depends on MEMCG && BLK_CGROUP
+	default y
+
 endif # CGROUPS
 
 config CHECKPOINT_RESTORE

mm/backing-dev.c

···
 	.name		= "noop",
 	.capabilities	= BDI_CAP_NO_ACCT_AND_WRITEBACK,
 };
+EXPORT_SYMBOL_GPL(noop_backing_dev_info);
 
 static struct class *bdi_class;
 
···
 	struct bdi_writeback *wb = &bdi->wb;
 	unsigned long background_thresh;
 	unsigned long dirty_thresh;
-	unsigned long bdi_thresh;
+	unsigned long wb_thresh;
 	unsigned long nr_dirty, nr_io, nr_more_io, nr_dirty_time;
 	struct inode *inode;
 
···
 	spin_unlock(&wb->list_lock);
 
 	global_dirty_limits(&background_thresh, &dirty_thresh);
-	bdi_thresh = bdi_dirty_limit(bdi, dirty_thresh);
+	wb_thresh = wb_calc_thresh(wb, dirty_thresh);
 
 #define K(x) ((x) << (PAGE_SHIFT - 10))
 	seq_printf(m,
···
 		   "b_dirty_time:       %10lu\n"
 		   "bdi_list:           %10u\n"
 		   "state:              %10lx\n",
-		   (unsigned long) K(bdi_stat(bdi, BDI_WRITEBACK)),
-		   (unsigned long) K(bdi_stat(bdi, BDI_RECLAIMABLE)),
-		   K(bdi_thresh),
+		   (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(bdi_stat(bdi, BDI_DIRTIED)),
-		   (unsigned long) K(bdi_stat(bdi, BDI_WRITTEN)),
-		   (unsigned long) K(bdi->write_bandwidth),
+		   (unsigned long) K(wb_stat(wb, WB_DIRTIED)),
+		   (unsigned long) K(wb_stat(wb, WB_WRITTEN)),
+		   (unsigned long) K(wb->write_bandwidth),
 		   nr_dirty,
 		   nr_io,
 		   nr_more_io,
 		   nr_dirty_time,
-		   !list_empty(&bdi->bdi_list), bdi->state);
+		   !list_empty(&bdi->bdi_list), bdi->wb.state);
 #undef K
 
 	return 0;
···
 }
 subsys_initcall(default_bdi_init);
 
-int bdi_has_dirty_io(struct backing_dev_info *bdi)
-{
-	return wb_has_dirty_io(&bdi->wb);
-}
-
 /*
- * This function is used when the first inode for this bdi is marked dirty. It
+ * 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
···
  * We have to be careful not to postpone flush work if it is scheduled for
  * earlier. Thus we use queue_delayed_work().
  */
-void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi)
+void wb_wakeup_delayed(struct bdi_writeback *wb)
 {
 	unsigned long timeout;
 
 	timeout = msecs_to_jiffies(dirty_writeback_interval * 10);
-	spin_lock_bh(&bdi->wb_lock);
-	if (test_bit(BDI_registered, &bdi->state))
-		queue_delayed_work(bdi_wq, &bdi->wb.dwork, timeout);
-	spin_unlock_bh(&bdi->wb_lock);
+	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);
 }
 
 /*
- * Remove bdi from bdi_list, and ensure that it is no longer visible
+ * Initial write bandwidth: 100 MB/s
  */
-static void bdi_remove_from_list(struct backing_dev_info *bdi)
-{
-	spin_lock_bh(&bdi_lock);
-	list_del_rcu(&bdi->bdi_list);
-	spin_unlock_bh(&bdi_lock);
+#define INIT_BW		(100 << (20 - PAGE_SHIFT))
 
-	synchronize_rcu_expedited();
+static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi,
+		   gfp_t gfp)
+{
+	int i, err;
+
+	memset(wb, 0, sizeof(*wb));
+
+	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);
+	spin_lock_init(&wb->list_lock);
+
+	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;
+
+	spin_lock_init(&wb->work_lock);
+	INIT_LIST_HEAD(&wb->work_list);
+	INIT_DELAYED_WORK(&wb->dwork, wb_workfn);
+
+	err = fprop_local_init_percpu(&wb->completions, gfp);
+	if (err)
+		return err;
+
+	for (i = 0; i < NR_WB_STAT_ITEMS; i++) {
+		err = percpu_counter_init(&wb->stat[i], 0, gfp);
+		if (err) {
+			while (--i)
+				percpu_counter_destroy(&wb->stat[i]);
+			fprop_local_destroy_percpu(&wb->completions);
+			return err;
+		}
+	}
+
+	return 0;
 }
+
+/*
+ * 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);
+		return;
+	}
+	spin_unlock_bh(&wb->work_lock);
+
+	/*
+	 * 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.
+	 */
+	mod_delayed_work(bdi_wq, &wb->dwork, 0);
+	flush_delayed_work(&wb->dwork);
+	WARN_ON(!list_empty(&wb->work_list));
+}
+
+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]);
+
+	fprop_local_destroy_percpu(&wb->completions);
+}
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+#include <linux/memcontrol.h>
+
+/*
+ * cgwb_lock protects bdi->cgwb_tree, bdi->cgwb_congested_tree,
+ * blkcg->cgwb_list, and memcg->cgwb_list.  bdi->cgwb_tree is also RCU
+ * protected.  cgwb_release_wait is used to wait for the completion of cgwb
+ * releases from bdi destruction path.
+ */
+static DEFINE_SPINLOCK(cgwb_lock);
+static DECLARE_WAIT_QUEUE_HEAD(cgwb_release_wait);
+
+/**
+ * wb_congested_get_create - get or create a wb_congested
+ * @bdi: associated bdi
+ * @blkcg_id: ID of the associated blkcg
+ * @gfp: allocation mask
+ *
+ * Look up the wb_congested for @blkcg_id on @bdi.  If missing, create one.
+ * The returned wb_congested has its reference count incremented.  Returns
+ * NULL on failure.
+ */
+struct bdi_writeback_congested *
+wb_congested_get_create(struct backing_dev_info *bdi, int blkcg_id, gfp_t gfp)
+{
+	struct bdi_writeback_congested *new_congested = NULL, *congested;
+	struct rb_node **node, *parent;
+	unsigned long flags;
+
+	if (blkcg_id == 1)
+		return &bdi->wb_congested;
+retry:
+	spin_lock_irqsave(&cgwb_lock, flags);
+
+	node = &bdi->cgwb_congested_tree.rb_node;
+	parent = NULL;
+
+	while (*node != NULL) {
+		parent = *node;
+		congested = container_of(parent, struct bdi_writeback_congested,
+					 rb_node);
+		if (congested->blkcg_id < blkcg_id)
+			node = &parent->rb_left;
+		else if (congested->blkcg_id > blkcg_id)
+			node = &parent->rb_right;
+		else
+			goto found;
+	}
+
+	if (new_congested) {
+		/* !found and storage for new one already allocated, insert */
+		congested = new_congested;
+		new_congested = NULL;
+		rb_link_node(&congested->rb_node, parent, node);
+		rb_insert_color(&congested->rb_node, &bdi->cgwb_congested_tree);
+		atomic_inc(&bdi->usage_cnt);
+		goto found;
+	}
+
+	spin_unlock_irqrestore(&cgwb_lock, flags);
+
+	/* allocate storage for new one and retry */
+	new_congested = kzalloc(sizeof(*new_congested), gfp);
+	if (!new_congested)
+		return NULL;
+
+	atomic_set(&new_congested->refcnt, 0);
+	new_congested->bdi = bdi;
+	new_congested->blkcg_id = blkcg_id;
+	goto retry;
+
+found:
+	atomic_inc(&congested->refcnt);
+	spin_unlock_irqrestore(&cgwb_lock, flags);
+	kfree(new_congested);
+	return congested;
+}
+
+/**
+ * wb_congested_put - put a wb_congested
+ * @congested: wb_congested to put
+ *
+ * Put @congested and destroy it if the refcnt reaches zero.
+ */
+void wb_congested_put(struct bdi_writeback_congested *congested)
+{
+	struct backing_dev_info *bdi = congested->bdi;
+	unsigned long flags;
+
+	if (congested->blkcg_id == 1)
+		return;
+
+	local_irq_save(flags);
+	if (!atomic_dec_and_lock(&congested->refcnt, &cgwb_lock)) {
+		local_irq_restore(flags);
+		return;
+	}
+
+	rb_erase(&congested->rb_node, &congested->bdi->cgwb_congested_tree);
+	spin_unlock_irqrestore(&cgwb_lock, flags);
+	kfree(congested);
+
+	if (atomic_dec_and_test(&bdi->usage_cnt))
+		wake_up_all(&cgwb_release_wait);
+}
+
+static void cgwb_release_workfn(struct work_struct *work)
+{
+	struct bdi_writeback *wb = container_of(work, struct bdi_writeback,
+						release_work);
+	struct backing_dev_info *bdi = wb->bdi;
+
+	wb_shutdown(wb);
+
+	css_put(wb->memcg_css);
+	css_put(wb->blkcg_css);
+	wb_congested_put(wb->congested);
+
+	fprop_local_destroy_percpu(&wb->memcg_completions);
+	percpu_ref_exit(&wb->refcnt);
+	wb_exit(wb);
+	kfree_rcu(wb, rcu);
+
+	if (atomic_dec_and_test(&bdi->usage_cnt))
+		wake_up_all(&cgwb_release_wait);
+}
+
+static void cgwb_release(struct percpu_ref *refcnt)
+{
+	struct bdi_writeback *wb = container_of(refcnt, struct bdi_writeback,
+						refcnt);
+	schedule_work(&wb->release_work);
+}
+
+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);
+	percpu_ref_kill(&wb->refcnt);
+}
+
+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_css = cgroup_get_e_css(memcg_css->cgroup, &blkio_cgrp_subsys);
+	blkcg = css_to_blkcg(blkcg_css);
+	memcg_cgwb_list = mem_cgroup_cgwb_list(memcg);
+	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)
+		return -ENOMEM;
+
+	ret = wb_init(wb, bdi, gfp);
+	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;
+
+	wb->congested = wb_congested_get_create(bdi, blkcg_css->id, gfp);
+	if (!wb->congested) {
+		ret = -ENOMEM;
+		goto err_fprop_exit;
+	}
+
+	wb->memcg_css = memcg_css;
+	wb->blkcg_css = blkcg_css;
+	INIT_WORK(&wb->release_work, cgwb_release_workfn);
+	set_bit(WB_registered, &wb->state);
+
+	/*
+	 * 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().
+	 */
+	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) {
+			atomic_inc(&bdi->usage_cnt);
+			list_add(&wb->memcg_node, memcg_cgwb_list);
+			list_add(&wb->blkcg_node, blkcg_cgwb_list);
+			css_get(memcg_css);
+			css_get(blkcg_css);
+		}
+	}
+	spin_unlock_irqrestore(&cgwb_lock, flags);
+	if (ret) {
+		if (ret == -EEXIST)
+			ret = 0;
+		goto err_put_congested;
+	}
+	goto out_put;
+
+err_put_congested:
+	wb_congested_put(wb->congested);
+err_fprop_exit:
+	fprop_local_destroy_percpu(&wb->memcg_completions);
+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;
+}
+
+/**
+ * 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.  The returned wb has its refcount incremented.
+ *
+ * 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_create(struct backing_dev_info *bdi,
+				    struct cgroup_subsys_state *memcg_css,
+				    gfp_t gfp)
+{
+	struct bdi_writeback *wb;
+
+	might_sleep_if(gfp & __GFP_WAIT);
+
+	if (!memcg_css->parent)
+		return &bdi->wb;
+
+	do {
+		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,
+						     &blkio_cgrp_subsys);
+			if (unlikely(wb->blkcg_css != blkcg_css ||
+				     !wb_tryget(wb)))
+				wb = NULL;
+			css_put(blkcg_css);
+		}
+		rcu_read_unlock();
+	} while (!wb && !cgwb_create(bdi, memcg_css, gfp));
+
+	return wb;
+}
+
+static void cgwb_bdi_init(struct backing_dev_info *bdi)
+{
+	bdi->wb.memcg_css = mem_cgroup_root_css;
+	bdi->wb.blkcg_css = blkcg_root_css;
+	bdi->wb_congested.blkcg_id = 1;
+	INIT_RADIX_TREE(&bdi->cgwb_tree, GFP_ATOMIC);
+	bdi->cgwb_congested_tree = RB_ROOT;
+	atomic_set(&bdi->usage_cnt, 1);
+}
+
+static void cgwb_bdi_destroy(struct backing_dev_info *bdi)
+{
+	struct radix_tree_iter iter;
+	void **slot;
+
+	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);
+	spin_unlock_irq(&cgwb_lock);
+
+	/*
+	 * All cgwb's and their congested states must be shutdown and
+	 * released before returning.  Drain the usage counter to wait for
+	 * all cgwb's and cgwb_congested's ever created on @bdi.
+	 */
+	atomic_dec(&bdi->usage_cnt);
+	wait_event(cgwb_release_wait, !atomic_read(&bdi->usage_cnt));
+}
+
+/**
+ * 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.
+ */
+void wb_memcg_offline(struct mem_cgroup *memcg)
+{
+	LIST_HEAD(to_destroy);
+	struct list_head *memcg_cgwb_list = mem_cgroup_cgwb_list(memcg);
+	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);
+}
+
+/**
+ * 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)
+{
+	LIST_HEAD(to_destroy);
+	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);
+}
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static void cgwb_bdi_init(struct backing_dev_info *bdi) { }
+static void cgwb_bdi_destroy(struct backing_dev_info *bdi) { }
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
+
+int bdi_init(struct backing_dev_info *bdi)
+{
+	int err;
+
+	bdi->dev = NULL;
+
+	bdi->min_ratio = 0;
+	bdi->max_ratio = 100;
+	bdi->max_prop_frac = FPROP_FRAC_BASE;
+	INIT_LIST_HEAD(&bdi->bdi_list);
+	init_waitqueue_head(&bdi->wb_waitq);
+
+	err = wb_init(&bdi->wb, bdi, GFP_KERNEL);
+	if (err)
+		return err;
+
+	bdi->wb_congested.state = 0;
+	bdi->wb.congested = &bdi->wb_congested;
+
+	cgwb_bdi_init(bdi);
+	return 0;
+}
+EXPORT_SYMBOL(bdi_init);
 
 int bdi_register(struct backing_dev_info *bdi, struct device *parent,
 		const char *fmt, ...)
···
 	bdi->dev = dev;
 
 	bdi_debug_register(bdi, dev_name(dev));
-	set_bit(BDI_registered, &bdi->state);
+	set_bit(WB_registered, &bdi->wb.state);
 
 	spin_lock_bh(&bdi_lock);
 	list_add_tail_rcu(&bdi->bdi_list, &bdi_list);
···
 EXPORT_SYMBOL(bdi_register_dev);
 
 /*
- * Remove bdi from the global list and shutdown any threads we have running
+ * Remove bdi from bdi_list, and ensure that it is no longer visible
  */
-static void bdi_wb_shutdown(struct backing_dev_info *bdi)
+static void bdi_remove_from_list(struct backing_dev_info *bdi)
 {
-	/* Make sure nobody queues further work */
-	spin_lock_bh(&bdi->wb_lock);
-	if (!test_and_clear_bit(BDI_registered, &bdi->state)) {
-		spin_unlock_bh(&bdi->wb_lock);
-		return;
-	}
-	spin_unlock_bh(&bdi->wb_lock);
+	spin_lock_bh(&bdi_lock);
+	list_del_rcu(&bdi->bdi_list);
+	spin_unlock_bh(&bdi_lock);
 
