1 files changed, 504 insertions, 323 deletions

diff --git a/block/blk-wbt.c b/block/blk-wbt.c
index 6a9a0f03a67b..eb8037bae0bd 100644
--- a/block/blk-wbt.c
+++ b/block/blk-wbt.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
  * buffered writeback throttling. loosely based on CoDel. We can't drop
  * packets for IO scheduling, so the logic is something like this:
@@ -24,11 +25,99 @@
 #include <linux/backing-dev.h>
 #include <linux/swap.h>
 
+#include "blk-stat.h"
 #include "blk-wbt.h"
+#include "blk-rq-qos.h"
+#include "elevator.h"
+#include "blk.h"
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/wbt.h>
 
+enum wbt_flags {
+	WBT_TRACKED		= 1,	/* write, tracked for throttling */
+	WBT_READ		= 2,	/* read */
+	WBT_SWAP		= 4,	/* write, from swap_writeout() */
+	WBT_DISCARD		= 8,	/* discard */
+
+	WBT_NR_BITS		= 4,	/* number of bits */
+};
+
+enum {
+	WBT_RWQ_BG		= 0,
+	WBT_RWQ_SWAP,
+	WBT_RWQ_DISCARD,
+	WBT_NUM_RWQ,
+};
+
+/*
+ * If current state is WBT_STATE_ON/OFF_DEFAULT, it can be covered to any other
+ * state, if current state is WBT_STATE_ON/OFF_MANUAL, it can only be covered
+ * to WBT_STATE_OFF/ON_MANUAL.
+ */
+enum {
+	WBT_STATE_ON_DEFAULT	= 1,	/* on by default */
+	WBT_STATE_ON_MANUAL	= 2,	/* on manually by sysfs */
+	WBT_STATE_OFF_DEFAULT	= 3,	/* off by default */
+	WBT_STATE_OFF_MANUAL	= 4,	/* off manually by sysfs */
+};
+
+struct rq_wb {
+	/*
+	 * Settings that govern how we throttle
+	 */
+	unsigned int wb_background;		/* background writeback */
+	unsigned int wb_normal;			/* normal writeback */
+
+	short enable_state;			/* WBT_STATE_* */
+
+	/*
+	 * Number of consecutive periods where we don't have enough
+	 * information to make a firm scale up/down decision.
+	 */
+	unsigned int unknown_cnt;
+
+	u64 win_nsec;				/* default window size */
+	u64 cur_win_nsec;			/* current window size */
+
+	struct blk_stat_callback *cb;
+
+	u64 sync_issue;
+	void *sync_cookie;
+
+	unsigned long last_issue;	/* issue time of last read rq */
+	unsigned long last_comp;	/* completion time of last read rq */
+	unsigned long min_lat_nsec;
+	struct rq_qos rqos;
+	struct rq_wait rq_wait[WBT_NUM_RWQ];
+	struct rq_depth rq_depth;
+};
+
+static inline struct rq_wb *RQWB(struct rq_qos *rqos)
+{
+	return container_of(rqos, struct rq_wb, rqos);
+}
+
+static inline void wbt_clear_state(struct request *rq)
+{
+	rq->wbt_flags = 0;
+}
+
+static inline enum wbt_flags wbt_flags(struct request *rq)
+{
+	return rq->wbt_flags;
+}
+
+static inline bool wbt_is_tracked(struct request *rq)
+{
+	return rq->wbt_flags & WBT_TRACKED;
+}
+
+static inline bool wbt_is_read(struct request *rq)
+{
+	return rq->wbt_flags & WBT_READ;
+}
+
 enum {
 	/*
 	 * Default setting, we'll scale up (to 75% of QD max) or down (min 1)
@@ -47,37 +136,17 @@ enum {
 	RWB_MIN_WRITE_SAMPLES	= 3,
 
 	/*
-	 * If we have this number of consecutive windows with not enough
-	 * information to scale up or down, scale up.
+	 * If we have this number of consecutive windows without enough
+	 * information to scale up or down, slowly return to center state
+	 * (step == 0).
 	 */
 	RWB_UNKNOWN_BUMP	= 5,
 };
 
 static inline bool rwb_enabled(struct rq_wb *rwb)
 {
-	return rwb && rwb->wb_normal != 0;
-}
-
-/*
- * Increment 'v', if 'v' is below 'below'. Returns true if we succeeded,
- * false if 'v' + 1 would be bigger than 'below'.
- */
-static bool atomic_inc_below(atomic_t *v, int below)
-{
-	int cur = atomic_read(v);
-
-	for (;;) {
-		int old;
-
-		if (cur >= below)
-			return false;
-		old = atomic_cmpxchg(v, cur, cur + 1);
-		if (old == cur)
-			break;
-		cur = old;
-	}
-
-	return true;
+	return rwb && rwb->enable_state != WBT_STATE_OFF_DEFAULT &&
+		      rwb->enable_state != WBT_STATE_OFF_MANUAL;
 }
 
 static void wb_timestamp(struct rq_wb *rwb, unsigned long *var)
@@ -96,14 +165,20 @@ static void wb_timestamp(struct rq_wb *rwb, unsigned long *var)
  */
 static bool wb_recent_wait(struct rq_wb *rwb)
 {
-	struct bdi_writeback *wb = &rwb->queue->backing_dev_info->wb;
+	struct backing_dev_info *bdi = rwb->rqos.disk->bdi;
 
