@@ -5,7 +5,7 @@
obj-$(CONFIG_BLOCK) := bio.o elevator.o blk-core.o blk-tag.o blk-sysfs.o \
blk-flush.o blk-settings.o blk-ioc.o blk-map.o \
blk-exec.o blk-merge.o blk-softirq.o blk-timeout.o \
- blk-lib.o blk-mq.o blk-mq-tag.o \
+ blk-lib.o blk-mq.o blk-mq-tag.o blk-wb.o \
blk-mq-sysfs.o blk-mq-cpu.o blk-mq-cpumap.o ioctl.o \
genhd.o scsi_ioctl.o partition-generic.o ioprio.o \
badblocks.o partitions/
@@ -39,6 +39,7 @@
#include "blk.h"
#include "blk-mq.h"
+#include "blk-wb.h"
EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap);
EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap);
@@ -863,6 +864,9 @@ blk_init_allocated_queue(struct request_queue *q, request_fn_proc *rfn,
*/
blk_queue_make_request(q, blk_queue_bio);
+ if (blk_wb_init(q))
+ goto fail;
+
q->sg_reserved_size = INT_MAX;
/* Protect q->elevator from elevator_change */
@@ -880,6 +884,7 @@ blk_init_allocated_queue(struct request_queue *q, request_fn_proc *rfn,
fail:
blk_free_flush_queue(q->fq);
+ blk_wb_exit(q);
return NULL;
}
EXPORT_SYMBOL(blk_init_allocated_queue);
@@ -1485,6 +1490,8 @@ void __blk_put_request(struct request_queue *q, struct request *req)
/* this is a bio leak */
WARN_ON(req->bio != NULL);
+ blk_wb_done(q->rq_wb, req);
+
/*
* Request may not have originated from ll_rw_blk. if not,
* it didn't come out of our reserved rq pools
@@ -1714,6 +1721,7 @@ static blk_qc_t blk_queue_bio(struct request_queue *q, struct bio *bio)
int el_ret, rw_flags, where = ELEVATOR_INSERT_SORT;
struct request *req;
unsigned int request_count = 0;
+ bool wb_acct;
/*
* low level driver can indicate that it wants pages above a
@@ -1766,6 +1774,8 @@ static blk_qc_t blk_queue_bio(struct request_queue *q, struct bio *bio)
}
get_rq:
+ wb_acct = blk_wb_wait(q->rq_wb, bio, q->queue_lock);
+
/*
* This sync check and mask will be re-done in init_request_from_bio(),
* but we need to set it earlier to expose the sync flag to the
@@ -1781,11 +1791,16 @@ get_rq:
*/
req = get_request(q, rw_flags, bio, GFP_NOIO);
if (IS_ERR(req)) {
+ if (wb_acct)
+ __blk_wb_done(q->rq_wb);
bio->bi_error = PTR_ERR(req);
bio_endio(bio);
goto out_unlock;
}
+ if (wb_acct)
+ req->cmd_flags |= REQ_BUF_INFLIGHT;
+
/*
* After dropping the lock and possibly sleeping here, our request
* may now be mergeable after it had proven unmergeable (above).