-	/*
-	 * Make sure nobody finds us on the bdi_list anymore
-	 */
-	bdi_remove_from_list(bdi);
-
-	/*
-	 * Drain work list and shutdown the delayed_work.  At this point,
-	 * @bdi->bdi_list is empty telling bdi_Writeback_workfn() that @bdi
-	 * is dying and its work_list needs to be drained no matter what.
-	 */
-	mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
-	flush_delayed_work(&bdi->wb.dwork);
+	synchronize_rcu_expedited();
 }
-
-static void bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi)
-{
-	memset(wb, 0, sizeof(*wb));
-
-	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);
-	spin_lock_init(&wb->list_lock);
-	INIT_DELAYED_WORK(&wb->dwork, bdi_writeback_workfn);
-}
-
-/*
- * Initial write bandwidth: 100 MB/s
- */
-#define INIT_BW		(100 << (20 - PAGE_SHIFT))
-
-int bdi_init(struct backing_dev_info *bdi)
-{
-	int i, err;
-
-	bdi->dev = NULL;
-
-	bdi->min_ratio = 0;
-	bdi->max_ratio = 100;
-	bdi->max_prop_frac = FPROP_FRAC_BASE;
-	spin_lock_init(&bdi->wb_lock);
-	INIT_LIST_HEAD(&bdi->bdi_list);
-	INIT_LIST_HEAD(&bdi->work_list);
-
-	bdi_wb_init(&bdi->wb, bdi);
-
-	for (i = 0; i < NR_BDI_STAT_ITEMS; i++) {
-		err = percpu_counter_init(&bdi->bdi_stat[i], 0, GFP_KERNEL);
-		if (err)
-			goto err;
-	}
-
-	bdi->dirty_exceeded = 0;
-
-	bdi->bw_time_stamp = jiffies;
-	bdi->written_stamp = 0;
-
-	bdi->balanced_dirty_ratelimit = INIT_BW;
-	bdi->dirty_ratelimit = INIT_BW;
-	bdi->write_bandwidth = INIT_BW;
-	bdi->avg_write_bandwidth = INIT_BW;
-
-	err = fprop_local_init_percpu(&bdi->completions, GFP_KERNEL);
-
-	if (err) {
-err:
-		while (i--)
-			percpu_counter_destroy(&bdi->bdi_stat[i]);
-	}
-
-	return err;
-}
-EXPORT_SYMBOL(bdi_init);
 
 void bdi_destroy(struct backing_dev_info *bdi)
 {
-	int i;
-
-	bdi_wb_shutdown(bdi);
-	bdi_set_min_ratio(bdi, 0);
-
-	WARN_ON(!list_empty(&bdi->work_list));
-	WARN_ON(delayed_work_pending(&bdi->wb.dwork));
+	/* make sure nobody finds us on the bdi_list anymore */
+	bdi_remove_from_list(bdi);
+	wb_shutdown(&bdi->wb);
+	cgwb_bdi_destroy(bdi);
 
 	if (bdi->dev) {
 		bdi_debug_unregister(bdi);
···
 		bdi->dev = NULL;
 	}
 
-	for (i = 0; i < NR_BDI_STAT_ITEMS; i++)
-		percpu_counter_destroy(&bdi->bdi_stat[i]);
-	fprop_local_destroy_percpu(&bdi->completions);
+	wb_exit(&bdi->wb);
 }
 EXPORT_SYMBOL(bdi_destroy);
 
···
 		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
 		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
 	};
-static atomic_t nr_bdi_congested[2];
+static atomic_t nr_wb_congested[2];
 
-void clear_bdi_congested(struct backing_dev_info *bdi, int sync)
+void clear_wb_congested(struct bdi_writeback_congested *congested, int sync)
 {
-	enum bdi_state bit;
 	wait_queue_head_t *wqh = &congestion_wqh[sync];
+	enum wb_state bit;
 
-	bit = sync ? BDI_sync_congested : BDI_async_congested;
-	if (test_and_clear_bit(bit, &bdi->state))
-		atomic_dec(&nr_bdi_congested[sync]);
+	bit = sync ? WB_sync_congested : WB_async_congested;
+	if (test_and_clear_bit(bit, &congested->state))
+		atomic_dec(&nr_wb_congested[sync]);
 	smp_mb__after_atomic();
 	if (waitqueue_active(wqh))
 		wake_up(wqh);
 }
-EXPORT_SYMBOL(clear_bdi_congested);
+EXPORT_SYMBOL(clear_wb_congested);
 
-void set_bdi_congested(struct backing_dev_info *bdi, int sync)
+void set_wb_congested(struct bdi_writeback_congested *congested, int sync)
 {
-	enum bdi_state bit;
+	enum wb_state bit;
 
-	bit = sync ? BDI_sync_congested : BDI_async_congested;
-	if (!test_and_set_bit(bit, &bdi->state))
-		atomic_inc(&nr_bdi_congested[sync]);
+	bit = sync ? WB_sync_congested : WB_async_congested;
+	if (!test_and_set_bit(bit, &congested->state))
+		atomic_inc(&nr_wb_congested[sync]);
 }
-EXPORT_SYMBOL(set_bdi_congested);
+EXPORT_SYMBOL(set_wb_congested);
 
 /**
  * congestion_wait - wait for a backing_dev to become uncongested
···
 	 * encountered in the current zone, yield if necessary instead
 	 * of sleeping on the congestion queue
 	 */
-	if (atomic_read(&nr_bdi_congested[sync]) == 0 ||
+	if (atomic_read(&nr_wb_congested[sync]) == 0 ||
 	    !test_bit(ZONE_CONGESTED, &zone->flags)) {
 		cond_resched();
 

mm/fadvise.c

···
 	case POSIX_FADV_NOREUSE:
 		break;
 	case POSIX_FADV_DONTNEED:
-		if (!bdi_write_congested(bdi))
+		if (!inode_write_congested(mapping->host))
 			__filemap_fdatawrite_range(mapping, offset, endbyte,
 						   WB_SYNC_NONE);
 

mm/filemap.c

···
  *    ->tree_lock		(page_remove_rmap->set_page_dirty)
  *    bdi.wb->list_lock		(page_remove_rmap->set_page_dirty)
  *    ->inode->i_lock		(page_remove_rmap->set_page_dirty)
+ *    ->memcg->move_lock	(page_remove_rmap->mem_cgroup_begin_page_stat)
  *    bdi.wb->list_lock		(zap_pte_range->set_page_dirty)
  *    ->inode->i_lock		(zap_pte_range->set_page_dirty)
  *    ->private_lock		(zap_pte_range->__set_page_dirty_buffers)
···
 /*
  * Delete a page from the page cache and free it. Caller has to make
  * sure the page is locked and that nobody else uses it - or that usage
- * is safe.  The caller must hold the mapping's tree_lock.
+ * is safe.  The caller must hold the mapping's tree_lock and
+ * mem_cgroup_begin_page_stat().
  */
-void __delete_from_page_cache(struct page *page, void *shadow)
+void __delete_from_page_cache(struct page *page, void *shadow,
+			      struct mem_cgroup *memcg)
 {
 	struct address_space *mapping = page->mapping;
 
···
 	 * anyway will be cleared before returning page into buddy allocator.
 	 */
 	if (WARN_ON_ONCE(PageDirty(page)))
-		account_page_cleaned(page, mapping);
+		account_page_cleaned(page, mapping, memcg,
+				     inode_to_wb(mapping->host));
 }
 
 /**
···
 void delete_from_page_cache(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
+	struct mem_cgroup *memcg;
+	unsigned long flags;
+
 	void (*freepage)(struct page *);
 
 	BUG_ON(!PageLocked(page));
 
 	freepage = mapping->a_ops->freepage;
-	spin_lock_irq(&mapping->tree_lock);
-	__delete_from_page_cache(page, NULL);
-	spin_unlock_irq(&mapping->tree_lock);
+
+	memcg = mem_cgroup_begin_page_stat(page);
+	spin_lock_irqsave(&mapping->tree_lock, flags);
+	__delete_from_page_cache(page, NULL, memcg);
+	spin_unlock_irqrestore(&mapping->tree_lock, flags);
+	mem_cgroup_end_page_stat(memcg);
 
 	if (freepage)
 		freepage(page);
···
 	if (!mapping_cap_writeback_dirty(mapping))
 		return 0;
 
+	wbc_attach_fdatawrite_inode(&wbc, mapping->host);
 	ret = do_writepages(mapping, &wbc);
+	wbc_detach_inode(&wbc);
 	return ret;
 }
 
···
 	if (!error) {
 		struct address_space *mapping = old->mapping;
 		void (*freepage)(struct page *);
+		struct mem_cgroup *memcg;
+		unsigned long flags;
 
 		pgoff_t offset = old->index;
 		freepage = mapping->a_ops->freepage;
···
 		new->mapping = mapping;
 		new->index = offset;
 
-		spin_lock_irq(&mapping->tree_lock);
-		__delete_from_page_cache(old, NULL);
+		memcg = mem_cgroup_begin_page_stat(old);
+		spin_lock_irqsave(&mapping->tree_lock, flags);
+		__delete_from_page_cache(old, NULL, memcg);
 		error = radix_tree_insert(&mapping->page_tree, offset, new);
 		BUG_ON(error);
 		mapping->nrpages++;
···
 			__inc_zone_page_state(new, NR_FILE_PAGES);
 		if (PageSwapBacked(new))
 			__inc_zone_page_state(new, NR_SHMEM);
-		spin_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
+		mem_cgroup_end_page_stat(memcg);
 		mem_cgroup_migrate(old, new, true);
 		radix_tree_preload_end();
 		if (freepage)

mm/madvise.c

···
 #include <linux/fs.h>
 #include <linux/file.h>
 #include <linux/blkdev.h>
+#include <linux/backing-dev.h>
 #include <linux/swap.h>
 #include <linux/swapops.h>
 

mm/memcontrol.c

···
 
 #define MEM_CGROUP_RECLAIM_RETRIES	5
 static struct mem_cgroup *root_mem_cgroup __read_mostly;
+struct cgroup_subsys_state *mem_cgroup_root_css __read_mostly;
 
 /* Whether the swap controller is active */
 #ifdef CONFIG_MEMCG_SWAP
···
 	"rss",
 	"rss_huge",
 	"mapped_file",
+	"dirty",
 	"writeback",
 	"swap",
 };
···
 	 * percpu counter.
 	 */
 	struct mem_cgroup_stat_cpu __percpu *stat;
-	/*
-	 * used when a cpu is offlined or other synchronizations
-	 * See mem_cgroup_read_stat().
-	 */
-	struct mem_cgroup_stat_cpu nocpu_base;
 	spinlock_t pcp_counter_lock;
 
 #if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_INET)
···
 	nodemask_t	scan_nodes;
 	atomic_t	numainfo_events;
 	atomic_t	numainfo_updating;
+#endif
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct list_head cgwb_list;
+	struct wb_domain cgwb_domain;
 #endif
 
 	/* List of events which userspace want to receive */
···
 	return &memcg->css;
 }
 
+/**
+ * mem_cgroup_css_from_page - css of the memcg associated with a page
+ * @page: page of interest
+ *
+ * If memcg is bound to the default hierarchy, css of the memcg associated
+ * with @page is returned.  The returned css remains associated with @page
+ * until it is released.
+ *
+ * If memcg is bound to a traditional hierarchy, the css of root_mem_cgroup
+ * is returned.
+ *
+ * XXX: The above description of behavior on the default hierarchy isn't
+ * strictly true yet as replace_page_cache_page() can modify the
+ * association before @page is released even on the default hierarchy;
+ * however, the current and planned usages don't mix the the two functions
+ * and replace_page_cache_page() will soon be updated to make the invariant
+ * actually true.
+ */
+struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page)
+{
+	struct mem_cgroup *memcg;
+
+	rcu_read_lock();
+
+	memcg = page->mem_cgroup;
+
+	if (!memcg || !cgroup_on_dfl(memcg->css.cgroup))
+		memcg = root_mem_cgroup;
+
+	rcu_read_unlock();
+	return &memcg->css;
+}
+
 static struct mem_cgroup_per_zone *
 mem_cgroup_page_zoneinfo(struct mem_cgroup *memcg, struct page *page)
 {
···
 	long val = 0;
 	int cpu;
 
-	get_online_cpus();
-	for_each_online_cpu(cpu)
+	for_each_possible_cpu(cpu)
 		val += per_cpu(memcg->stat->count[idx], cpu);
-#ifdef CONFIG_HOTPLUG_CPU
-	spin_lock(&memcg->pcp_counter_lock);
-	val += memcg->nocpu_base.count[idx];
-	spin_unlock(&memcg->pcp_counter_lock);
-#endif
-	put_online_cpus();
 	return val;
 }
 
···
 	unsigned long val = 0;
 	int cpu;
 
-	get_online_cpus();
-	for_each_online_cpu(cpu)
+	for_each_possible_cpu(cpu)
 		val += per_cpu(memcg->stat->events[idx], cpu);
-#ifdef CONFIG_HOTPLUG_CPU
-	spin_lock(&memcg->pcp_counter_lock);
-	val += memcg->nocpu_base.events[idx];
-	spin_unlock(&memcg->pcp_counter_lock);
-#endif
-	put_online_cpus();
 	return val;
 }
 
···
 
 	return memcg;
 }
+EXPORT_SYMBOL(mem_cgroup_begin_page_stat);
 
 /**
  * mem_cgroup_end_page_stat - finish a page state statistics transaction
···
 
 	rcu_read_unlock();
 }
+EXPORT_SYMBOL(mem_cgroup_end_page_stat);
 
 /**
  * mem_cgroup_update_page_stat - update page state statistics
···
 	mutex_unlock(&percpu_charge_mutex);
 }
 