-	return time_before(jiffies, wb->dirty_sleep + HZ);
+	return time_before(jiffies, bdi->last_bdp_sleep + HZ);
 }
 
-static inline struct rq_wait *get_rq_wait(struct rq_wb *rwb, bool is_kswapd)
+static inline struct rq_wait *get_rq_wait(struct rq_wb *rwb,
+					  enum wbt_flags wb_acct)
 {
-	return &rwb->rq_wait[is_kswapd];
+	if (wb_acct & WBT_SWAP)
+		return &rwb->rq_wait[WBT_RWQ_SWAP];
+	else if (wb_acct & WBT_DISCARD)
+		return &rwb->rq_wait[WBT_RWQ_DISCARD];
+
+	return &rwb->rq_wait[WBT_RWQ_BG];
 }
 
 static void rwb_wake_all(struct rq_wb *rwb)
@@ -113,36 +188,27 @@ static void rwb_wake_all(struct rq_wb *rwb)
 	for (i = 0; i < WBT_NUM_RWQ; i++) {
 		struct rq_wait *rqw = &rwb->rq_wait[i];
 
-		if (waitqueue_active(&rqw->wait))
+		if (wq_has_sleeper(&rqw->wait))
 			wake_up_all(&rqw->wait);
 	}
 }
 
-void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct)
+static void wbt_rqw_done(struct rq_wb *rwb, struct rq_wait *rqw,
+			 enum wbt_flags wb_acct)
 {
-	struct rq_wait *rqw;
 	int inflight, limit;
 
-	if (!(wb_acct & WBT_TRACKED))
-		return;
-
-	rqw = get_rq_wait(rwb, wb_acct & WBT_KSWAPD);
 	inflight = atomic_dec_return(&rqw->inflight);
 
 	/*
-	 * wbt got disabled with IO in flight. Wake up any potential
-	 * waiters, we don't have to do more than that.
+	 * For discards, our limit is always the background. For writes, if
+	 * the device does write back caching, drop further down before we
+	 * wake people up.
 	 */
-	if (unlikely(!rwb_enabled(rwb))) {
-		rwb_wake_all(rwb);
-		return;
-	}
-
-	/*
-	 * If the device does write back caching, drop further down
-	 * before we wake people up.
-	 */
-	if (rwb->wc && !wb_recent_wait(rwb))
+	if (wb_acct & WBT_DISCARD)
+		limit = rwb->wb_background;
+	else if (blk_queue_write_cache(rwb->rqos.disk->queue) &&
+		 !wb_recent_wait(rwb))
 		limit = 0;
 	else
 		limit = rwb->wb_normal;
@@ -153,7 +219,7 @@ void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct)
 	if (inflight && inflight >= limit)
 		return;
 
-	if (waitqueue_active(&rqw->wait)) {
+	if (wq_has_sleeper(&rqw->wait)) {
 		int diff = limit - inflight;
 
 		if (!inflight || diff >= rwb->wb_background / 2)
@@ -161,90 +227,40 @@ void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct)
 	}
 }
 
-/*
- * Called on completion of a request. Note that it's also called when
- * a request is merged, when the request gets freed.
- */
-void wbt_done(struct rq_wb *rwb, struct blk_issue_stat *stat)
+static void __wbt_done(struct rq_qos *rqos, enum wbt_flags wb_acct)
 {
-	if (!rwb)
-		return;
+	struct rq_wb *rwb = RQWB(rqos);
+	struct rq_wait *rqw;
 
-	if (!wbt_is_tracked(stat)) {
-		if (rwb->sync_cookie == stat) {
-			rwb->sync_issue = 0;
-			rwb->sync_cookie = NULL;
-		}
+	if (!(wb_acct & WBT_TRACKED))
+		return;
 
-		if (wbt_is_read(stat))
-			wb_timestamp(rwb, &rwb->last_comp);
-		wbt_clear_state(stat);
-	} else {
-		WARN_ON_ONCE(stat == rwb->sync_cookie);
-		__wbt_done(rwb, wbt_stat_to_mask(stat));
-		wbt_clear_state(stat);
-	}
+	rqw = get_rq_wait(rwb, wb_acct);
+	wbt_rqw_done(rwb, rqw, wb_acct);
 }
 
 /*
- * Return true, if we can't increase the depth further by scaling
+ * Called on completion of a request. Note that it's also called when
+ * a request is merged, when the request gets freed.
  */
-static bool calc_wb_limits(struct rq_wb *rwb)
+static void wbt_done(struct rq_qos *rqos, struct request *rq)
 {
-	unsigned int depth;
-	bool ret = false;
+	struct rq_wb *rwb = RQWB(rqos);
 
-	if (!rwb->min_lat_nsec) {
-		rwb->wb_max = rwb->wb_normal = rwb->wb_background = 0;
-		return false;
-	}
+	if (!wbt_is_tracked(rq)) {
+		if (wbt_is_read(rq)) {
+			if (rwb->sync_cookie == rq) {
+				rwb->sync_issue = 0;
+				rwb->sync_cookie = NULL;
+			}
 