@@ -29,6 +29,7 @@
#include "blk.h"
#include "blk-mq.h"
#include "blk-mq-tag.h"
+#include "blk-wb.h"
static DEFINE_MUTEX(all_q_mutex);
static LIST_HEAD(all_q_list);
@@ -274,6 +275,9 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx,
if (rq->cmd_flags & REQ_MQ_INFLIGHT)
atomic_dec(&hctx->nr_active);
+
+ blk_wb_done(q->rq_wb, rq);
+
rq->cmd_flags = 0;
clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
@@ -1253,6 +1257,7 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
struct blk_plug *plug;
struct request *same_queue_rq = NULL;
blk_qc_t cookie;
+ bool wb_acct;
blk_queue_bounce(q, &bio);
@@ -1270,9 +1275,17 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
} else
request_count = blk_plug_queued_count(q);
+ wb_acct = blk_wb_wait(q->rq_wb, bio, NULL);
+
rq = blk_mq_map_request(q, bio, &data);
- if (unlikely(!rq))
+ if (unlikely(!rq)) {
+ if (wb_acct)
+ __blk_wb_done(q->rq_wb);
return BLK_QC_T_NONE;
+ }
+
+ if (wb_acct)
+ rq->cmd_flags |= REQ_BUF_INFLIGHT;
cookie = blk_tag_to_qc_t(rq->tag, data.hctx->queue_num);
@@ -1349,6 +1362,7 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio)
struct blk_map_ctx data;
struct request *rq;
blk_qc_t cookie;
+ bool wb_acct;
blk_queue_bounce(q, &bio);
@@ -1363,9 +1377,17 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio)
blk_attempt_plug_merge(q, bio, &request_count, NULL))
return BLK_QC_T_NONE;
+ wb_acct = blk_wb_wait(q->rq_wb, bio, NULL);
+
rq = blk_mq_map_request(q, bio, &data);
- if (unlikely(!rq))
+ if (unlikely(!rq)) {
+ if (wb_acct)
+ __blk_wb_done(q->rq_wb);
return BLK_QC_T_NONE;
+ }
+
+ if (wb_acct)
+ rq->cmd_flags |= REQ_BUF_INFLIGHT;
cookie = blk_tag_to_qc_t(rq->tag, data.hctx->queue_num);
@@ -2062,6 +2084,9 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
*/
q->nr_requests = set->queue_depth;
+ if (blk_wb_init(q))
+ goto err_hctxs;
+
if (set->ops->complete)
blk_queue_softirq_done(q, set->ops->complete);
@@ -2097,6 +2122,8 @@ void blk_mq_free_queue(struct request_queue *q)
list_del_init(&q->all_q_node);
mutex_unlock(&all_q_mutex);
+ blk_wb_exit(q);
+
blk_mq_del_queue_tag_set(q);
blk_mq_exit_hw_queues(q, set, set->nr_hw_queues);
@@ -13,6 +13,7 @@
#include <linux/gfp.h>
#include "blk.h"
+#include "blk-wb.h"
unsigned long blk_max_low_pfn;
EXPORT_SYMBOL(blk_max_low_pfn);
@@ -823,6 +824,8 @@ EXPORT_SYMBOL(blk_queue_update_dma_alignment);
void blk_set_queue_depth(struct request_queue *q, unsigned int depth)
{
q->queue_depth = depth;
+ if (q->rq_wb)
+ blk_wb_update_limits(q->rq_wb, depth);
}
EXPORT_SYMBOL(blk_set_queue_depth);
@@ -13,6 +13,7 @@
#include "blk.h"
#include "blk-mq.h"
+#include "blk-wb.h"
struct queue_sysfs_entry {
struct attribute attr;
@@ -347,6 +348,76 @@ static ssize_t queue_poll_store(struct request_queue *q, const char *page,
return ret;
}
+static ssize_t queue_wb_stats_show(struct request_queue *q, char *page)
+{
+ struct rq_wb *wb = q->rq_wb;
+
+ if (!q->rq_wb)
+ return -EINVAL;
+
+ return sprintf(page, "idle=%d, normal=%d, max=%d, inflight=%d, wait=%d,"
+ " timer=%d, bdp_wait=%d\n", wb->wb_idle,
+ wb->wb_normal, wb->wb_max,
+ atomic_read(&wb->inflight),
+ waitqueue_active(&wb->wait),
+ timer_pending(&wb->timer),
+ *wb->bdp_wait);
+}
+
+static ssize_t queue_wb_perc_show(struct request_queue *q, char *page)
+{
+ if (!q->rq_wb)
+ return -EINVAL;
+
+ return queue_var_show(q->rq_wb->perc, page);
+}
+
+static ssize_t queue_wb_perc_store(struct request_queue *q, const char *page,
+ size_t count)
+{
+ unsigned long perc;
+ ssize_t ret;
+
+ if (!q->rq_wb)
+ return -EINVAL;
+
+ ret = queue_var_store(&perc, page, count);
+ if (ret < 0)
+ return ret;
+ if (perc > 100)
+ return -EINVAL;
+