-/*
- * This function drains percpu counter value from DEAD cpu and
- * move it to local cpu. Note that this function can be preempted.
- */
-static void mem_cgroup_drain_pcp_counter(struct mem_cgroup *memcg, int cpu)
-{
-	int i;
-
-	spin_lock(&memcg->pcp_counter_lock);
-	for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) {
-		long x = per_cpu(memcg->stat->count[i], cpu);
-
-		per_cpu(memcg->stat->count[i], cpu) = 0;
-		memcg->nocpu_base.count[i] += x;
-	}
-	for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) {
-		unsigned long x = per_cpu(memcg->stat->events[i], cpu);
-
-		per_cpu(memcg->stat->events[i], cpu) = 0;
-		memcg->nocpu_base.events[i] += x;
-	}
-	spin_unlock(&memcg->pcp_counter_lock);
-}
-
 static int memcg_cpu_hotplug_callback(struct notifier_block *nb,
 					unsigned long action,
 					void *hcpu)
 {
 	int cpu = (unsigned long)hcpu;
 	struct memcg_stock_pcp *stock;
-	struct mem_cgroup *iter;
 
 	if (action == CPU_ONLINE)
 		return NOTIFY_OK;
 
 	if (action != CPU_DEAD && action != CPU_DEAD_FROZEN)
 		return NOTIFY_OK;
-
-	for_each_mem_cgroup(iter)
-		mem_cgroup_drain_pcp_counter(iter, cpu);
 
 	stock = &per_cpu(memcg_stock, cpu);
 	drain_stock(stock);
···
 }
 #endif
 
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg)
+{
+	return &memcg->cgwb_list;
+}
+
+static int memcg_wb_domain_init(struct mem_cgroup *memcg, gfp_t gfp)
+{
+	return wb_domain_init(&memcg->cgwb_domain, gfp);
+}
+
+static void memcg_wb_domain_exit(struct mem_cgroup *memcg)
+{
+	wb_domain_exit(&memcg->cgwb_domain);
+}
+
+static void memcg_wb_domain_size_changed(struct mem_cgroup *memcg)
+{
+	wb_domain_size_changed(&memcg->cgwb_domain);
+}
+
+struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb)
+{
+	struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css);
+
+	if (!memcg->css.parent)
+		return NULL;
+
+	return &memcg->cgwb_domain;
+}
+
+/**
+ * mem_cgroup_wb_stats - retrieve writeback related stats from its memcg
+ * @wb: bdi_writeback in question
+ * @pavail: out parameter for number of available pages
+ * @pdirty: out parameter for number of dirty pages
+ * @pwriteback: out parameter for number of pages under writeback
+ *
+ * Determine the numbers of available, dirty, and writeback pages in @wb's
+ * memcg.  Dirty and writeback are self-explanatory.  Available is a bit
+ * more involved.
+ *
+ * A memcg's headroom is "min(max, high) - used".  The available memory is
+ * calculated as the lowest headroom of itself and the ancestors plus the
+ * number of pages already being used for file pages.  Note that this
+ * doesn't consider the actual amount of available memory in the system.
+ * The caller should further cap *@pavail accordingly.
+ */
+void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pavail,
+			 unsigned long *pdirty, unsigned long *pwriteback)
+{
+	struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css);
+	struct mem_cgroup *parent;
+	unsigned long head_room = PAGE_COUNTER_MAX;
+	unsigned long file_pages;
+
+	*pdirty = mem_cgroup_read_stat(memcg, MEM_CGROUP_STAT_DIRTY);
+
+	/* this should eventually include NR_UNSTABLE_NFS */
+	*pwriteback = mem_cgroup_read_stat(memcg, MEM_CGROUP_STAT_WRITEBACK);
+
+	file_pages = mem_cgroup_nr_lru_pages(memcg, (1 << LRU_INACTIVE_FILE) |
+						    (1 << LRU_ACTIVE_FILE));
+	while ((parent = parent_mem_cgroup(memcg))) {
+		unsigned long ceiling = min(memcg->memory.limit, memcg->high);
+		unsigned long used = page_counter_read(&memcg->memory);
+
+		head_room = min(head_room, ceiling - min(ceiling, used));
+		memcg = parent;
+	}
+
+	*pavail = file_pages + head_room;
+}
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+static int memcg_wb_domain_init(struct mem_cgroup *memcg, gfp_t gfp)
+{
+	return 0;
+}
+
+static void memcg_wb_domain_exit(struct mem_cgroup *memcg)
+{
+}
+
+static void memcg_wb_domain_size_changed(struct mem_cgroup *memcg)
+{
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
+
 /*
  * DO NOT USE IN NEW FILES.
  *
···
 	memcg->stat = alloc_percpu(struct mem_cgroup_stat_cpu);
 	if (!memcg->stat)
 		goto out_free;
+
+	if (memcg_wb_domain_init(memcg, GFP_KERNEL))
+		goto out_free_stat;
+
 	spin_lock_init(&memcg->pcp_counter_lock);
 	return memcg;
 
+out_free_stat:
+	free_percpu(memcg->stat);
 out_free:
 	kfree(memcg);
 	return NULL;
···
 		free_mem_cgroup_per_zone_info(memcg, node);
 
 	free_percpu(memcg->stat);
+	memcg_wb_domain_exit(memcg);
 	kfree(memcg);
 }
 
···
 	/* root ? */
 	if (parent_css == NULL) {
 		root_mem_cgroup = memcg;
+		mem_cgroup_root_css = &memcg->css;
 		page_counter_init(&memcg->memory, NULL);
 		memcg->high = PAGE_COUNTER_MAX;
 		memcg->soft_limit = PAGE_COUNTER_MAX;
···
 #ifdef CONFIG_MEMCG_KMEM
 	memcg->kmemcg_id = -1;
 #endif
-
+#ifdef CONFIG_CGROUP_WRITEBACK
+	INIT_LIST_HEAD(&memcg->cgwb_list);
+#endif
 	return &memcg->css;
 
 free_out:
···
 	vmpressure_cleanup(&memcg->vmpressure);
 
 	memcg_deactivate_kmem(memcg);
+
+	wb_memcg_offline(memcg);
 }
 
 static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
···
 	memcg->low = 0;
 	memcg->high = PAGE_COUNTER_MAX;
 	memcg->soft_limit = PAGE_COUNTER_MAX;
+	memcg_wb_domain_size_changed(memcg);
 }
 
 #ifdef CONFIG_MMU
···
 {
 	unsigned long flags;
 	int ret;
+	bool anon;
 
 	VM_BUG_ON(from == to);
 	VM_BUG_ON_PAGE(PageLRU(page), page);
···
 	if (page->mem_cgroup != from)
 		goto out_unlock;
 
+	anon = PageAnon(page);
+
 	spin_lock_irqsave(&from->move_lock, flags);
 
-	if (!PageAnon(page) && page_mapped(page)) {
+	if (!anon && page_mapped(page)) {
 		__this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_FILE_MAPPED],
 			       nr_pages);
 		__this_cpu_add(to->stat->count[MEM_CGROUP_STAT_FILE_MAPPED],
 			       nr_pages);
+	}
+
+	/*
+	 * move_lock grabbed above and caller set from->moving_account, so
+	 * mem_cgroup_update_page_stat() will serialize updates to PageDirty.
+	 * So mapping should be stable for dirty pages.
+	 */
+	if (!anon && PageDirty(page)) {
+		struct address_space *mapping = page_mapping(page);
+
+		if (mapping_cap_account_dirty(mapping)) {
+			__this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_DIRTY],
+				       nr_pages);
+			__this_cpu_add(to->stat->count[MEM_CGROUP_STAT_DIRTY],
+				       nr_pages);
+		}
 	}
 
 	if (PageWriteback(page)) {
···
 
 	memcg->high = high;
 
+	memcg_wb_domain_size_changed(memcg);
 	return nbytes;
 }
 
···
 	if (err)
 		return err;
 
+	memcg_wb_domain_size_changed(memcg);
 	return nbytes;
 }
 

mm/page-writeback.c

···
 
 /* End of sysctl-exported parameters */
 
-unsigned long global_dirty_limit;
+struct wb_domain global_wb_domain;
 
-/*
- * Scale the writeback cache size proportional to the relative writeout speeds.
- *
- * We do this by keeping a floating proportion between BDIs, based on page
- * writeback completions [end_page_writeback()]. Those devices that write out
- * pages fastest will get the larger share, while the slower will get a smaller
- * share.
- *
- * We use page writeout completions because we are interested in getting rid of
- * dirty pages. Having them written out is the primary goal.
- *
- * We introduce a concept of time, a period over which we measure these events,
- * because demand can/will vary over time. The length of this period itself is
- * measured in page writeback completions.
- *
- */
-static struct fprop_global writeout_completions;
+/* consolidated parameters for balance_dirty_pages() and its subroutines */
+struct dirty_throttle_control {
+#ifdef CONFIG_CGROUP_WRITEBACK
+	struct wb_domain	*dom;
+	struct dirty_throttle_control *gdtc;	/* only set in memcg dtc's */
+#endif
+	struct bdi_writeback	*wb;
+	struct fprop_local_percpu *wb_completions;
 
-static void writeout_period(unsigned long t);
-/* Timer for aging of writeout_completions */
-static struct timer_list writeout_period_timer =
-		TIMER_DEFERRED_INITIALIZER(writeout_period, 0, 0);
-static unsigned long writeout_period_time = 0;
+	unsigned long		avail;		/* dirtyable */
+	unsigned long		dirty;		/* file_dirty + write + nfs */
+	unsigned long		thresh;		/* dirty threshold */
+	unsigned long		bg_thresh;	/* dirty background threshold */
+
+	unsigned long		wb_dirty;	/* per-wb counterparts */
+	unsigned long		wb_thresh;
+	unsigned long		wb_bg_thresh;
+
+	unsigned long		pos_ratio;
+};
+
+#define DTC_INIT_COMMON(__wb)	.wb = (__wb),				\
+				.wb_completions = &(__wb)->completions
 
 /*
  * Length of period for aging writeout fractions of bdis. This is an
···
  * reflect changes in current writeout rate.
  */
 #define VM_COMPLETIONS_PERIOD_LEN (3*HZ)
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+
+#define GDTC_INIT(__wb)		.dom = &global_wb_domain,		\
+				DTC_INIT_COMMON(__wb)
+#define GDTC_INIT_NO_WB		.dom = &global_wb_domain
+#define MDTC_INIT(__wb, __gdtc)	.dom = mem_cgroup_wb_domain(__wb),	\
+				.gdtc = __gdtc,				\
+				DTC_INIT_COMMON(__wb)
+
+static bool mdtc_valid(struct dirty_throttle_control *dtc)
+{
+	return dtc->dom;
+}
+
+static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc)
+{
+	return dtc->dom;
+}
+
+static struct dirty_throttle_control *mdtc_gdtc(struct dirty_throttle_control *mdtc)
+{
+	return mdtc->gdtc;
+}
+
+static struct fprop_local_percpu *wb_memcg_completions(struct bdi_writeback *wb)
+{
+	return &wb->memcg_completions;
+}
+
+static void wb_min_max_ratio(struct bdi_writeback *wb,
+			     unsigned long *minp, unsigned long *maxp)
+{
+	unsigned long this_bw = wb->avg_write_bandwidth;
+	unsigned long tot_bw = atomic_long_read(&wb->bdi->tot_write_bandwidth);
+	unsigned long long min = wb->bdi->min_ratio;
+	unsigned long long max = wb->bdi->max_ratio;
+
+	/*
+	 * @wb may already be clean by the time control reaches here and
+	 * the total may not include its bw.
+	 */
+	if (this_bw < tot_bw) {
+		if (min) {
+			min *= this_bw;
+			do_div(min, tot_bw);
+		}
+		if (max < 100) {
+			max *= this_bw;
+			do_div(max, tot_bw);
+		}
+	}
+
+	*minp = min;
+	*maxp = max;
+}
+
+#else	/* CONFIG_CGROUP_WRITEBACK */
+
+#define GDTC_INIT(__wb)		DTC_INIT_COMMON(__wb)
+#define GDTC_INIT_NO_WB
+#define MDTC_INIT(__wb, __gdtc)
+
+static bool mdtc_valid(struct dirty_throttle_control *dtc)
+{
+	return false;
+}
+
+static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc)
+{
+	return &global_wb_domain;
+}
+
+static struct dirty_throttle_control *mdtc_gdtc(struct dirty_throttle_control *mdtc)
+{
+	return NULL;
+}
+
+static struct fprop_local_percpu *wb_memcg_completions(struct bdi_writeback *wb)
+{
+	return NULL;
+}
+
+static void wb_min_max_ratio(struct bdi_writeback *wb,
+			     unsigned long *minp, unsigned long *maxp)
+{
+	*minp = wb->bdi->min_ratio;
+	*maxp = wb->bdi->max_ratio;
+}
+
+#endif	/* CONFIG_CGROUP_WRITEBACK */
 
 /*
  * In a memory zone, there is a certain amount of pages we consider
···
 	return x + 1;	/* Ensure that we never return 0 */
 }
 
-/*
- * global_dirty_limits - background-writeback and dirty-throttling thresholds
+/**
+ * domain_dirty_limits - calculate thresh and bg_thresh for a wb_domain
+ * @dtc: dirty_throttle_control of interest
  *
- * Calculate the dirty thresholds based on sysctl parameters
- * - vm.dirty_background_ratio  or  vm.dirty_background_bytes
- * - vm.dirty_ratio             or  vm.dirty_bytes
- * The dirty limits will be lifted by 1/4 for PF_LESS_THROTTLE (ie. nfsd) and
+ * Calculate @dtc->thresh and ->bg_thresh considering
+ * vm_dirty_{bytes|ratio} and dirty_background_{bytes|ratio}.  The caller
+ * must ensure that @dtc->avail is set before calling this function.  The
+ * dirty limits will be lifted by 1/4 for PF_LESS_THROTTLE (ie. nfsd) and
  * real-time tasks.
+ */
+static void domain_dirty_limits(struct dirty_throttle_control *dtc)
+{
+	const unsigned long available_memory = dtc->avail;
+	struct dirty_throttle_control *gdtc = mdtc_gdtc(dtc);
+	unsigned long bytes = vm_dirty_bytes;
+	unsigned long bg_bytes = dirty_background_bytes;
+	unsigned long ratio = vm_dirty_ratio;
+	unsigned long bg_ratio = dirty_background_ratio;
+	unsigned long thresh;
+	unsigned long bg_thresh;
+	struct task_struct *tsk;
+
+	/* gdtc is !NULL iff @dtc is for memcg domain */
+	if (gdtc) {
+		unsigned long global_avail = gdtc->avail;
+
+		/*
+		 * The byte settings can't be applied directly to memcg
+		 * domains.  Convert them to ratios by scaling against
+		 * globally available memory.
+		 */
+		if (bytes)
+			ratio = min(DIV_ROUND_UP(bytes, PAGE_SIZE) * 100 /
+				    global_avail, 100UL);
+		if (bg_bytes)
+			bg_ratio = min(DIV_ROUND_UP(bg_bytes, PAGE_SIZE) * 100 /
+				       global_avail, 100UL);
+		bytes = bg_bytes = 0;
+	}
+
+	if (bytes)
+		thresh = DIV_ROUND_UP(bytes, PAGE_SIZE);
+	else
+		thresh = (ratio * available_memory) / 100;
+
+	if (bg_bytes)
+		bg_thresh = DIV_ROUND_UP(bg_bytes, PAGE_SIZE);
+	else
+		bg_thresh = (bg_ratio * available_memory) / 100;
+
+	if (bg_thresh >= thresh)
+		bg_thresh = thresh / 2;
+	tsk = current;
+	if (tsk->flags & PF_LESS_THROTTLE || rt_task(tsk)) {
+		bg_thresh += bg_thresh / 4;
+		thresh += thresh / 4;
+	}
+	dtc->thresh = thresh;
+	dtc->bg_thresh = bg_thresh;
+
+	/* we should eventually report the domain in the TP */
+	if (!gdtc)
+		trace_global_dirty_state(bg_thresh, thresh);
+}
+
+/**
+ * global_dirty_limits - background-writeback and dirty-throttling thresholds
+ * @pbackground: out parameter for bg_thresh
+ * @pdirty: out parameter for thresh
+ *
+ * Calculate bg_thresh and thresh for global_wb_domain.  See
+ * domain_dirty_limits() for details.
  */
 void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty)
 {
-	const unsigned long available_memory = global_dirtyable_memory();
-	unsigned long background;
-	unsigned long dirty;
-	struct task_struct *tsk;
+	struct dirty_throttle_control gdtc = { GDTC_INIT_NO_WB };
 
-	if (vm_dirty_bytes)
-		dirty = DIV_ROUND_UP(vm_dirty_bytes, PAGE_SIZE);
-	else
-		dirty = (vm_dirty_ratio * available_memory) / 100;
+	gdtc.avail = global_dirtyable_memory();
+	domain_dirty_limits(&gdtc);
 
-	if (dirty_background_bytes)
-		background = DIV_ROUND_UP(dirty_background_bytes, PAGE_SIZE);
-	else
-		background = (dirty_background_ratio * available_memory) / 100;
-
-	if (background >= dirty)
-		background = dirty / 2;
-	tsk = current;
-	if (tsk->flags & PF_LESS_THROTTLE || rt_task(tsk)) {
-		background += background / 4;
-		dirty += dirty / 4;
-	}
-	*pbackground = background;
-	*pdirty = dirty;
-	trace_global_dirty_state(background, dirty);
+	*pbackground = gdtc.bg_thresh;
+	*pdirty = gdtc.thresh;
 }
 