-	/*
-	 * For QD=1 devices, this is a special case. It's important for those
-	 * to have one request ready when one completes, so force a depth of
-	 * 2 for those devices. On the backend, it'll be a depth of 1 anyway,
-	 * since the device can't have more than that in flight. If we're
-	 * scaling down, then keep a setting of 1/1/1.
-	 */
-	if (rwb->queue_depth == 1) {
-		if (rwb->scale_step > 0)
-			rwb->wb_max = rwb->wb_normal = 1;
-		else {
-			rwb->wb_max = rwb->wb_normal = 2;
-			ret = true;
+			wb_timestamp(rwb, &rwb->last_comp);
 		}
-		rwb->wb_background = 1;
 	} else {
-		/*
-		 * scale_step == 0 is our default state. If we have suffered
-		 * latency spikes, step will be > 0, and we shrink the
-		 * allowed write depths. If step is < 0, we're only doing
-		 * writes, and we allow a temporarily higher depth to
-		 * increase performance.
-		 */
-		depth = min_t(unsigned int, RWB_DEF_DEPTH, rwb->queue_depth);
-		if (rwb->scale_step > 0)
-			depth = 1 + ((depth - 1) >> min(31, rwb->scale_step));
-		else if (rwb->scale_step < 0) {
-			unsigned int maxd = 3 * rwb->queue_depth / 4;
-
-			depth = 1 + ((depth - 1) << -rwb->scale_step);
-			if (depth > maxd) {
-				depth = maxd;
-				ret = true;
-			}
-		}
-
-		/*
-		 * Set our max/normal/bg queue depths based on how far
-		 * we have scaled down (->scale_step).
-		 */
-		rwb->wb_max = depth;
-		rwb->wb_normal = (rwb->wb_max + 1) / 2;
-		rwb->wb_background = (rwb->wb_max + 3) / 4;
+		WARN_ON_ONCE(rq == rwb->sync_cookie);
+		__wbt_done(rqos, wbt_flags(rq));
 	}
-
-	return ret;
+	wbt_clear_state(rq);
 }
 
 static inline bool stat_sample_valid(struct blk_rq_stat *stat)
@@ -261,13 +277,22 @@ static inline bool stat_sample_valid(struct blk_rq_stat *stat)
 
 static u64 rwb_sync_issue_lat(struct rq_wb *rwb)
 {
-	u64 now, issue = ACCESS_ONCE(rwb->sync_issue);
+	u64 issue = READ_ONCE(rwb->sync_issue);
 
 	if (!issue || !rwb->sync_cookie)
 		return 0;
 
-	now = ktime_to_ns(ktime_get());
-	return now - issue;
+	return blk_time_get_ns() - issue;
+}
+
+static inline unsigned int wbt_inflight(struct rq_wb *rwb)
+{
+	unsigned int i, ret = 0;
+
+	for (i = 0; i < WBT_NUM_RWQ; i++)
+		ret += atomic_read(&rwb->rq_wait[i].inflight);
+
+	return ret;
 }
 
 enum {
@@ -279,7 +304,8 @@ enum {
 
 static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat)
 {
-	struct backing_dev_info *bdi = rwb->queue->backing_dev_info;
+	struct backing_dev_info *bdi = rwb->rqos.disk->bdi;
+	struct rq_depth *rqd = &rwb->rq_depth;
 	u64 thislat;
 
 	/*
@@ -323,7 +349,7 @@ static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat)
 		return LAT_EXCEEDED;
 	}
 
-	if (rwb->scale_step)
+	if (rqd->scale_step)
 		trace_wbt_stat(bdi, stat);
 
 	return LAT_OK;
@@ -331,58 +357,50 @@ static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat)
 
 static void rwb_trace_step(struct rq_wb *rwb, const char *msg)
 {
-	struct backing_dev_info *bdi = rwb->queue->backing_dev_info;
+	struct backing_dev_info *bdi = rwb->rqos.disk->bdi;
+	struct rq_depth *rqd = &rwb->rq_depth;
 
-	trace_wbt_step(bdi, msg, rwb->scale_step, rwb->cur_win_nsec,
-			rwb->wb_background, rwb->wb_normal, rwb->wb_max);
+	trace_wbt_step(bdi, msg, rqd->scale_step, rwb->cur_win_nsec,
+			rwb->wb_background, rwb->wb_normal, rqd->max_depth);
+}
+
+static void calc_wb_limits(struct rq_wb *rwb)
+{
+	if (rwb->min_lat_nsec == 0) {
+		rwb->wb_normal = rwb->wb_background = 0;
+	} else if (rwb->rq_depth.max_depth <= 2) {
+		rwb->wb_normal = rwb->rq_depth.max_depth;
+		rwb->wb_background = 1;
+	} else {
+		rwb->wb_normal = (rwb->rq_depth.max_depth + 1) / 2;
+		rwb->wb_background = (rwb->rq_depth.max_depth + 3) / 4;
+	}
 }
 
 static void scale_up(struct rq_wb *rwb)
 {
-	/*
-	 * Hit max in previous round, stop here
-	 */
-	if (rwb->scaled_max)
+	if (!rq_depth_scale_up(&rwb->rq_depth))
 		return;
-
-	rwb->scale_step--;
+	calc_wb_limits(rwb);
 	rwb->unknown_cnt = 0;
-
-	rwb->scaled_max = calc_wb_limits(rwb);
-
 	rwb_wake_all(rwb);
-
-	rwb_trace_step(rwb, "step up");
+	rwb_trace_step(rwb, tracepoint_string("scale up"));
 }
 
-/*
- * Scale rwb down. If 'hard_throttle' is set, do it quicker, since we
- * had a latency violation.
- */
 static void scale_down(struct rq_wb *rwb, bool hard_throttle)
 {
-	/*
-	 * Stop scaling down when we've hit the limit. This also prevents
-	 * ->scale_step from going to crazy values, if the device can't
-	 * keep up.
-	 */
-	if (rwb->wb_max == 1)
+	if (!rq_depth_scale_down(&rwb->rq_depth, hard_throttle))
 		return;
-
-	if (rwb->scale_step < 0 && hard_throttle)
-		rwb->scale_step = 0;
-	else
-		rwb->scale_step++;
-
-	rwb->scaled_max = false;
-	rwb->unknown_cnt = 0;
 	calc_wb_limits(rwb);
-	rwb_trace_step(rwb, "step down");
+	rwb->unknown_cnt = 0;
+	rwb_trace_step(rwb, tracepoint_string("scale down"));
 }
 
 static void rwb_arm_timer(struct rq_wb *rwb)
 {
-	if (rwb->scale_step > 0) {
+	struct rq_depth *rqd = &rwb->rq_depth;
+
+	if (rqd->scale_step > 0) {
 		/*
 		 * We should speed this up, using some variant of a fast
 		 * integer inverse square root calculation. Since we only do
@@ -390,7 +408,7 @@ static void rwb_arm_timer(struct rq_wb *rwb)
 		 * though.
 		 */
 		rwb->cur_win_nsec = div_u64(rwb->win_nsec << 4,
-					int_sqrt((rwb->scale_step + 1) << 8));
+					int_sqrt((rqd->scale_step + 1) << 8));
 	} else {
 		/*
 		 * For step < 0, we don't want to increase/decrease the
@@ -405,13 +423,16 @@ static void rwb_arm_timer(struct rq_wb *rwb)
 static void wb_timer_fn(struct blk_stat_callback *cb)
 {
 	struct rq_wb *rwb = cb->data;
+	struct rq_depth *rqd = &rwb->rq_depth;
 	unsigned int inflight = wbt_inflight(rwb);
 	int status;
 
+	if (!rwb->rqos.disk)
+		return;
+
 	status = latency_exceeded(rwb, cb->stat);
 
-	trace_wbt_timer(rwb->queue->backing_dev_info, status, rwb->scale_step,
-			inflight);
+	trace_wbt_timer(rwb->rqos.disk->bdi, status, rqd->scale_step, inflight);
 
 	/*
 	 * If we exceeded the latency target, step down. If we did not,
@@ -427,9 +448,9 @@ static void wb_timer_fn(struct blk_stat_callback *cb)
 		break;
 	case LAT_UNKNOWN_WRITES:
 		/*
-		 * We started a the center step, but don't have a valid
-		 * read/write sample, but we do have writes going on.
-		 * Allow step to go negative, to increase write perf.
+		 * We don't have a valid read/write sample, but we do have
+		 * writes going on. Allow step to go negative, to increase
+		 * write performance.
 		 */
 		scale_up(rwb);
 		break;
@@ -441,9 +462,9 @@ static void wb_timer_fn(struct blk_stat_callback *cb)
 		 * currently don't have a valid read/write sample. For that
 		 * case, slowly return to center state (step == 0).
 		 */
-		if (rwb->scale_step > 0)
+		if (rqd->scale_step > 0)
 			scale_up(rwb);
-		else if (rwb->scale_step < 0)
+		else if (rqd->scale_step < 0)
 			scale_down(rwb, false);
 		break;
 	default:
@@ -453,19 +474,54 @@ static void wb_timer_fn(struct blk_stat_callback *cb)
 	/*
 	 * Re-arm timer, if we have IO in flight
 	 */
-	if (rwb->scale_step || inflight)
+	if (rqd->scale_step || inflight)
 		rwb_arm_timer(rwb);
 }
 
-void wbt_update_limits(struct rq_wb *rwb)
+static void wbt_update_limits(struct rq_wb *rwb)
 {
-	rwb->scale_step = 0;
-	rwb->scaled_max = false;
+	struct rq_depth *rqd = &rwb->rq_depth;
+
+	rqd->scale_step = 0;
+	rqd->scaled_max = false;
+
+	rq_depth_calc_max_depth(rqd);
 	calc_wb_limits(rwb);
 