 /**
···
 	return cur_time;
 }
 
-/*
- * Increment the BDI's writeout completion count and the global writeout
- * completion count. Called from test_clear_page_writeback().
- */
-static inline void __bdi_writeout_inc(struct backing_dev_info *bdi)
+static void wb_domain_writeout_inc(struct wb_domain *dom,
+				   struct fprop_local_percpu *completions,
+				   unsigned int max_prop_frac)
 {
-	__inc_bdi_stat(bdi, BDI_WRITTEN);
-	__fprop_inc_percpu_max(&writeout_completions, &bdi->completions,
-			       bdi->max_prop_frac);
+	__fprop_inc_percpu_max(&dom->completions, completions,
+			       max_prop_frac);
 	/* First event after period switching was turned off? */
-	if (!unlikely(writeout_period_time)) {
+	if (!unlikely(dom->period_time)) {
 		/*
 		 * We can race with other __bdi_writeout_inc calls here but
 		 * it does not cause any harm since the resulting time when
 		 * timer will fire and what is in writeout_period_time will be
 		 * roughly the same.
 		 */
-		writeout_period_time = wp_next_time(jiffies);
-		mod_timer(&writeout_period_timer, writeout_period_time);
+		dom->period_time = wp_next_time(jiffies);
+		mod_timer(&dom->period_timer, dom->period_time);
 	}
 }
 
-void bdi_writeout_inc(struct backing_dev_info *bdi)
+/*
+ * Increment @wb's writeout completion count and the global writeout
+ * completion count. Called from test_clear_page_writeback().
+ */
+static inline void __wb_writeout_inc(struct bdi_writeback *wb)
+{
+	struct wb_domain *cgdom;
+
+	__inc_wb_stat(wb, WB_WRITTEN);
+	wb_domain_writeout_inc(&global_wb_domain, &wb->completions,
+			       wb->bdi->max_prop_frac);
+
+	cgdom = mem_cgroup_wb_domain(wb);
+	if (cgdom)
+		wb_domain_writeout_inc(cgdom, wb_memcg_completions(wb),
+				       wb->bdi->max_prop_frac);
+}
+
+void wb_writeout_inc(struct bdi_writeback *wb)
 {
 	unsigned long flags;
 
 	local_irq_save(flags);
-	__bdi_writeout_inc(bdi);
+	__wb_writeout_inc(wb);
 	local_irq_restore(flags);
 }
-EXPORT_SYMBOL_GPL(bdi_writeout_inc);
-
-/*
- * Obtain an accurate fraction of the BDI's portion.
- */
-static void bdi_writeout_fraction(struct backing_dev_info *bdi,
-		long *numerator, long *denominator)
-{
-	fprop_fraction_percpu(&writeout_completions, &bdi->completions,
-				numerator, denominator);
-}
+EXPORT_SYMBOL_GPL(wb_writeout_inc);
 
 /*
  * On idle system, we can be called long after we scheduled because we use
···
  */
 static void writeout_period(unsigned long t)
 {
-	int miss_periods = (jiffies - writeout_period_time) /
+	struct wb_domain *dom = (void *)t;
+	int miss_periods = (jiffies - dom->period_time) /
 						 VM_COMPLETIONS_PERIOD_LEN;
 
-	if (fprop_new_period(&writeout_completions, miss_periods + 1)) {
-		writeout_period_time = wp_next_time(writeout_period_time +
+	if (fprop_new_period(&dom->completions, miss_periods + 1)) {
+		dom->period_time = wp_next_time(dom->period_time +
 				miss_periods * VM_COMPLETIONS_PERIOD_LEN);
-		mod_timer(&writeout_period_timer, writeout_period_time);
+		mod_timer(&dom->period_timer, dom->period_time);
 	} else {
 		/*
 		 * Aging has zeroed all fractions. Stop wasting CPU on period
 		 * updates.
 		 */
-		writeout_period_time = 0;
+		dom->period_time = 0;
 	}
 }
+
+int wb_domain_init(struct wb_domain *dom, gfp_t gfp)
+{
+	memset(dom, 0, sizeof(*dom));
+
+	spin_lock_init(&dom->lock);
+
+	init_timer_deferrable(&dom->period_timer);
+	dom->period_timer.function = writeout_period;
+	dom->period_timer.data = (unsigned long)dom;
+
+	dom->dirty_limit_tstamp = jiffies;
+
+	return fprop_global_init(&dom->completions, gfp);
+}
+
+#ifdef CONFIG_CGROUP_WRITEBACK
+void wb_domain_exit(struct wb_domain *dom)
+{
+	del_timer_sync(&dom->period_timer);
+	fprop_global_destroy(&dom->completions);
+}
+#endif
 
 /*
  * bdi_min_ratio keeps the sum of the minimum dirty shares of all
···
 	return (thresh + bg_thresh) / 2;
 }
 
-static unsigned long hard_dirty_limit(unsigned long thresh)
+static unsigned long hard_dirty_limit(struct wb_domain *dom,
+				      unsigned long thresh)
 {
-	return max(thresh, global_dirty_limit);
+	return max(thresh, dom->dirty_limit);
+}
+
+/* memory available to a memcg domain is capped by system-wide clean memory */
+static void mdtc_cap_avail(struct dirty_throttle_control *mdtc)
+{
+	struct dirty_throttle_control *gdtc = mdtc_gdtc(mdtc);
+	unsigned long clean = gdtc->avail - min(gdtc->avail, gdtc->dirty);
+
+	mdtc->avail = min(mdtc->avail, clean);
 }
 
 /**
- * bdi_dirty_limit - @bdi's share of dirty throttling threshold
- * @bdi: the backing_dev_info to query
- * @dirty: global dirty limit in pages
+ * __wb_calc_thresh - @wb's share of dirty throttling threshold
+ * @dtc: dirty_throttle_context of interest
  *
- * Returns @bdi's dirty limit in pages. The term "dirty" in the context of
+ * Returns @wb's dirty limit in pages. The term "dirty" in the context of
  * dirty balancing includes all PG_dirty, PG_writeback and NFS unstable pages.
  *
  * Note that balance_dirty_pages() will only seriously take it as a hard limit
···
  * control. For example, when the device is completely stalled due to some error
  * conditions, or when there are 1000 dd tasks writing to a slow 10MB/s USB key.
  * In the other normal situations, it acts more gently by throttling the tasks
- * more (rather than completely block them) when the bdi dirty pages go high.
+ * more (rather than completely block them) when the wb dirty pages go high.
  *
  * It allocates high/low dirty limits to fast/slow devices, in order to prevent
  * - starving fast devices
  * - piling up dirty pages (that will take long time to sync) on slow devices
  *
- * The bdi's share of dirty limit will be adapting to its throughput and
+ * The wb's share of dirty limit will be adapting to its throughput and
  * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set.
  */
-unsigned long bdi_dirty_limit(struct backing_dev_info *bdi, unsigned long dirty)
+static unsigned long __wb_calc_thresh(struct dirty_throttle_control *dtc)
 {
-	u64 bdi_dirty;
+	struct wb_domain *dom = dtc_dom(dtc);
+	unsigned long thresh = dtc->thresh;
+	u64 wb_thresh;
 	long numerator, denominator;
+	unsigned long wb_min_ratio, wb_max_ratio;
 
 	/*
-	 * Calculate this BDI's share of the dirty ratio.
+	 * Calculate this BDI's share of the thresh ratio.
 	 */
-	bdi_writeout_fraction(bdi, &numerator, &denominator);
+	fprop_fraction_percpu(&dom->completions, dtc->wb_completions,
+			      &numerator, &denominator);
 
-	bdi_dirty = (dirty * (100 - bdi_min_ratio)) / 100;
-	bdi_dirty *= numerator;
-	do_div(bdi_dirty, denominator);
+	wb_thresh = (thresh * (100 - bdi_min_ratio)) / 100;
+	wb_thresh *= numerator;
+	do_div(wb_thresh, denominator);
 
-	bdi_dirty += (dirty * bdi->min_ratio) / 100;
-	if (bdi_dirty > (dirty * bdi->max_ratio) / 100)
-		bdi_dirty = dirty * bdi->max_ratio / 100;
+	wb_min_max_ratio(dtc->wb, &wb_min_ratio, &wb_max_ratio);
 
-	return bdi_dirty;
+	wb_thresh += (thresh * wb_min_ratio) / 100;
+	if (wb_thresh > (thresh * wb_max_ratio) / 100)
+		wb_thresh = thresh * wb_max_ratio / 100;
+
+	return wb_thresh;
+}
+
+unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh)
+{
+	struct dirty_throttle_control gdtc = { GDTC_INIT(wb),
+					       .thresh = thresh };
+	return __wb_calc_thresh(&gdtc);
 }
 
 /*
···
  *
  * (o) global/bdi setpoints
  *
- * We want the dirty pages be balanced around the global/bdi setpoints.
+ * We want the dirty pages be balanced around the global/wb setpoints.
  * When the number of dirty pages is higher/lower than the setpoint, the
  * dirty position control ratio (and hence task dirty ratelimit) will be
  * decreased/increased to bring the dirty pages back to the setpoint.
···
  *     if (dirty < setpoint) scale up   pos_ratio
  *     if (dirty > setpoint) scale down pos_ratio
  *
- *     if (bdi_dirty < bdi_setpoint) scale up   pos_ratio
- *     if (bdi_dirty > bdi_setpoint) scale down pos_ratio
+ *     if (wb_dirty < wb_setpoint) scale up   pos_ratio
+ *     if (wb_dirty > wb_setpoint) scale down pos_ratio
  *
  *     task_ratelimit = dirty_ratelimit * pos_ratio >> RATELIMIT_CALC_SHIFT
  *
···
  *   0 +------------.------------------.----------------------*------------->
  *           freerun^          setpoint^                 limit^   dirty pages
  *
- * (o) bdi control line
+ * (o) wb control line
  *
  *     ^ pos_ratio
  *     |
···
  *     |                      .                           .
  *     |                      .                             .
  *   0 +----------------------.-------------------------------.------------->
- *                bdi_setpoint^                    x_intercept^
+ *                wb_setpoint^                    x_intercept^
  *
- * The bdi control line won't drop below pos_ratio=1/4, so that bdi_dirty can
+ * The wb control line won't drop below pos_ratio=1/4, so that wb_dirty can
  * be smoothly throttled down to normal if it starts high in situations like
  * - start writing to a slow SD card and a fast disk at the same time. The SD
- *   card's bdi_dirty may rush to many times higher than bdi_setpoint.
- * - the bdi dirty thresh drops quickly due to change of JBOD workload
+ *   card's wb_dirty may rush to many times higher than wb_setpoint.
+ * - the wb dirty thresh drops quickly due to change of JBOD workload
  */
-static unsigned long bdi_position_ratio(struct backing_dev_info *bdi,
-					unsigned long thresh,
-					unsigned long bg_thresh,
-					unsigned long dirty,
-					unsigned long bdi_thresh,
-					unsigned long bdi_dirty)
+static void wb_position_ratio(struct dirty_throttle_control *dtc)
 {
-	unsigned long write_bw = bdi->avg_write_bandwidth;
-	unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh);
-	unsigned long limit = hard_dirty_limit(thresh);
+	struct bdi_writeback *wb = dtc->wb;
+	unsigned long write_bw = wb->avg_write_bandwidth;
+	unsigned long freerun = dirty_freerun_ceiling(dtc->thresh, dtc->bg_thresh);
+	unsigned long limit = hard_dirty_limit(dtc_dom(dtc), dtc->thresh);
+	unsigned long wb_thresh = dtc->wb_thresh;
 	unsigned long x_intercept;
 	unsigned long setpoint;		/* dirty pages' target balance point */
-	unsigned long bdi_setpoint;
+	unsigned long wb_setpoint;
 	unsigned long span;
 	long long pos_ratio;		/* for scaling up/down the rate limit */
 	long x;
 
-	if (unlikely(dirty >= limit))
-		return 0;
+	dtc->pos_ratio = 0;
+
+	if (unlikely(dtc->dirty >= limit))
+		return;
 
 	/*
 	 * global setpoint
···
 	 * See comment for pos_ratio_polynom().
 	 */
 	setpoint = (freerun + limit) / 2;
-	pos_ratio = pos_ratio_polynom(setpoint, dirty, limit);
+	pos_ratio = pos_ratio_polynom(setpoint, dtc->dirty, limit);
 
 	/*
 	 * The strictlimit feature is a tool preventing mistrusted filesystems
 	 * from growing a large number of dirty pages before throttling. For
-	 * such filesystems balance_dirty_pages always checks bdi counters
-	 * against bdi limits. Even if global "nr_dirty" is under "freerun".
+	 * such filesystems balance_dirty_pages always checks wb counters
+	 * against wb limits. Even if global "nr_dirty" is under "freerun".
 	 * This is especially important for fuse which sets bdi->max_ratio to
 	 * 1% by default. Without strictlimit feature, fuse writeback may
 	 * consume arbitrary amount of RAM because it is accounted in
 	 * NR_WRITEBACK_TEMP which is not involved in calculating "nr_dirty".
 	 *
-	 * Here, in bdi_position_ratio(), we calculate pos_ratio based on
-	 * two values: bdi_dirty and bdi_thresh. Let's consider an example:
+	 * Here, in wb_position_ratio(), we calculate pos_ratio based on
+	 * two values: wb_dirty and wb_thresh. Let's consider an example:
 	 * total amount of RAM is 16GB, bdi->max_ratio is equal to 1%, global
 	 * limits are set by default to 10% and 20% (background and throttle).
-	 * Then bdi_thresh is 1% of 20% of 16GB. This amounts to ~8K pages.
-	 * bdi_dirty_limit(bdi, bg_thresh) is about ~4K pages. bdi_setpoint is
-	 * about ~6K pages (as the average of background and throttle bdi
+	 * Then wb_thresh is 1% of 20% of 16GB. This amounts to ~8K pages.
+	 * wb_calc_thresh(wb, bg_thresh) is about ~4K pages. wb_setpoint is
+	 * about ~6K pages (as the average of background and throttle wb
 	 * limits). The 3rd order polynomial will provide positive feedback if
-	 * bdi_dirty is under bdi_setpoint and vice versa.
+	 * wb_dirty is under wb_setpoint and vice versa.
 	 *
 	 * Note, that we cannot use global counters in these calculations
-	 * because we want to throttle process writing to a strictlimit BDI
+	 * because we want to throttle process writing to a strictlimit wb
 	 * much earlier than global "freerun" is reached (~23MB vs. ~2.3GB
 	 * in the example above).
 	 */
-	if (unlikely(bdi->capabilities & BDI_CAP_STRICTLIMIT)) {
-		long long bdi_pos_ratio;
-		unsigned long bdi_bg_thresh;
+	if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) {
+		long long wb_pos_ratio;
 
-		if (bdi_dirty < 8)
-			return min_t(long long, pos_ratio * 2,
-				     2 << RATELIMIT_CALC_SHIFT);
+		if (dtc->wb_dirty < 8) {
+			dtc->pos_ratio = min_t(long long, pos_ratio * 2,
+					   2 << RATELIMIT_CALC_SHIFT);
+			return;
+		}
 
-		if (bdi_dirty >= bdi_thresh)
-			return 0;
+		if (dtc->wb_dirty >= wb_thresh)
+			return;
 
-		bdi_bg_thresh = div_u64((u64)bdi_thresh * bg_thresh, thresh);
-		bdi_setpoint = dirty_freerun_ceiling(bdi_thresh,
-						     bdi_bg_thresh);
+		wb_setpoint = dirty_freerun_ceiling(wb_thresh,
+						    dtc->wb_bg_thresh);
 