 	rwb_wake_all(rwb);
 }
 
+bool wbt_disabled(struct request_queue *q)
+{
+	struct rq_qos *rqos = wbt_rq_qos(q);
+
+	return !rqos || !rwb_enabled(RQWB(rqos));
+}
+
+u64 wbt_get_min_lat(struct request_queue *q)
+{
+	struct rq_qos *rqos = wbt_rq_qos(q);
+	if (!rqos)
+		return 0;
+	return RQWB(rqos)->min_lat_nsec;
+}
+
+void wbt_set_min_lat(struct request_queue *q, u64 val)
+{
+	struct rq_qos *rqos = wbt_rq_qos(q);
+	if (!rqos)
+		return;
+
+	RQWB(rqos)->min_lat_nsec = val;
+	if (val)
+		RQWB(rqos)->enable_state = WBT_STATE_ON_MANUAL;
+	else
+		RQWB(rqos)->enable_state = WBT_STATE_OFF_MANUAL;
+
+	wbt_update_limits(RQWB(rqos));
+}
+
+
 static bool close_io(struct rq_wb *rwb)
 {
 	const unsigned long now = jiffies;
@@ -474,23 +530,26 @@ static bool close_io(struct rq_wb *rwb)
 		time_before(now, rwb->last_comp + HZ / 10);
 }
 
-#define REQ_HIPRIO	(REQ_SYNC | REQ_META | REQ_PRIO)
+#define REQ_HIPRIO	(REQ_SYNC | REQ_META | REQ_PRIO | REQ_SWAP)
 
-static inline unsigned int get_limit(struct rq_wb *rwb, unsigned long rw)
+static inline unsigned int get_limit(struct rq_wb *rwb, blk_opf_t opf)
 {
 	unsigned int limit;
 
+	if ((opf & REQ_OP_MASK) == REQ_OP_DISCARD)
+		return rwb->wb_background;
+
 	/*
 	 * At this point we know it's a buffered write. If this is
-	 * kswapd trying to free memory, or REQ_SYNC is set, set, then
+	 * swap trying to free memory, or REQ_SYNC is set, then
 	 * it's WB_SYNC_ALL writeback, and we'll use the max limit for
 	 * that. If the write is marked as a background write, then use
 	 * the idle limit, or go to normal if we haven't had competing
 	 * IO for a bit.
 	 */
-	if ((rw & REQ_HIPRIO) || wb_recent_wait(rwb) || current_is_kswapd())
-		limit = rwb->wb_max;
-	else if ((rw & REQ_BACKGROUND) || close_io(rwb)) {
+	if ((opf & REQ_HIPRIO) || wb_recent_wait(rwb))
+		limit = rwb->rq_depth.max_depth;
+	else if ((opf & REQ_BACKGROUND) || close_io(rwb)) {
 		/*
 		 * If less than 100ms since we completed unrelated IO,
 		 * limit us to half the depth for background writeback.
@@ -502,185 +561,171 @@ static inline unsigned int get_limit(struct rq_wb *rwb, unsigned long rw)
 	return limit;
 }
 
-static inline bool may_queue(struct rq_wb *rwb, struct rq_wait *rqw,
-			     wait_queue_entry_t *wait, unsigned long rw)
-{
-	/*
-	 * inc it here even if disabled, since we'll dec it at completion.
-	 * this only happens if the task was sleeping in __wbt_wait(),
-	 * and someone turned it off at the same time.
-	 */
-	if (!rwb_enabled(rwb)) {
-		atomic_inc(&rqw->inflight);
-		return true;
-	}
+struct wbt_wait_data {
+	struct rq_wb *rwb;
+	enum wbt_flags wb_acct;
+	blk_opf_t opf;
+};
 
-	/*
-	 * If the waitqueue is already active and we are not the next
-	 * in line to be woken up, wait for our turn.
-	 */
-	if (waitqueue_active(&rqw->wait) &&
-	    rqw->wait.head.next != &wait->entry)
-		return false;
+static bool wbt_inflight_cb(struct rq_wait *rqw, void *private_data)
+{
+	struct wbt_wait_data *data = private_data;
+	return rq_wait_inc_below(rqw, get_limit(data->rwb, data->opf));
+}
 
-	return atomic_inc_below(&rqw->inflight, get_limit(rwb, rw));
+static void wbt_cleanup_cb(struct rq_wait *rqw, void *private_data)
+{
+	struct wbt_wait_data *data = private_data;
+	wbt_rqw_done(data->rwb, rqw, data->wb_acct);
 }
 
 /*
  * Block if we will exceed our limit, or if we are currently waiting for
  * the timer to kick off queuing again.
  */
-static void __wbt_wait(struct rq_wb *rwb, unsigned long rw, spinlock_t *lock)
-	__releases(lock)
-	__acquires(lock)
+static void __wbt_wait(struct rq_wb *rwb, enum wbt_flags wb_acct,
+		       blk_opf_t opf)
 {
-	struct rq_wait *rqw = get_rq_wait(rwb, current_is_kswapd());
-	DEFINE_WAIT(wait);
-
-	if (may_queue(rwb, rqw, &wait, rw))
-		return;
-
-	do {
-		prepare_to_wait_exclusive(&rqw->wait, &wait,
-						TASK_UNINTERRUPTIBLE);
+	struct rq_wait *rqw = get_rq_wait(rwb, wb_acct);
+	struct wbt_wait_data data = {
+		.rwb = rwb,
+		.wb_acct = wb_acct,
+		.opf = opf,
+	};
 