-		if (bdi_setpoint == 0 || bdi_setpoint == bdi_thresh)
-			return 0;
+		if (wb_setpoint == 0 || wb_setpoint == wb_thresh)
+			return;
 
-		bdi_pos_ratio = pos_ratio_polynom(bdi_setpoint, bdi_dirty,
-						  bdi_thresh);
+		wb_pos_ratio = pos_ratio_polynom(wb_setpoint, dtc->wb_dirty,
+						 wb_thresh);
 
 		/*
-		 * Typically, for strictlimit case, bdi_setpoint << setpoint
-		 * and pos_ratio >> bdi_pos_ratio. In the other words global
+		 * Typically, for strictlimit case, wb_setpoint << setpoint
+		 * and pos_ratio >> wb_pos_ratio. In the other words global
 		 * state ("dirty") is not limiting factor and we have to
-		 * make decision based on bdi counters. But there is an
+		 * make decision based on wb counters. But there is an
 		 * important case when global pos_ratio should get precedence:
 		 * global limits are exceeded (e.g. due to activities on other
-		 * BDIs) while given strictlimit BDI is below limit.
+		 * wb's) while given strictlimit wb is below limit.
 		 *
-		 * "pos_ratio * bdi_pos_ratio" would work for the case above,
+		 * "pos_ratio * wb_pos_ratio" would work for the case above,
 		 * but it would look too non-natural for the case of all
-		 * activity in the system coming from a single strictlimit BDI
+		 * activity in the system coming from a single strictlimit wb
 		 * with bdi->max_ratio == 100%.
 		 *
 		 * Note that min() below somewhat changes the dynamics of the
 		 * control system. Normally, pos_ratio value can be well over 3
-		 * (when globally we are at freerun and bdi is well below bdi
+		 * (when globally we are at freerun and wb is well below wb
 		 * setpoint). Now the maximum pos_ratio in the same situation
 		 * is 2. We might want to tweak this if we observe the control
 		 * system is too slow to adapt.
 		 */
-		return min(pos_ratio, bdi_pos_ratio);
+		dtc->pos_ratio = min(pos_ratio, wb_pos_ratio);
+		return;
 	}
 
 	/*
 	 * We have computed basic pos_ratio above based on global situation. If
-	 * the bdi is over/under its share of dirty pages, we want to scale
+	 * the wb is over/under its share of dirty pages, we want to scale
 	 * pos_ratio further down/up. That is done by the following mechanism.
 	 */
 
 	/*
-	 * bdi setpoint
+	 * wb setpoint
 	 *
-	 *        f(bdi_dirty) := 1.0 + k * (bdi_dirty - bdi_setpoint)
+	 *        f(wb_dirty) := 1.0 + k * (wb_dirty - wb_setpoint)
 	 *
-	 *                        x_intercept - bdi_dirty
+	 *                        x_intercept - wb_dirty
 	 *                     := --------------------------
-	 *                        x_intercept - bdi_setpoint
+	 *                        x_intercept - wb_setpoint
 	 *
-	 * The main bdi control line is a linear function that subjects to
+	 * The main wb control line is a linear function that subjects to
 	 *
-	 * (1) f(bdi_setpoint) = 1.0
-	 * (2) k = - 1 / (8 * write_bw)  (in single bdi case)
-	 *     or equally: x_intercept = bdi_setpoint + 8 * write_bw
+	 * (1) f(wb_setpoint) = 1.0
+	 * (2) k = - 1 / (8 * write_bw)  (in single wb case)
+	 *     or equally: x_intercept = wb_setpoint + 8 * write_bw
 	 *
-	 * For single bdi case, the dirty pages are observed to fluctuate
+	 * For single wb case, the dirty pages are observed to fluctuate
 	 * regularly within range
-	 *        [bdi_setpoint - write_bw/2, bdi_setpoint + write_bw/2]
+	 *        [wb_setpoint - write_bw/2, wb_setpoint + write_bw/2]
 	 * for various filesystems, where (2) can yield in a reasonable 12.5%
 	 * fluctuation range for pos_ratio.
 	 *
-	 * For JBOD case, bdi_thresh (not bdi_dirty!) could fluctuate up to its
+	 * For JBOD case, wb_thresh (not wb_dirty!) could fluctuate up to its
 	 * own size, so move the slope over accordingly and choose a slope that
-	 * yields 100% pos_ratio fluctuation on suddenly doubled bdi_thresh.
+	 * yields 100% pos_ratio fluctuation on suddenly doubled wb_thresh.
 	 */
-	if (unlikely(bdi_thresh > thresh))
-		bdi_thresh = thresh;
+	if (unlikely(wb_thresh > dtc->thresh))
+		wb_thresh = dtc->thresh;
 	/*
-	 * It's very possible that bdi_thresh is close to 0 not because the
+	 * It's very possible that wb_thresh is close to 0 not because the
 	 * device is slow, but that it has remained inactive for long time.
 	 * Honour such devices a reasonable good (hopefully IO efficient)
 	 * threshold, so that the occasional writes won't be blocked and active
 	 * writes can rampup the threshold quickly.
 	 */
-	bdi_thresh = max(bdi_thresh, (limit - dirty) / 8);
+	wb_thresh = max(wb_thresh, (limit - dtc->dirty) / 8);
 	/*
-	 * scale global setpoint to bdi's:
-	 *	bdi_setpoint = setpoint * bdi_thresh / thresh
+	 * scale global setpoint to wb's:
+	 *	wb_setpoint = setpoint * wb_thresh / thresh
 	 */
-	x = div_u64((u64)bdi_thresh << 16, thresh | 1);
-	bdi_setpoint = setpoint * (u64)x >> 16;
+	x = div_u64((u64)wb_thresh << 16, dtc->thresh | 1);
+	wb_setpoint = setpoint * (u64)x >> 16;
 	/*
-	 * Use span=(8*write_bw) in single bdi case as indicated by
-	 * (thresh - bdi_thresh ~= 0) and transit to bdi_thresh in JBOD case.
+	 * Use span=(8*write_bw) in single wb case as indicated by
+	 * (thresh - wb_thresh ~= 0) and transit to wb_thresh in JBOD case.
 	 *
-	 *        bdi_thresh                    thresh - bdi_thresh
-	 * span = ---------- * (8 * write_bw) + ------------------- * bdi_thresh
-	 *          thresh                            thresh
+	 *        wb_thresh                    thresh - wb_thresh
+	 * span = --------- * (8 * write_bw) + ------------------ * wb_thresh
+	 *         thresh                           thresh
 	 */
-	span = (thresh - bdi_thresh + 8 * write_bw) * (u64)x >> 16;
-	x_intercept = bdi_setpoint + span;
+	span = (dtc->thresh - wb_thresh + 8 * write_bw) * (u64)x >> 16;
+	x_intercept = wb_setpoint + span;
 
-	if (bdi_dirty < x_intercept - span / 4) {
-		pos_ratio = div64_u64(pos_ratio * (x_intercept - bdi_dirty),
-				      (x_intercept - bdi_setpoint) | 1);
+	if (dtc->wb_dirty < x_intercept - span / 4) {
+		pos_ratio = div64_u64(pos_ratio * (x_intercept - dtc->wb_dirty),
+				      (x_intercept - wb_setpoint) | 1);
 	} else
 		pos_ratio /= 4;
 
 	/*
-	 * bdi reserve area, safeguard against dirty pool underrun and disk idle
+	 * wb reserve area, safeguard against dirty pool underrun and disk idle
 	 * It may push the desired control point of global dirty pages higher
 	 * than setpoint.
 	 */
-	x_intercept = bdi_thresh / 2;
-	if (bdi_dirty < x_intercept) {
-		if (bdi_dirty > x_intercept / 8)
-			pos_ratio = div_u64(pos_ratio * x_intercept, bdi_dirty);
+	x_intercept = wb_thresh / 2;
+	if (dtc->wb_dirty < x_intercept) {
+		if (dtc->wb_dirty > x_intercept / 8)
+			pos_ratio = div_u64(pos_ratio * x_intercept,
+					    dtc->wb_dirty);
 		else
 			pos_ratio *= 8;
 	}
 
-	return pos_ratio;
+	dtc->pos_ratio = pos_ratio;
 }
 
-static void bdi_update_write_bandwidth(struct backing_dev_info *bdi,
-				       unsigned long elapsed,
-				       unsigned long written)
+static void wb_update_write_bandwidth(struct bdi_writeback *wb,
+				      unsigned long elapsed,
+				      unsigned long written)
 {
 	const unsigned long period = roundup_pow_of_two(3 * HZ);
-	unsigned long avg = bdi->avg_write_bandwidth;
-	unsigned long old = bdi->write_bandwidth;
+	unsigned long avg = wb->avg_write_bandwidth;
+	unsigned long old = wb->write_bandwidth;
 	u64 bw;
 
 	/*
···
 	 * @written may have decreased due to account_page_redirty().
 	 * Avoid underflowing @bw calculation.
 	 */
-	bw = written - min(written, bdi->written_stamp);
+	bw = written - min(written, wb->written_stamp);
 	bw *= HZ;
 	if (unlikely(elapsed > period)) {
 		do_div(bw, elapsed);
 		avg = bw;
 		goto out;
 	}
-	bw += (u64)bdi->write_bandwidth * (period - elapsed);
+	bw += (u64)wb->write_bandwidth * (period - elapsed);
 	bw >>= ilog2(period);
 
 	/*
···
 		avg += (old - avg) >> 3;
 
 out:
-	bdi->write_bandwidth = bw;
-	bdi->avg_write_bandwidth = avg;
+	/* keep avg > 0 to guarantee that tot > 0 if there are dirty wbs */
+	avg = max(avg, 1LU);
+	if (wb_has_dirty_io(wb)) {
+		long delta = avg - wb->avg_write_bandwidth;
+		WARN_ON_ONCE(atomic_long_add_return(delta,
+					&wb->bdi->tot_write_bandwidth) <= 0);
+	}
+	wb->write_bandwidth = bw;
+	wb->avg_write_bandwidth = avg;
 }
 
-/*
- * The global dirtyable memory and dirty threshold could be suddenly knocked
- * down by a large amount (eg. on the startup of KVM in a swapless system).
- * This may throw the system into deep dirty exceeded state and throttle
- * heavy/light dirtiers alike. To retain good responsiveness, maintain
- * global_dirty_limit for tracking slowly down to the knocked down dirty
- * threshold.
- */
-static void update_dirty_limit(unsigned long thresh, unsigned long dirty)
+static void update_dirty_limit(struct dirty_throttle_control *dtc)
 {
-	unsigned long limit = global_dirty_limit;
+	struct wb_domain *dom = dtc_dom(dtc);
+	unsigned long thresh = dtc->thresh;
+	unsigned long limit = dom->dirty_limit;
 
 	/*
 	 * Follow up in one step.
···
 	/*
 	 * Follow down slowly. Use the higher one as the target, because thresh
 	 * may drop below dirty. This is exactly the reason to introduce
-	 * global_dirty_limit which is guaranteed to lie above the dirty pages.
+	 * dom->dirty_limit which is guaranteed to lie above the dirty pages.
 	 */
-	thresh = max(thresh, dirty);
+	thresh = max(thresh, dtc->dirty);
 	if (limit > thresh) {
 		limit -= (limit - thresh) >> 5;
 		goto update;
 	}
 	return;
 update:
-	global_dirty_limit = limit;
+	dom->dirty_limit = limit;
 }
 
-static void global_update_bandwidth(unsigned long thresh,
-				    unsigned long dirty,
+static void domain_update_bandwidth(struct dirty_throttle_control *dtc,
 				    unsigned long now)
 {
-	static DEFINE_SPINLOCK(dirty_lock);
-	static unsigned long update_time = INITIAL_JIFFIES;
+	struct wb_domain *dom = dtc_dom(dtc);
 
 	/*
 	 * check locklessly first to optimize away locking for the most time
 	 */
-	if (time_before(now, update_time + BANDWIDTH_INTERVAL))
+	if (time_before(now, dom->dirty_limit_tstamp + BANDWIDTH_INTERVAL))
 		return;
 
-	spin_lock(&dirty_lock);
-	if (time_after_eq(now, update_time + BANDWIDTH_INTERVAL)) {
-		update_dirty_limit(thresh, dirty);
-		update_time = now;
+	spin_lock(&dom->lock);
+	if (time_after_eq(now, dom->dirty_limit_tstamp + BANDWIDTH_INTERVAL)) {
+		update_dirty_limit(dtc);
+		dom->dirty_limit_tstamp = now;
 	}
-	spin_unlock(&dirty_lock);
+	spin_unlock(&dom->lock);
 }
 
 /*
- * Maintain bdi->dirty_ratelimit, the base dirty throttle rate.
+ * Maintain wb->dirty_ratelimit, the base dirty throttle rate.
  *
- * Normal bdi tasks will be curbed at or below it in long term.
+ * Normal wb tasks will be curbed at or below it in long term.
  * Obviously it should be around (write_bw / N) when there are N dd tasks.
  */
-static void bdi_update_dirty_ratelimit(struct backing_dev_info *bdi,
-				       unsigned long thresh,
-				       unsigned long bg_thresh,
-				       unsigned long dirty,
-				       unsigned long bdi_thresh,
-				       unsigned long bdi_dirty,
-				       unsigned long dirtied,
-				       unsigned long elapsed)
+static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc,
+				      unsigned long dirtied,
+				      unsigned long elapsed)
 {
-	unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh);
-	unsigned long limit = hard_dirty_limit(thresh);
+	struct bdi_writeback *wb = dtc->wb;
+	unsigned long dirty = dtc->dirty;
+	unsigned long freerun = dirty_freerun_ceiling(dtc->thresh, dtc->bg_thresh);
+	unsigned long limit = hard_dirty_limit(dtc_dom(dtc), dtc->thresh);
 	unsigned long setpoint = (freerun + limit) / 2;
-	unsigned long write_bw = bdi->avg_write_bandwidth;
-	unsigned long dirty_ratelimit = bdi->dirty_ratelimit;
+	unsigned long write_bw = wb->avg_write_bandwidth;
+	unsigned long dirty_ratelimit = wb->dirty_ratelimit;
 	unsigned long dirty_rate;
 	unsigned long task_ratelimit;
 	unsigned long balanced_dirty_ratelimit;
-	unsigned long pos_ratio;
 	unsigned long step;
 	unsigned long x;
 
···
 	 * The dirty rate will match the writeout rate in long term, except
 	 * when dirty pages are truncated by userspace or re-dirtied by FS.
 	 */
-	dirty_rate = (dirtied - bdi->dirtied_stamp) * HZ / elapsed;
+	dirty_rate = (dirtied - wb->dirtied_stamp) * HZ / elapsed;
 
-	pos_ratio = bdi_position_ratio(bdi, thresh, bg_thresh, dirty,
-				       bdi_thresh, bdi_dirty);
 	/*
 	 * task_ratelimit reflects each dd's dirty rate for the past 200ms.
 	 */
 	task_ratelimit = (u64)dirty_ratelimit *
-					pos_ratio >> RATELIMIT_CALC_SHIFT;
+					dtc->pos_ratio >> RATELIMIT_CALC_SHIFT;
 	task_ratelimit++; /* it helps rampup dirty_ratelimit from tiny values */
 
 	/*
 	 * A linear estimation of the "balanced" throttle rate. The theory is,
-	 * if there are N dd tasks, each throttled at task_ratelimit, the bdi's
+	 * if there are N dd tasks, each throttled at task_ratelimit, the wb's
 	 * dirty_rate will be measured to be (N * task_ratelimit). So the below
 	 * formula will yield the balanced rate limit (write_bw / N).
 	 *
···
 	/*
 	 * We could safely do this and return immediately:
 	 *
-	 *	bdi->dirty_ratelimit = balanced_dirty_ratelimit;
+	 *	wb->dirty_ratelimit = balanced_dirty_ratelimit;
 	 *
 	 * However to get a more stable dirty_ratelimit, the below elaborated
 	 * code makes use of task_ratelimit to filter out singular points and
···
 	step = 0;
 