-		if (may_queue(rwb, rqw, &wait, rw))
-			break;
-
-		if (lock) {
-			spin_unlock_irq(lock);
-			io_schedule();
-			spin_lock_irq(lock);
-		} else
-			io_schedule();
-	} while (1);
-
-	finish_wait(&rqw->wait, &wait);
+	rq_qos_wait(rqw, &data, wbt_inflight_cb, wbt_cleanup_cb);
 }
 
-static inline bool wbt_should_throttle(struct rq_wb *rwb, struct bio *bio)
+static inline bool wbt_should_throttle(struct bio *bio)
 {
-	const int op = bio_op(bio);
-
-	/*
-	 * If not a WRITE, do nothing
-	 */
-	if (op != REQ_OP_WRITE)
-		return false;
-
-	/*
-	 * Don't throttle WRITE_ODIRECT
-	 */
-	if ((bio->bi_opf & (REQ_SYNC | REQ_IDLE)) == (REQ_SYNC | REQ_IDLE))
+	switch (bio_op(bio)) {
+	case REQ_OP_WRITE:
+		/*
+		 * Don't throttle WRITE_ODIRECT
+		 */
+		if ((bio->bi_opf & (REQ_SYNC | REQ_IDLE)) ==
+		    (REQ_SYNC | REQ_IDLE))
+			return false;
+		fallthrough;
+	case REQ_OP_DISCARD:
+		return true;
+	default:
 		return false;
-
-	return true;
+	}
 }
 
-/*
- * Returns true if the IO request should be accounted, false if not.
- * May sleep, if we have exceeded the writeback limits. Caller can pass
- * in an irq held spinlock, if it holds one when calling this function.
- * If we do sleep, we'll release and re-grab it.
- */
-enum wbt_flags wbt_wait(struct rq_wb *rwb, struct bio *bio, spinlock_t *lock)
+static enum wbt_flags bio_to_wbt_flags(struct rq_wb *rwb, struct bio *bio)
 {
-	unsigned int ret = 0;
+	enum wbt_flags flags = 0;
 
 	if (!rwb_enabled(rwb))
 		return 0;
 
-	if (bio_op(bio) == REQ_OP_READ)
-		ret = WBT_READ;
+	if (bio_op(bio) == REQ_OP_READ) {
+		flags = WBT_READ;
+	} else if (wbt_should_throttle(bio)) {
+		if (bio->bi_opf & REQ_SWAP)
+			flags |= WBT_SWAP;
+		if (bio_op(bio) == REQ_OP_DISCARD)
+			flags |= WBT_DISCARD;
+		flags |= WBT_TRACKED;
+	}
+	return flags;
+}
+
+static void wbt_cleanup(struct rq_qos *rqos, struct bio *bio)
+{
+	struct rq_wb *rwb = RQWB(rqos);
+	enum wbt_flags flags = bio_to_wbt_flags(rwb, bio);
+	__wbt_done(rqos, flags);
+}
+
+/* May sleep, if we have exceeded the writeback limits. */
+static void wbt_wait(struct rq_qos *rqos, struct bio *bio)
+{
+	struct rq_wb *rwb = RQWB(rqos);
+	enum wbt_flags flags;
 
-	if (!wbt_should_throttle(rwb, bio)) {
-		if (ret & WBT_READ)
+	flags = bio_to_wbt_flags(rwb, bio);
+	if (!(flags & WBT_TRACKED)) {
+		if (flags & WBT_READ)
 			wb_timestamp(rwb, &rwb->last_issue);
-		return ret;
+		return;
 	}
 
-	__wbt_wait(rwb, bio->bi_opf, lock);
+	__wbt_wait(rwb, flags, bio->bi_opf);
 
 	if (!blk_stat_is_active(rwb->cb))
 		rwb_arm_timer(rwb);
+}
 
-	if (current_is_kswapd())
-		ret |= WBT_KSWAPD;
-
-	return ret | WBT_TRACKED;
+static void wbt_track(struct rq_qos *rqos, struct request *rq, struct bio *bio)
+{
+	struct rq_wb *rwb = RQWB(rqos);
+	rq->wbt_flags |= bio_to_wbt_flags(rwb, bio);
 }
 
-void wbt_issue(struct rq_wb *rwb, struct blk_issue_stat *stat)
+static void wbt_issue(struct rq_qos *rqos, struct request *rq)
 {
+	struct rq_wb *rwb = RQWB(rqos);
+
 	if (!rwb_enabled(rwb))
 		return;
 
 	/*
-	 * Track sync issue, in case it takes a long time to complete. Allows
-	 * us to react quicker, if a sync IO takes a long time to complete.
-	 * Note that this is just a hint. 'stat' can go away when the
-	 * request completes, so it's important we never dereference it. We
-	 * only use the address to compare with, which is why we store the
-	 * sync_issue time locally.
+	 * Track sync issue, in case it takes a long time to complete. Allows us
+	 * to react quicker, if a sync IO takes a long time to complete. Note
+	 * that this is just a hint. The request can go away when it completes,
+	 * so it's important we never dereference it. We only use the address to
+	 * compare with, which is why we store the sync_issue time locally.
 	 */
-	if (wbt_is_read(stat) && !rwb->sync_issue) {
-		rwb->sync_cookie = stat;
-		rwb->sync_issue = blk_stat_time(stat);
+	if (wbt_is_read(rq) && !rwb->sync_issue) {
+		rwb->sync_cookie = rq;
+		rwb->sync_issue = rq->io_start_time_ns;
 	}
 }
 
-void wbt_requeue(struct rq_wb *rwb, struct blk_issue_stat *stat)
+static void wbt_requeue(struct rq_qos *rqos, struct request *rq)
 {
+	struct rq_wb *rwb = RQWB(rqos);
 	if (!rwb_enabled(rwb))
 		return;
-	if (stat == rwb->sync_cookie) {
+	if (rq == rwb->sync_cookie) {
 		rwb->sync_issue = 0;
 		rwb->sync_cookie = NULL;
 	}
 }
 
-void wbt_set_queue_depth(struct rq_wb *rwb, unsigned int depth)
-{
-	if (rwb) {
-		rwb->queue_depth = depth;
-		wbt_update_limits(rwb);
-	}
-}
-
-void wbt_set_write_cache(struct rq_wb *rwb, bool write_cache_on)
-{
-	if (rwb)
-		rwb->wc = write_cache_on;
-}
-
 /*
- * Disable wbt, if enabled by default. Only called from CFQ.
+ * Enable wbt if defaults are configured that way
  */
-void wbt_disable_default(struct request_queue *q)
+void wbt_enable_default(struct gendisk *disk)
 {
-	struct rq_wb *rwb = q->rq_wb;
+	struct request_queue *q = disk->queue;
+	struct rq_qos *rqos;
+	bool enable = IS_ENABLED(CONFIG_BLK_WBT_MQ);
 
-	if (rwb && rwb->enable_state == WBT_STATE_ON_DEFAULT)
-		wbt_exit(q);
-}
-EXPORT_SYMBOL_GPL(wbt_disable_default);
+	mutex_lock(&disk->rqos_state_mutex);
+
+	if (blk_queue_disable_wbt(q))
+		enable = false;
 
-/*
- * Enable wbt if defaults are configured that way
- */
-void wbt_enable_default(struct request_queue *q)
-{
 	/* Throttling already enabled? */
-	if (q->rq_wb)
+	rqos = wbt_rq_qos(q);
+	if (rqos) {
+		if (enable && RQWB(rqos)->enable_state == WBT_STATE_OFF_DEFAULT)
+			RQWB(rqos)->enable_state = WBT_STATE_ON_DEFAULT;
+		mutex_unlock(&disk->rqos_state_mutex);
 		return;
+	}
+	mutex_unlock(&disk->rqos_state_mutex);
 
 	/* Queue not registered? Maybe shutting down... */
-	if (!test_bit(QUEUE_FLAG_REGISTERED, &q->queue_flags))
+	if (!blk_queue_registered(q))
 		return;
 
-	if ((q->mq_ops && IS_ENABLED(CONFIG_BLK_WBT_MQ)) ||
-	    (q->request_fn && IS_ENABLED(CONFIG_BLK_WBT_SQ)))
-		wbt_init(q);
+	if (queue_is_mq(q) && enable)
+		wbt_init(disk);
 }
 EXPORT_SYMBOL_GPL(wbt_enable_default);
 
@@ -698,15 +743,159 @@ u64 wbt_default_latency_nsec(struct request_queue *q)
 
 static int wbt_data_dir(const struct request *rq)
 {
-	return rq_data_dir(rq);
+	const enum req_op op = req_op(rq);
+
+	if (op == REQ_OP_READ)
+		return READ;
+	else if (op_is_write(op))
+		return WRITE;
+
+	/* don't account */
+	return -1;
+}
+
+static void wbt_queue_depth_changed(struct rq_qos *rqos)
+{
+	RQWB(rqos)->rq_depth.queue_depth = blk_queue_depth(rqos->disk->queue);
+	wbt_update_limits(RQWB(rqos));
 }
 
-int wbt_init(struct request_queue *q)
+static void wbt_exit(struct rq_qos *rqos)
 {
+	struct rq_wb *rwb = RQWB(rqos);
+
+	blk_stat_remove_callback(rqos->disk->queue, rwb->cb);
+	blk_stat_free_callback(rwb->cb);
+	kfree(rwb);
+}
+
+/*
+ * Disable wbt, if enabled by default.
+ */
+void wbt_disable_default(struct gendisk *disk)
+{
+	struct rq_qos *rqos = wbt_rq_qos(disk->queue);
 	struct rq_wb *rwb;
+	if (!rqos)
+		return;
+	mutex_lock(&disk->rqos_state_mutex);
+	rwb = RQWB(rqos);
+	if (rwb->enable_state == WBT_STATE_ON_DEFAULT) {
+		blk_stat_deactivate(rwb->cb);
+		rwb->enable_state = WBT_STATE_OFF_DEFAULT;
+	}
+	mutex_unlock(&disk->rqos_state_mutex);
+}
+EXPORT_SYMBOL_GPL(wbt_disable_default);
+
+#ifdef CONFIG_BLK_DEBUG_FS
+static int wbt_curr_win_nsec_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
+
+	seq_printf(m, "%llu\n", rwb->cur_win_nsec);
+	return 0;
+}
+
+static int wbt_enabled_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
+
+	seq_printf(m, "%d\n", rwb->enable_state);
+	return 0;
+}
+
+static int wbt_id_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+
+	seq_printf(m, "%u\n", rqos->id);
+	return 0;
+}
+
+static int wbt_inflight_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
 	int i;
 