 	/*
-	 * For strictlimit case, calculations above were based on bdi counters
-	 * and limits (starting from pos_ratio = bdi_position_ratio() and up to
+	 * For strictlimit case, calculations above were based on wb counters
+	 * and limits (starting from pos_ratio = wb_position_ratio() and up to
 	 * balanced_dirty_ratelimit = task_ratelimit * write_bw / dirty_rate).
-	 * Hence, to calculate "step" properly, we have to use bdi_dirty as
-	 * "dirty" and bdi_setpoint as "setpoint".
+	 * Hence, to calculate "step" properly, we have to use wb_dirty as
+	 * "dirty" and wb_setpoint as "setpoint".
 	 *
-	 * We rampup dirty_ratelimit forcibly if bdi_dirty is low because
-	 * it's possible that bdi_thresh is close to zero due to inactivity
-	 * of backing device (see the implementation of bdi_dirty_limit()).
+	 * We rampup dirty_ratelimit forcibly if wb_dirty is low because
+	 * it's possible that wb_thresh is close to zero due to inactivity
+	 * of backing device.
 	 */
-	if (unlikely(bdi->capabilities & BDI_CAP_STRICTLIMIT)) {
-		dirty = bdi_dirty;
-		if (bdi_dirty < 8)
-			setpoint = bdi_dirty + 1;
+	if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) {
+		dirty = dtc->wb_dirty;
+		if (dtc->wb_dirty < 8)
+			setpoint = dtc->wb_dirty + 1;
 		else
-			setpoint = (bdi_thresh +
-				    bdi_dirty_limit(bdi, bg_thresh)) / 2;
+			setpoint = (dtc->wb_thresh + dtc->wb_bg_thresh) / 2;
 	}
 
 	if (dirty < setpoint) {
-		x = min3(bdi->balanced_dirty_ratelimit,
+		x = min3(wb->balanced_dirty_ratelimit,
 			 balanced_dirty_ratelimit, task_ratelimit);
 		if (dirty_ratelimit < x)
 			step = x - dirty_ratelimit;
 	} else {
-		x = max3(bdi->balanced_dirty_ratelimit,
+		x = max3(wb->balanced_dirty_ratelimit,
 			 balanced_dirty_ratelimit, task_ratelimit);
 		if (dirty_ratelimit > x)
 			step = dirty_ratelimit - x;
···
 	else
 		dirty_ratelimit -= step;
 
-	bdi->dirty_ratelimit = max(dirty_ratelimit, 1UL);
-	bdi->balanced_dirty_ratelimit = balanced_dirty_ratelimit;
+	wb->dirty_ratelimit = max(dirty_ratelimit, 1UL);
+	wb->balanced_dirty_ratelimit = balanced_dirty_ratelimit;
 
-	trace_bdi_dirty_ratelimit(bdi, dirty_rate, task_ratelimit);
+	trace_bdi_dirty_ratelimit(wb->bdi, dirty_rate, task_ratelimit);
 }
 
-void __bdi_update_bandwidth(struct backing_dev_info *bdi,
-			    unsigned long thresh,
-			    unsigned long bg_thresh,
-			    unsigned long dirty,
-			    unsigned long bdi_thresh,
-			    unsigned long bdi_dirty,
-			    unsigned long start_time)
+static void __wb_update_bandwidth(struct dirty_throttle_control *gdtc,
+				  struct dirty_throttle_control *mdtc,
+				  unsigned long start_time,
+				  bool update_ratelimit)
 {
+	struct bdi_writeback *wb = gdtc->wb;
 	unsigned long now = jiffies;
-	unsigned long elapsed = now - bdi->bw_time_stamp;
+	unsigned long elapsed = now - wb->bw_time_stamp;
 	unsigned long dirtied;
 	unsigned long written;
+
+	lockdep_assert_held(&wb->list_lock);
 
 	/*
 	 * rate-limit, only update once every 200ms.
···
 	if (elapsed < BANDWIDTH_INTERVAL)
 		return;
 
-	dirtied = percpu_counter_read(&bdi->bdi_stat[BDI_DIRTIED]);
-	written = percpu_counter_read(&bdi->bdi_stat[BDI_WRITTEN]);
+	dirtied = percpu_counter_read(&wb->stat[WB_DIRTIED]);
+	written = percpu_counter_read(&wb->stat[WB_WRITTEN]);
 
 	/*
 	 * Skip quiet periods when disk bandwidth is under-utilized.
 	 * (at least 1s idle time between two flusher runs)
 	 */
-	if (elapsed > HZ && time_before(bdi->bw_time_stamp, start_time))
+	if (elapsed > HZ && time_before(wb->bw_time_stamp, start_time))
 		goto snapshot;
 
-	if (thresh) {
-		global_update_bandwidth(thresh, dirty, now);
-		bdi_update_dirty_ratelimit(bdi, thresh, bg_thresh, dirty,
-					   bdi_thresh, bdi_dirty,
-					   dirtied, elapsed);
+	if (update_ratelimit) {
+		domain_update_bandwidth(gdtc, now);
+		wb_update_dirty_ratelimit(gdtc, dirtied, elapsed);
+
+		/*
+		 * @mdtc is always NULL if !CGROUP_WRITEBACK but the
+		 * compiler has no way to figure that out.  Help it.
+		 */
+		if (IS_ENABLED(CONFIG_CGROUP_WRITEBACK) && mdtc) {
+			domain_update_bandwidth(mdtc, now);
+			wb_update_dirty_ratelimit(mdtc, dirtied, elapsed);
+		}
 	}
-	bdi_update_write_bandwidth(bdi, elapsed, written);
+	wb_update_write_bandwidth(wb, elapsed, written);
 
 snapshot:
-	bdi->dirtied_stamp = dirtied;
-	bdi->written_stamp = written;
-	bdi->bw_time_stamp = now;
+	wb->dirtied_stamp = dirtied;
+	wb->written_stamp = written;
+	wb->bw_time_stamp = now;
 }
 
-static void bdi_update_bandwidth(struct backing_dev_info *bdi,
-				 unsigned long thresh,
-				 unsigned long bg_thresh,
-				 unsigned long dirty,
-				 unsigned long bdi_thresh,
-				 unsigned long bdi_dirty,
-				 unsigned long start_time)
+void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time)
 {
-	if (time_is_after_eq_jiffies(bdi->bw_time_stamp + BANDWIDTH_INTERVAL))
-		return;
-	spin_lock(&bdi->wb.list_lock);
-	__bdi_update_bandwidth(bdi, thresh, bg_thresh, dirty,
-			       bdi_thresh, bdi_dirty, start_time);
-	spin_unlock(&bdi->wb.list_lock);
+	struct dirty_throttle_control gdtc = { GDTC_INIT(wb) };
+
+	__wb_update_bandwidth(&gdtc, NULL, start_time, false);
 }
 
 /*
···
 	return 1;
 }
 
-static unsigned long bdi_max_pause(struct backing_dev_info *bdi,
-				   unsigned long bdi_dirty)
+static unsigned long wb_max_pause(struct bdi_writeback *wb,
+				  unsigned long wb_dirty)
 {
-	unsigned long bw = bdi->avg_write_bandwidth;
+	unsigned long bw = wb->avg_write_bandwidth;
 	unsigned long t;
 
 	/*
···
 	 *
 	 * 8 serves as the safety ratio.
 	 */
-	t = bdi_dirty / (1 + bw / roundup_pow_of_two(1 + HZ / 8));
+	t = wb_dirty / (1 + bw / roundup_pow_of_two(1 + HZ / 8));
 	t++;
 
 	return min_t(unsigned long, t, MAX_PAUSE);
 }
 
-static long bdi_min_pause(struct backing_dev_info *bdi,
-			  long max_pause,
-			  unsigned long task_ratelimit,
-			  unsigned long dirty_ratelimit,
-			  int *nr_dirtied_pause)
+static long wb_min_pause(struct bdi_writeback *wb,
+			 long max_pause,
+			 unsigned long task_ratelimit,
+			 unsigned long dirty_ratelimit,
+			 int *nr_dirtied_pause)
 {
-	long hi = ilog2(bdi->avg_write_bandwidth);
-	long lo = ilog2(bdi->dirty_ratelimit);
+	long hi = ilog2(wb->avg_write_bandwidth);
+	long lo = ilog2(wb->dirty_ratelimit);
 	long t;		/* target pause */
 	long pause;	/* estimated next pause */
 	int pages;	/* target nr_dirtied_pause */
···
 	return pages >= DIRTY_POLL_THRESH ? 1 + t / 2 : t;
 }
 
-static inline void bdi_dirty_limits(struct backing_dev_info *bdi,
-				    unsigned long dirty_thresh,
-				    unsigned long background_thresh,
-				    unsigned long *bdi_dirty,
-				    unsigned long *bdi_thresh,
-				    unsigned long *bdi_bg_thresh)
+static inline void wb_dirty_limits(struct dirty_throttle_control *dtc)
 {
-	unsigned long bdi_reclaimable;
+	struct bdi_writeback *wb = dtc->wb;
+	unsigned long wb_reclaimable;
 
 	/*
-	 * bdi_thresh is not treated as some limiting factor as
+	 * wb_thresh is not treated as some limiting factor as
 	 * dirty_thresh, due to reasons
-	 * - in JBOD setup, bdi_thresh can fluctuate a lot
+	 * - in JBOD setup, wb_thresh can fluctuate a lot
 	 * - in a system with HDD and USB key, the USB key may somehow
-	 *   go into state (bdi_dirty >> bdi_thresh) either because
-	 *   bdi_dirty starts high, or because bdi_thresh drops low.
+	 *   go into state (wb_dirty >> wb_thresh) either because
+	 *   wb_dirty starts high, or because wb_thresh drops low.
 	 *   In this case we don't want to hard throttle the USB key
-	 *   dirtiers for 100 seconds until bdi_dirty drops under
-	 *   bdi_thresh. Instead the auxiliary bdi control line in
-	 *   bdi_position_ratio() will let the dirtier task progress
-	 *   at some rate <= (write_bw / 2) for bringing down bdi_dirty.
+	 *   dirtiers for 100 seconds until wb_dirty drops under
+	 *   wb_thresh. Instead the auxiliary wb control line in
+	 *   wb_position_ratio() will let the dirtier task progress
+	 *   at some rate <= (write_bw / 2) for bringing down wb_dirty.
 	 */
-	*bdi_thresh = bdi_dirty_limit(bdi, dirty_thresh);
-
-	if (bdi_bg_thresh)
-		*bdi_bg_thresh = dirty_thresh ? div_u64((u64)*bdi_thresh *
-							background_thresh,
-							dirty_thresh) : 0;
+	dtc->wb_thresh = __wb_calc_thresh(dtc);
+	dtc->wb_bg_thresh = dtc->thresh ?
+		div_u64((u64)dtc->wb_thresh * dtc->bg_thresh, dtc->thresh) : 0;
 
 	/*
 	 * In order to avoid the stacked BDI deadlock we need
···
 	 * actually dirty; with m+n sitting in the percpu
 	 * deltas.
 	 */
-	if (*bdi_thresh < 2 * bdi_stat_error(bdi)) {
-		bdi_reclaimable = bdi_stat_sum(bdi, BDI_RECLAIMABLE);
-		*bdi_dirty = bdi_reclaimable +
-			bdi_stat_sum(bdi, BDI_WRITEBACK);
+	if (dtc->wb_thresh < 2 * wb_stat_error(wb)) {
+		wb_reclaimable = wb_stat_sum(wb, WB_RECLAIMABLE);
+		dtc->wb_dirty = wb_reclaimable + wb_stat_sum(wb, WB_WRITEBACK);
 	} else {
-		bdi_reclaimable = bdi_stat(bdi, BDI_RECLAIMABLE);
-		*bdi_dirty = bdi_reclaimable +
-			bdi_stat(bdi, BDI_WRITEBACK);
+		wb_reclaimable = wb_stat(wb, WB_RECLAIMABLE);
+		dtc->wb_dirty = wb_reclaimable + wb_stat(wb, WB_WRITEBACK);
 	}
 }
 
···
  * perform some writeout.
  */
 static void balance_dirty_pages(struct address_space *mapping,
+				struct bdi_writeback *wb,
 				unsigned long pages_dirtied)
 {
+	struct dirty_throttle_control gdtc_stor = { GDTC_INIT(wb) };
+	struct dirty_throttle_control mdtc_stor = { MDTC_INIT(wb, &gdtc_stor) };
+	struct dirty_throttle_control * const gdtc = &gdtc_stor;
+	struct dirty_throttle_control * const mdtc = mdtc_valid(&mdtc_stor) ?
+						     &mdtc_stor : NULL;
+	struct dirty_throttle_control *sdtc;
 	unsigned long nr_reclaimable;	/* = file_dirty + unstable_nfs */
-	unsigned long nr_dirty;  /* = file_dirty + writeback + unstable_nfs */
-	unsigned long background_thresh;
-	unsigned long dirty_thresh;
 	long period;
 	long pause;
 	long max_pause;
···
 	bool dirty_exceeded = false;
 	unsigned long task_ratelimit;
 	unsigned long dirty_ratelimit;
-	unsigned long pos_ratio;
-	struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+	struct backing_dev_info *bdi = wb->bdi;
 	bool strictlimit = bdi->capabilities & BDI_CAP_STRICTLIMIT;
 	unsigned long start_time = jiffies;
 
 	for (;;) {
 		unsigned long now = jiffies;
-		unsigned long uninitialized_var(bdi_thresh);
-		unsigned long thresh;
-		unsigned long uninitialized_var(bdi_dirty);
-		unsigned long dirty;
-		unsigned long bg_thresh;
+		unsigned long dirty, thresh, bg_thresh;
+		unsigned long m_dirty, m_thresh, m_bg_thresh;
 
 		/*
 		 * Unstable writes are a feature of certain networked
···
 		 */
 		nr_reclaimable = global_page_state(NR_FILE_DIRTY) +
 					global_page_state(NR_UNSTABLE_NFS);
-		nr_dirty = nr_reclaimable + global_page_state(NR_WRITEBACK);
+		gdtc->avail = global_dirtyable_memory();
+		gdtc->dirty = nr_reclaimable + global_page_state(NR_WRITEBACK);
 
-		global_dirty_limits(&background_thresh, &dirty_thresh);
+		domain_dirty_limits(gdtc);
 
 		if (unlikely(strictlimit)) {
-			bdi_dirty_limits(bdi, dirty_thresh, background_thresh,
-					 &bdi_dirty, &bdi_thresh, &bg_thresh);
+			wb_dirty_limits(gdtc);
 
-			dirty = bdi_dirty;
-			thresh = bdi_thresh;
+			dirty = gdtc->wb_dirty;
+			thresh = gdtc->wb_thresh;
+			bg_thresh = gdtc->wb_bg_thresh;
 		} else {
-			dirty = nr_dirty;
-			thresh = dirty_thresh;
-			bg_thresh = background_thresh;
+			dirty = gdtc->dirty;
+			thresh = gdtc->thresh;
+			bg_thresh = gdtc->bg_thresh;
+		}
+
+		if (mdtc) {
+			unsigned long writeback;
+
+			/*
+			 * If @wb belongs to !root memcg, repeat the same
+			 * basic calculations for the memcg domain.
+			 */
+			mem_cgroup_wb_stats(wb, &mdtc->avail, &mdtc->dirty,
+					    &writeback);
+			mdtc_cap_avail(mdtc);
+			mdtc->dirty += writeback;
+
+			domain_dirty_limits(mdtc);
+
+			if (unlikely(strictlimit)) {
+				wb_dirty_limits(mdtc);
+				m_dirty = mdtc->wb_dirty;
+				m_thresh = mdtc->wb_thresh;
+				m_bg_thresh = mdtc->wb_bg_thresh;
+			} else {
+				m_dirty = mdtc->dirty;
+				m_thresh = mdtc->thresh;
+				m_bg_thresh = mdtc->bg_thresh;
+			}
 		}
 
 		/*
 		 * Throttle it only when the background writeback cannot
 		 * catch-up. This avoids (excessively) small writeouts
-		 * when the bdi limits are ramping up in case of !strictlimit.
+		 * when the wb limits are ramping up in case of !strictlimit.
 		 *
-		 * In strictlimit case make decision based on the bdi counters
-		 * and limits. Small writeouts when the bdi limits are ramping
+		 * In strictlimit case make decision based on the wb counters
+		 * and limits. Small writeouts when the wb limits are ramping
 		 * up are the price we consciously pay for strictlimit-ing.
+		 *
+		 * If memcg domain is in effect, @dirty should be under
+		 * both global and memcg freerun ceilings.
 		 */
-		if (dirty <= dirty_freerun_ceiling(thresh, bg_thresh)) {
+		if (dirty <= dirty_freerun_ceiling(thresh, bg_thresh) &&
+		    (!mdtc ||
+		     m_dirty <= dirty_freerun_ceiling(m_thresh, m_bg_thresh))) {
+			unsigned long intv = dirty_poll_interval(dirty, thresh);
+			unsigned long m_intv = ULONG_MAX;
+
 			current->dirty_paused_when = now;
 			current->nr_dirtied = 0;
-			current->nr_dirtied_pause =
-				dirty_poll_interval(dirty, thresh);
+			if (mdtc)
+				m_intv = dirty_poll_interval(m_dirty, m_thresh);
+			current->nr_dirtied_pause = min(intv, m_intv);
 			break;
 		}
 
-		if (unlikely(!writeback_in_progress(bdi)))
-			bdi_start_background_writeback(bdi);
+		if (unlikely(!writeback_in_progress(wb)))
+			wb_start_background_writeback(wb);
 
+		/*
+		 * Calculate global domain's pos_ratio and select the
+		 * global dtc by default.
+		 */
 		if (!strictlimit)
-			bdi_dirty_limits(bdi, dirty_thresh, background_thresh,
-					 &bdi_dirty, &bdi_thresh, NULL);
+			wb_dirty_limits(gdtc);
 
-		dirty_exceeded = (bdi_dirty > bdi_thresh) &&
-				 ((nr_dirty > dirty_thresh) || strictlimit);
-		if (dirty_exceeded && !bdi->dirty_exceeded)
-			bdi->dirty_exceeded = 1;
+		dirty_exceeded = (gdtc->wb_dirty > gdtc->wb_thresh) &&
+			((gdtc->dirty > gdtc->thresh) || strictlimit);
 
-		bdi_update_bandwidth(bdi, dirty_thresh, background_thresh,
-				     nr_dirty, bdi_thresh, bdi_dirty,
-				     start_time);
+		wb_position_ratio(gdtc);
+		sdtc = gdtc;
 
-		dirty_ratelimit = bdi->dirty_ratelimit;
-		pos_ratio = bdi_position_ratio(bdi, dirty_thresh,
-					       background_thresh, nr_dirty,
-					       bdi_thresh, bdi_dirty);
-		task_ratelimit = ((u64)dirty_ratelimit * pos_ratio) >>
+		if (mdtc) {
+			/*
+			 * If memcg domain is in effect, calculate its
+			 * pos_ratio.  @wb should satisfy constraints from
+			 * both global and memcg domains.  Choose the one
+			 * w/ lower pos_ratio.
+			 */
+			if (!strictlimit)
+				wb_dirty_limits(mdtc);
+
+			dirty_exceeded |= (mdtc->wb_dirty > mdtc->wb_thresh) &&
+				((mdtc->dirty > mdtc->thresh) || strictlimit);
+
+			wb_position_ratio(mdtc);
+			if (mdtc->pos_ratio < gdtc->pos_ratio)
+				sdtc = mdtc;
+		}
+
+		if (dirty_exceeded && !wb->dirty_exceeded)
+			wb->dirty_exceeded = 1;
+
+		if (time_is_before_jiffies(wb->bw_time_stamp +
+					   BANDWIDTH_INTERVAL)) {
+			spin_lock(&wb->list_lock);
+			__wb_update_bandwidth(gdtc, mdtc, start_time, true);
+			spin_unlock(&wb->list_lock);
+		}
+
+		/* throttle according to the chosen dtc */
+		dirty_ratelimit = wb->dirty_ratelimit;
+		task_ratelimit = ((u64)dirty_ratelimit * sdtc->pos_ratio) >>
 							RATELIMIT_CALC_SHIFT;
-		max_pause = bdi_max_pause(bdi, bdi_dirty);
-		min_pause = bdi_min_pause(bdi, max_pause,
-					  task_ratelimit, dirty_ratelimit,
-					  &nr_dirtied_pause);
+		max_pause = wb_max_pause(wb, sdtc->wb_dirty);
+		min_pause = wb_min_pause(wb, max_pause,
+					 task_ratelimit, dirty_ratelimit,
+					 &nr_dirtied_pause);
 
 		if (unlikely(task_ratelimit == 0)) {
 			period = max_pause;
···
 		 */
 		if (pause < min_pause) {
 			trace_balance_dirty_pages(bdi,
-						  dirty_thresh,
-						  background_thresh,
-						  nr_dirty,
-						  bdi_thresh,
-						  bdi_dirty,
+						  sdtc->thresh,
+						  sdtc->bg_thresh,
+						  sdtc->dirty,
+						  sdtc->wb_thresh,
+						  sdtc->wb_dirty,
 						  dirty_ratelimit,
 						  task_ratelimit,
 						  pages_dirtied,
···
 
 pause:
 		trace_balance_dirty_pages(bdi,
-					  dirty_thresh,
-					  background_thresh,
-					  nr_dirty,
-					  bdi_thresh,
-					  bdi_dirty,
+					  sdtc->thresh,
+					  sdtc->bg_thresh,
+					  sdtc->dirty,
+					  sdtc->wb_thresh,
+					  sdtc->wb_dirty,
 					  dirty_ratelimit,
 					  task_ratelimit,
 					  pages_dirtied,
···
 		current->nr_dirtied_pause = nr_dirtied_pause;
 
 		/*
-		 * This is typically equal to (nr_dirty < dirty_thresh) and can
-		 * also keep "1000+ dd on a slow USB stick" under control.
+		 * This is typically equal to (dirty < thresh) and can also
+		 * keep "1000+ dd on a slow USB stick" under control.
 		 */
 		if (task_ratelimit)
 			break;
 
 		/*
 		 * In the case of an unresponding NFS server and the NFS dirty
-		 * pages exceeds dirty_thresh, give the other good bdi's a pipe
+		 * pages exceeds dirty_thresh, give the other good wb's a pipe
 		 * to go through, so that tasks on them still remain responsive.
 		 *
 		 * In theory 1 page is enough to keep the comsumer-producer
 		 * pipe going: the flusher cleans 1 page => the task dirties 1
-		 * more page. However bdi_dirty has accounting errors.  So use
-		 * the larger and more IO friendly bdi_stat_error.
+		 * more page. However wb_dirty has accounting errors.  So use
+		 * the larger and more IO friendly wb_stat_error.
 		 */
-		if (bdi_dirty <= bdi_stat_error(bdi))
+		if (sdtc->wb_dirty <= wb_stat_error(wb))
 			break;
 
 		if (fatal_signal_pending(current))
 			break;
 	}
 
-	if (!dirty_exceeded && bdi->dirty_exceeded)
-		bdi->dirty_exceeded = 0;
+	if (!dirty_exceeded && wb->dirty_exceeded)
+		wb->dirty_exceeded = 0;
 
-	if (writeback_in_progress(bdi))
+	if (writeback_in_progress(wb))
 		return;
 
 	/*
···
 	if (laptop_mode)
 		return;
 
-	if (nr_reclaimable > background_thresh)
-		bdi_start_background_writeback(bdi);
+	if (nr_reclaimable > gdtc->bg_thresh)
+		wb_start_background_writeback(wb);
 }
 
 static DEFINE_PER_CPU(int, bdp_ratelimits);
···
  */
 void balance_dirty_pages_ratelimited(struct address_space *mapping)
 {
-	struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+	struct inode *inode = mapping->host;
+	struct backing_dev_info *bdi = inode_to_bdi(inode);
+	struct bdi_writeback *wb = NULL;
 	int ratelimit;
 	int *p;
 
 	if (!bdi_cap_account_dirty(bdi))
 		return;
 
+	if (inode_cgwb_enabled(inode))
+		wb = wb_get_create_current(bdi, GFP_KERNEL);
+	if (!wb)
+		wb = &bdi->wb;
+
 	ratelimit = current->nr_dirtied_pause;
-	if (bdi->dirty_exceeded)
+	if (wb->dirty_exceeded)
 		ratelimit = min(ratelimit, 32 >> (PAGE_SHIFT - 10));
 
 	preempt_disable();
···
 	preempt_enable();
 
 	if (unlikely(current->nr_dirtied >= ratelimit))
-		balance_dirty_pages(mapping, current->nr_dirtied);
+		balance_dirty_pages(mapping, wb, current->nr_dirtied);
+
+	wb_put(wb);
 }
 EXPORT_SYMBOL(balance_dirty_pages_ratelimited);
+
+/**
+ * wb_over_bg_thresh - does @wb need to be written back?
+ * @wb: bdi_writeback of interest
+ *
+ * Determines whether background writeback should keep writing @wb or it's
+ * clean enough.  Returns %true if writeback should continue.
+ */
+bool wb_over_bg_thresh(struct bdi_writeback *wb)
+{
+	struct dirty_throttle_control gdtc_stor = { GDTC_INIT(wb) };
+	struct dirty_throttle_control mdtc_stor = { MDTC_INIT(wb, &gdtc_stor) };
+	struct dirty_throttle_control * const gdtc = &gdtc_stor;
+	struct dirty_throttle_control * const mdtc = mdtc_valid(&mdtc_stor) ?
+						     &mdtc_stor : NULL;
+
+	/*
+	 * Similar to balance_dirty_pages() but ignores pages being written
+	 * as we're trying to decide whether to put more under writeback.
+	 */
+	gdtc->avail = global_dirtyable_memory();
+	gdtc->dirty = global_page_state(NR_FILE_DIRTY) +
+		      global_page_state(NR_UNSTABLE_NFS);
+	domain_dirty_limits(gdtc);
+
+	if (gdtc->dirty > gdtc->bg_thresh)
+		return true;
+
+	if (wb_stat(wb, WB_RECLAIMABLE) > __wb_calc_thresh(gdtc))
+		return true;
+
+	if (mdtc) {
+		unsigned long writeback;
+
+		mem_cgroup_wb_stats(wb, &mdtc->avail, &mdtc->dirty, &writeback);
+		mdtc_cap_avail(mdtc);
+		domain_dirty_limits(mdtc);	/* ditto, ignore writeback */
+
+		if (mdtc->dirty > mdtc->bg_thresh)
+			return true;
+
+		if (wb_stat(wb, WB_RECLAIMABLE) > __wb_calc_thresh(mdtc))
+			return true;
+	}
+
+	return false;
+}
 
 void throttle_vm_writeout(gfp_t gfp_mask)
 {
···
 
         for ( ; ; ) {
 		global_dirty_limits(&background_thresh, &dirty_thresh);
-		dirty_thresh = hard_dirty_limit(dirty_thresh);
+		dirty_thresh = hard_dirty_limit(&global_wb_domain, dirty_thresh);
 
                 /*
                  * Boost the allowable dirty threshold a bit for page
···
 	struct request_queue *q = (struct request_queue *)data;
 	int nr_pages = global_page_state(NR_FILE_DIRTY) +
 		global_page_state(NR_UNSTABLE_NFS);
+	struct bdi_writeback *wb;
+	struct wb_iter iter;
 
 	/*
 	 * We want to write everything out, not just down to the dirty
 	 * threshold
 	 */
-	if (bdi_has_dirty_io(&q->backing_dev_info))
-		bdi_start_writeback(&q->backing_dev_info, nr_pages,
-					WB_REASON_LAPTOP_TIMER);
+	if (!bdi_has_dirty_io(&q->backing_dev_info))
+		return;
+
+	bdi_for_each_wb(wb, &q->backing_dev_info, &iter, 0)
+		if (wb_has_dirty_io(wb))
+			wb_start_writeback(wb, nr_pages, true,
+					   WB_REASON_LAPTOP_TIMER);
 }
 
 /*
···
 
 void writeback_set_ratelimit(void)
 {
+	struct wb_domain *dom = &global_wb_domain;
 	unsigned long background_thresh;
 	unsigned long dirty_thresh;
+
 	global_dirty_limits(&background_thresh, &dirty_thresh);
-	global_dirty_limit = dirty_thresh;
+	dom->dirty_limit = dirty_thresh;
 	ratelimit_pages = dirty_thresh / (num_online_cpus() * 32);
 	if (ratelimit_pages < 16)
 		ratelimit_pages = 16;
···
 	writeback_set_ratelimit();
 	register_cpu_notifier(&ratelimit_nb);
 
-	fprop_global_init(&writeout_completions, GFP_KERNEL);
+	BUG_ON(wb_domain_init(&global_wb_domain, GFP_KERNEL));
 }
 
 /**
···
 
 /*
  * Helper function for set_page_dirty family.
+ *
+ * Caller must hold mem_cgroup_begin_page_stat().
+ *
  * NOTE: This relies on being atomic wrt interrupts.
  */
-void account_page_dirtied(struct page *page, struct address_space *mapping)
+void account_page_dirtied(struct page *page, struct address_space *mapping,
+			  struct mem_cgroup *memcg)
 {
+	struct inode *inode = mapping->host;
+
 	trace_writeback_dirty_page(page, mapping);
 
 	if (mapping_cap_account_dirty(mapping)) {
-		struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+		struct bdi_writeback *wb;
 
+		inode_attach_wb(inode, page);
+		wb = inode_to_wb(inode);
+
+		mem_cgroup_inc_page_stat(memcg, MEM_CGROUP_STAT_DIRTY);
 		__inc_zone_page_state(page, NR_FILE_DIRTY);
 		__inc_zone_page_state(page, NR_DIRTIED);
-		__inc_bdi_stat(bdi, BDI_RECLAIMABLE);
-		__inc_bdi_stat(bdi, BDI_DIRTIED);
+		__inc_wb_stat(wb, WB_RECLAIMABLE);
+		__inc_wb_stat(wb, WB_DIRTIED);
 		task_io_account_write(PAGE_CACHE_SIZE);
 		current->nr_dirtied++;
 		this_cpu_inc(bdp_ratelimits);
···
 /*
  * Helper function for deaccounting dirty page without writeback.
  *
- * Doing this should *normally* only ever be done when a page
- * is truncated, and is not actually mapped anywhere at all. However,
- * fs/buffer.c does this when it notices that somebody has cleaned
- * out all the buffers on a page without actually doing it through
- * the VM. Can you say "ext3 is horribly ugly"? Thought you could.
+ * Caller must hold mem_cgroup_begin_page_stat().
  */
-void account_page_cleaned(struct page *page, struct address_space *mapping)
+void account_page_cleaned(struct page *page, struct address_space *mapping,
+			  struct mem_cgroup *memcg, struct bdi_writeback *wb)
 {
 	if (mapping_cap_account_dirty(mapping)) {
+		mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY);
 		dec_zone_page_state(page, NR_FILE_DIRTY);
-		dec_bdi_stat(inode_to_bdi(mapping->host), BDI_RECLAIMABLE);
+		dec_wb_stat(wb, WB_RECLAIMABLE);
 		task_io_account_cancelled_write(PAGE_CACHE_SIZE);
 	}
 }
-EXPORT_SYMBOL(account_page_cleaned);
 