-	BUILD_BUG_ON(WBT_NR_BITS > BLK_STAT_RES_BITS);
+	for (i = 0; i < WBT_NUM_RWQ; i++)
+		seq_printf(m, "%d: inflight %d\n", i,
+			   atomic_read(&rwb->rq_wait[i].inflight));
+	return 0;
+}
+
+static int wbt_min_lat_nsec_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
+
+	seq_printf(m, "%lu\n", rwb->min_lat_nsec);
+	return 0;
+}
+
+static int wbt_unknown_cnt_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
+
+	seq_printf(m, "%u\n", rwb->unknown_cnt);
+	return 0;
+}
+
+static int wbt_normal_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
+
+	seq_printf(m, "%u\n", rwb->wb_normal);
+	return 0;
+}
+
+static int wbt_background_show(void *data, struct seq_file *m)
+{
+	struct rq_qos *rqos = data;
+	struct rq_wb *rwb = RQWB(rqos);
+
+	seq_printf(m, "%u\n", rwb->wb_background);
+	return 0;
+}
+
+static const struct blk_mq_debugfs_attr wbt_debugfs_attrs[] = {
+	{"curr_win_nsec", 0400, wbt_curr_win_nsec_show},
+	{"enabled", 0400, wbt_enabled_show},
+	{"id", 0400, wbt_id_show},
+	{"inflight", 0400, wbt_inflight_show},
+	{"min_lat_nsec", 0400, wbt_min_lat_nsec_show},
+	{"unknown_cnt", 0400, wbt_unknown_cnt_show},
+	{"wb_normal", 0400, wbt_normal_show},
+	{"wb_background", 0400, wbt_background_show},
+	{},
+};
+#endif
+
+static const struct rq_qos_ops wbt_rqos_ops = {
+	.throttle = wbt_wait,
+	.issue = wbt_issue,
+	.track = wbt_track,
+	.requeue = wbt_requeue,
+	.done = wbt_done,
+	.cleanup = wbt_cleanup,
+	.queue_depth_changed = wbt_queue_depth_changed,
+	.exit = wbt_exit,
+#ifdef CONFIG_BLK_DEBUG_FS
+	.debugfs_attrs = wbt_debugfs_attrs,
+#endif
+};
+
+int wbt_init(struct gendisk *disk)
+{
+	struct request_queue *q = disk->queue;
+	struct rq_wb *rwb;
+	int i;
+	int ret;
 
 	rwb = kzalloc(sizeof(*rwb), GFP_KERNEL);
 	if (!rwb)
@@ -718,41 +907,33 @@ int wbt_init(struct request_queue *q)
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < WBT_NUM_RWQ; i++) {
-		atomic_set(&rwb->rq_wait[i].inflight, 0);
-		init_waitqueue_head(&rwb->rq_wait[i].wait);
-	}
+	for (i = 0; i < WBT_NUM_RWQ; i++)
+		rq_wait_init(&rwb->rq_wait[i]);
 
-	rwb->wc = 1;
-	rwb->queue_depth = RWB_DEF_DEPTH;
 	rwb->last_comp = rwb->last_issue = jiffies;
-	rwb->queue = q;
 	rwb->win_nsec = RWB_WINDOW_NSEC;
 	rwb->enable_state = WBT_STATE_ON_DEFAULT;
+	rwb->rq_depth.default_depth = RWB_DEF_DEPTH;
+	rwb->min_lat_nsec = wbt_default_latency_nsec(q);
+	rwb->rq_depth.queue_depth = blk_queue_depth(q);
 	wbt_update_limits(rwb);
 
 	/*
 	 * Assign rwb and add the stats callback.
 	 */
-	q->rq_wb = rwb;
-	blk_stat_add_callback(q, rwb->cb);
+	mutex_lock(&q->rq_qos_mutex);
+	ret = rq_qos_add(&rwb->rqos, disk, RQ_QOS_WBT, &wbt_rqos_ops);
+	mutex_unlock(&q->rq_qos_mutex);
+	if (ret)
+		goto err_free;
 
-	rwb->min_lat_nsec = wbt_default_latency_nsec(q);
-
-	wbt_set_queue_depth(rwb, blk_queue_depth(q));
-	wbt_set_write_cache(rwb, test_bit(QUEUE_FLAG_WC, &q->queue_flags));
+	blk_stat_add_callback(q, rwb->cb);
 
 	return 0;
-}
 
-void wbt_exit(struct request_queue *q)
-{
-	struct rq_wb *rwb = q->rq_wb;
+err_free:
+	blk_stat_free_callback(rwb->cb);
+	kfree(rwb);
+	return ret;
 
-	if (rwb) {
-		blk_stat_remove_callback(q, rwb->cb);
-		blk_stat_free_callback(rwb->cb);
-		q->rq_wb = NULL;
-		kfree(rwb);
-	}
 }