 /*
  * For address_spaces which do not use buffers.  Just tag the page as dirty in
···
  */
 int __set_page_dirty_nobuffers(struct page *page)
 {
+	struct mem_cgroup *memcg;
+
+	memcg = mem_cgroup_begin_page_stat(page);
 	if (!TestSetPageDirty(page)) {
 		struct address_space *mapping = page_mapping(page);
 		unsigned long flags;
 
-		if (!mapping)
+		if (!mapping) {
+			mem_cgroup_end_page_stat(memcg);
 			return 1;
+		}
 
 		spin_lock_irqsave(&mapping->tree_lock, flags);
 		BUG_ON(page_mapping(page) != mapping);
 		WARN_ON_ONCE(!PagePrivate(page) && !PageUptodate(page));
-		account_page_dirtied(page, mapping);
+		account_page_dirtied(page, mapping, memcg);
 		radix_tree_tag_set(&mapping->page_tree, page_index(page),
 				   PAGECACHE_TAG_DIRTY);
 		spin_unlock_irqrestore(&mapping->tree_lock, flags);
+		mem_cgroup_end_page_stat(memcg);
+
 		if (mapping->host) {
 			/* !PageAnon && !swapper_space */
 			__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 		}
 		return 1;
 	}
+	mem_cgroup_end_page_stat(memcg);
 	return 0;
 }
 EXPORT_SYMBOL(__set_page_dirty_nobuffers);
···
 void account_page_redirty(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
+
 	if (mapping && mapping_cap_account_dirty(mapping)) {
+		struct inode *inode = mapping->host;
+		struct bdi_writeback *wb;
+		bool locked;
+
+		wb = unlocked_inode_to_wb_begin(inode, &locked);
 		current->nr_dirtied--;
 		dec_zone_page_state(page, NR_DIRTIED);
-		dec_bdi_stat(inode_to_bdi(mapping->host), BDI_DIRTIED);
+		dec_wb_stat(wb, WB_DIRTIED);
+		unlocked_inode_to_wb_end(inode, locked);
 	}
 }
 EXPORT_SYMBOL(account_page_redirty);
···
 EXPORT_SYMBOL(set_page_dirty_lock);
 
 /*
+ * This cancels just the dirty bit on the kernel page itself, it does NOT
+ * actually remove dirty bits on any mmap's that may be around. It also
+ * leaves the page tagged dirty, so any sync activity will still find it on
+ * the dirty lists, and in particular, clear_page_dirty_for_io() will still
+ * look at the dirty bits in the VM.
+ *
+ * Doing this should *normally* only ever be done when a page is truncated,
+ * and is not actually mapped anywhere at all. However, fs/buffer.c does
+ * this when it notices that somebody has cleaned out all the buffers on a
+ * page without actually doing it through the VM. Can you say "ext3 is
+ * horribly ugly"? Thought you could.
+ */
+void cancel_dirty_page(struct page *page)
+{
+	struct address_space *mapping = page_mapping(page);
+
+	if (mapping_cap_account_dirty(mapping)) {
+		struct inode *inode = mapping->host;
+		struct bdi_writeback *wb;
+		struct mem_cgroup *memcg;
+		bool locked;
+
+		memcg = mem_cgroup_begin_page_stat(page);
+		wb = unlocked_inode_to_wb_begin(inode, &locked);
+
+		if (TestClearPageDirty(page))
+			account_page_cleaned(page, mapping, memcg, wb);
+
+		unlocked_inode_to_wb_end(inode, locked);
+		mem_cgroup_end_page_stat(memcg);
+	} else {
+		ClearPageDirty(page);
+	}
+}
+EXPORT_SYMBOL(cancel_dirty_page);
+
+/*
  * Clear a page's dirty flag, while caring for dirty memory accounting.
  * Returns true if the page was previously dirty.
  *
···
 int clear_page_dirty_for_io(struct page *page)
 {
 	struct address_space *mapping = page_mapping(page);
+	int ret = 0;
 
 	BUG_ON(!PageLocked(page));
 
 	if (mapping && mapping_cap_account_dirty(mapping)) {
+		struct inode *inode = mapping->host;
+		struct bdi_writeback *wb;
+		struct mem_cgroup *memcg;
+		bool locked;
+
 		/*
 		 * Yes, Virginia, this is indeed insane.
 		 *
···
 		 * always locked coming in here, so we get the desired
 		 * exclusion.
 		 */
+		memcg = mem_cgroup_begin_page_stat(page);
+		wb = unlocked_inode_to_wb_begin(inode, &locked);
 		if (TestClearPageDirty(page)) {
+			mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY);
 			dec_zone_page_state(page, NR_FILE_DIRTY);
-			dec_bdi_stat(inode_to_bdi(mapping->host),
-					BDI_RECLAIMABLE);
-			return 1;
+			dec_wb_stat(wb, WB_RECLAIMABLE);
+			ret = 1;
 		}
-		return 0;
+		unlocked_inode_to_wb_end(inode, locked);
+		mem_cgroup_end_page_stat(memcg);
+		return ret;
 	}
 	return TestClearPageDirty(page);
 }
···
 
 	memcg = mem_cgroup_begin_page_stat(page);
 	if (mapping) {
-		struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+		struct inode *inode = mapping->host;
+		struct backing_dev_info *bdi = inode_to_bdi(inode);
 		unsigned long flags;
 
 		spin_lock_irqsave(&mapping->tree_lock, flags);
···
 						page_index(page),
 						PAGECACHE_TAG_WRITEBACK);
 			if (bdi_cap_account_writeback(bdi)) {
-				__dec_bdi_stat(bdi, BDI_WRITEBACK);
-				__bdi_writeout_inc(bdi);
+				struct bdi_writeback *wb = inode_to_wb(inode);
+
+				__dec_wb_stat(wb, WB_WRITEBACK);
+				__wb_writeout_inc(wb);
 			}
 		}
 		spin_unlock_irqrestore(&mapping->tree_lock, flags);
···
 
 	memcg = mem_cgroup_begin_page_stat(page);
 	if (mapping) {
-		struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+		struct inode *inode = mapping->host;
+		struct backing_dev_info *bdi = inode_to_bdi(inode);
 		unsigned long flags;
 
 		spin_lock_irqsave(&mapping->tree_lock, flags);
···
 						page_index(page),
 						PAGECACHE_TAG_WRITEBACK);
 			if (bdi_cap_account_writeback(bdi))
-				__inc_bdi_stat(bdi, BDI_WRITEBACK);
+				__inc_wb_stat(inode_to_wb(inode), WB_WRITEBACK);
 		}
 		if (!PageDirty(page))
 			radix_tree_tag_clear(&mapping->page_tree,

mm/readahead.c

···
 	/*
 	 * Defer asynchronous read-ahead on IO congestion.
 	 */
-	if (bdi_read_congested(inode_to_bdi(mapping->host)))
+	if (inode_read_congested(mapping->host))
 		return;
 
 	/* do read-ahead */

mm/rmap.c

···
  *             swap_lock (in swap_duplicate, swap_info_get)
  *               mmlist_lock (in mmput, drain_mmlist and others)
  *               mapping->private_lock (in __set_page_dirty_buffers)
+ *                 mem_cgroup_{begin,end}_page_stat (memcg->move_lock)
+ *                   mapping->tree_lock (widely used)
  *               inode->i_lock (in set_page_dirty's __mark_inode_dirty)
  *               bdi.wb->list_lock (in set_page_dirty's __mark_inode_dirty)
  *                 sb_lock (within inode_lock in fs/fs-writeback.c)

mm/truncate.c

···
 	 * the VM has canceled the dirty bit (eg ext3 journaling).
 	 * Hence dirty accounting check is placed after invalidation.
 	 */
-	if (TestClearPageDirty(page))
-		account_page_cleaned(page, mapping);
-
+	cancel_dirty_page(page);
 	ClearPageMappedToDisk(page);
 	delete_from_page_cache(page);
 	return 0;
···
 static int
 invalidate_complete_page2(struct address_space *mapping, struct page *page)
 {
+	struct mem_cgroup *memcg;
+	unsigned long flags;
+
 	if (page->mapping != mapping)
 		return 0;
 
 	if (page_has_private(page) && !try_to_release_page(page, GFP_KERNEL))
 		return 0;
 
-	spin_lock_irq(&mapping->tree_lock);
+	memcg = mem_cgroup_begin_page_stat(page);
+	spin_lock_irqsave(&mapping->tree_lock, flags);
 	if (PageDirty(page))
 		goto failed;
 
 	BUG_ON(page_has_private(page));
-	__delete_from_page_cache(page, NULL);
-	spin_unlock_irq(&mapping->tree_lock);
+	__delete_from_page_cache(page, NULL, memcg);
+	spin_unlock_irqrestore(&mapping->tree_lock, flags);
+	mem_cgroup_end_page_stat(memcg);
 
 	if (mapping->a_ops->freepage)
 		mapping->a_ops->freepage(page);
···
 	page_cache_release(page);	/* pagecache ref */
 	return 1;
 failed:
-	spin_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irqrestore(&mapping->tree_lock, flags);
+	mem_cgroup_end_page_stat(memcg);
 	return 0;
 }
 

mm/vmscan.c

···
 {
 	return !sc->target_mem_cgroup;
 }
+
+/**
+ * sane_reclaim - is the usual dirty throttling mechanism operational?
+ * @sc: scan_control in question
+ *
+ * The normal page dirty throttling mechanism in balance_dirty_pages() is
+ * completely broken with the legacy memcg and direct stalling in
+ * shrink_page_list() is used for throttling instead, which lacks all the
+ * niceties such as fairness, adaptive pausing, bandwidth proportional
+ * allocation and configurability.
+ *
+ * This function tests whether the vmscan currently in progress can assume
+ * that the normal dirty throttling mechanism is operational.
+ */
+static bool sane_reclaim(struct scan_control *sc)
+{
+	struct mem_cgroup *memcg = sc->target_mem_cgroup;
+
+	if (!memcg)
+		return true;
+#ifdef CONFIG_CGROUP_WRITEBACK
+	if (cgroup_on_dfl(mem_cgroup_css(memcg)->cgroup))
+		return true;
+#endif
+	return false;
+}
 #else
 static bool global_reclaim(struct scan_control *sc)
+{
+	return true;
+}
+
+static bool sane_reclaim(struct scan_control *sc)
 {
 	return true;
 }
···
 	return page_count(page) - page_has_private(page) == 2;
 }
 
-static int may_write_to_queue(struct backing_dev_info *bdi,
-			      struct scan_control *sc)
+static int may_write_to_inode(struct inode *inode, struct scan_control *sc)
 {
 	if (current->flags & PF_SWAPWRITE)
 		return 1;
-	if (!bdi_write_congested(bdi))
+	if (!inode_write_congested(inode))
 		return 1;
-	if (bdi == current->backing_dev_info)
+	if (inode_to_bdi(inode) == current->backing_dev_info)
 		return 1;
 	return 0;
 }
···
 	}
 	if (mapping->a_ops->writepage == NULL)
 		return PAGE_ACTIVATE;
-	if (!may_write_to_queue(inode_to_bdi(mapping->host), sc))
+	if (!may_write_to_inode(mapping->host, sc))
 		return PAGE_KEEP;
 
 	if (clear_page_dirty_for_io(page)) {
···
 static int __remove_mapping(struct address_space *mapping, struct page *page,
 			    bool reclaimed)
 {
+	unsigned long flags;
+	struct mem_cgroup *memcg;
+
 	BUG_ON(!PageLocked(page));
 	BUG_ON(mapping != page_mapping(page));
 
-	spin_lock_irq(&mapping->tree_lock);
+	memcg = mem_cgroup_begin_page_stat(page);
+	spin_lock_irqsave(&mapping->tree_lock, flags);
 	/*
 	 * The non racy check for a busy page.
 	 *
···
 		swp_entry_t swap = { .val = page_private(page) };
 		mem_cgroup_swapout(page, swap);
 		__delete_from_swap_cache(page);
-		spin_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
+		mem_cgroup_end_page_stat(memcg);
 		swapcache_free(swap);
 	} else {
 		void (*freepage)(struct page *);
···
 		if (reclaimed && page_is_file_cache(page) &&
 		    !mapping_exiting(mapping))
 			shadow = workingset_eviction(mapping, page);
-		__delete_from_page_cache(page, shadow);
-		spin_unlock_irq(&mapping->tree_lock);
+		__delete_from_page_cache(page, shadow, memcg);
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
+		mem_cgroup_end_page_stat(memcg);
 
 		if (freepage != NULL)
 			freepage(page);
···
 	return 1;
 
 cannot_free:
-	spin_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irqrestore(&mapping->tree_lock, flags);
+	mem_cgroup_end_page_stat(memcg);
 	return 0;
 }
 
···
 		 */
 		mapping = page_mapping(page);
 		if (((dirty || writeback) && mapping &&
-		     bdi_write_congested(inode_to_bdi(mapping->host))) ||
+		     inode_write_congested(mapping->host)) ||
 		    (writeback && PageReclaim(page)))
 			nr_congested++;
 
···
 		 *    note that the LRU is being scanned too quickly and the
 		 *    caller can stall after page list has been processed.
 		 *
-		 * 2) Global reclaim encounters a page, memcg encounters a
-		 *    page that is not marked for immediate reclaim or
-		 *    the caller does not have __GFP_IO. In this case mark
-		 *    the page for immediate reclaim and continue scanning.
+		 * 2) Global or new memcg reclaim encounters a page that is
+		 *    not marked for immediate reclaim or the caller does not
+		 *    have __GFP_IO. In this case mark the page for immediate
+		 *    reclaim and continue scanning.
 		 *
 		 *    __GFP_IO is checked  because a loop driver thread might
 		 *    enter reclaim, and deadlock if it waits on a page for
···
 		 *    grab_cache_page_write_begin(,,AOP_FLAG_NOFS), so testing
 		 *    may_enter_fs here is liable to OOM on them.
 		 *
-		 * 3) memcg encounters a page that is not already marked
+		 * 3) Legacy memcg encounters a page that is not already marked
 		 *    PageReclaim. memcg does not have any dirty pages
 		 *    throttling so we could easily OOM just because too many
 		 *    pages are in writeback and there is nothing else to
···
 				goto keep_locked;
 
 			/* Case 2 above */
-			} else if (global_reclaim(sc) ||
+			} else if (sane_reclaim(sc) ||
 			    !PageReclaim(page) || !(sc->gfp_mask & __GFP_IO)) {
 				/*
 				 * This is slightly racy - end_page_writeback()
···
 	if (current_is_kswapd())
 		return 0;
 
-	if (!global_reclaim(sc))
+	if (!sane_reclaim(sc))
 		return 0;
 
 	if (file) {
···
 		set_bit(ZONE_WRITEBACK, &zone->flags);
 
 	/*
-	 * memcg will stall in page writeback so only consider forcibly
-	 * stalling for global reclaim
+	 * Legacy memcg will stall in page writeback so avoid forcibly
+	 * stalling here.
 	 */
-	if (global_reclaim(sc)) {
+	if (sane_reclaim(sc)) {
 		/*
 		 * Tag a zone as congested if all the dirty pages scanned were
 		 * backed by a congested BDI and wait_iff_congested will stall.