+ q->rq_wb->perc = perc;
+ blk_wb_update_limits(q->rq_wb, blk_queue_depth(q));
+ return ret;
+}
+
+static ssize_t queue_wb_cache_delay_show(struct request_queue *q, char *page)
+{
+ if (!q->rq_wb)
+ return -EINVAL;
+
+ return queue_var_show(q->rq_wb->cache_delay_usecs, page);
+}
+
+static ssize_t queue_wb_cache_delay_store(struct request_queue *q,
+ const char *page, size_t count)
+{
+ unsigned long var;
+ ssize_t ret;
+
+ if (!q->rq_wb)
+ return -EINVAL;
+
+ ret = queue_var_store(&var, page, count);
+ if (ret < 0)
+ return ret;
+
+ q->rq_wb->cache_delay_usecs = var;
+ q->rq_wb->cache_delay = usecs_to_jiffies(var);
+ return ret;
+}
+
static ssize_t queue_wc_show(struct request_queue *q, char *page)
{
if (test_bit(QUEUE_FLAG_WC, &q->queue_flags))
@@ -516,6 +587,21 @@ static struct queue_sysfs_entry queue_wc_entry = {
.store = queue_wc_store,
};
+static struct queue_sysfs_entry queue_wb_stats_entry = {
+ .attr = {.name = "wb_stats", .mode = S_IRUGO },
+ .show = queue_wb_stats_show,
+};
+static struct queue_sysfs_entry queue_wb_cache_delay_entry = {
+ .attr = {.name = "wb_cache_usecs", .mode = S_IRUGO | S_IWUSR },
+ .show = queue_wb_cache_delay_show,
+ .store = queue_wb_cache_delay_store,
+};
+static struct queue_sysfs_entry queue_wb_perc_entry = {
+ .attr = {.name = "wb_percent", .mode = S_IRUGO | S_IWUSR },
+ .show = queue_wb_perc_show,
+ .store = queue_wb_perc_store,
+};
+
static struct attribute *default_attrs[] = {
&queue_requests_entry.attr,
&queue_ra_entry.attr,
@@ -542,6 +628,9 @@ static struct attribute *default_attrs[] = {
&queue_random_entry.attr,
&queue_poll_entry.attr,
&queue_wc_entry.attr,
+ &queue_wb_stats_entry.attr,
+ &queue_wb_cache_delay_entry.attr,
+ &queue_wb_perc_entry.attr,
NULL,
};
new file mode 100644
@@ -0,0 +1,238 @@
+/*
+ * buffered writeback throttling
+ *
+ * Copyright (C) 2016 Jens Axboe
+ *
+ * Things that need changing:
+ *
+ * - Auto-detection of optimal wb_percent setting. A lower setting
+ * is appropriate on rotating storage (wb_percent=15 gives good
+ * separation, while still getting full bandwidth with wb cache).
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
+
+#include "blk.h"
+#include "blk-wb.h"
+
+static inline bool rwb_enabled(struct rq_wb *rwb)
+{
+ return rwb->wb_normal != 0;
+}
+
+void __blk_wb_done(struct rq_wb *rwb)
+{
+ int inflight, limit = rwb->wb_normal;
+
+ inflight = atomic_dec_return(&rwb->inflight);
+ if (inflight >= limit)
+ return;
+
+ /*
+ * If the device does caching, we can still flood it with IO
+ * even at a low depth. If caching is on, delay a bit before
+ * submitting the next, if we're still purely background
+ * activity.
+ */
+ if (test_bit(QUEUE_FLAG_WC, &rwb->q->queue_flags) && !*rwb->bdp_wait &&
+ time_before(jiffies, rwb->last_comp + rwb->cache_delay)) {
+ if (!timer_pending(&rwb->timer))
+ mod_timer(&rwb->timer, jiffies + rwb->cache_delay);
+ return;
+ }
+
+ if (waitqueue_active(&rwb->wait)) {
+ int diff = limit - inflight;
+
+ if (diff >= rwb->wb_idle / 2)
+ wake_up_nr(&rwb->wait, 1);
+ }
+}
+
+/*
+ * Called on completion of a request. Note that it's also called when
+ * a request is merged, when the request gets freed.
+ */
+void blk_wb_done(struct rq_wb *rwb, struct request *rq)
+{
+ if (!(rq->cmd_flags & REQ_BUF_INFLIGHT)) {
+ if (rwb_enabled(rwb)) {
+ const unsigned long cur = jiffies;
+
+ if (cur != rwb->last_comp)
+ rwb->last_comp = cur;
+ }
+ } else
+ __blk_wb_done(rwb);
+}
+
+/*
+ * 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;
+}
+
+static inline unsigned int get_limit(struct rq_wb *rwb, unsigned int rw)
+{
+ unsigned int limit;
+
+ /*
+ * At this point we know it's a buffered write. If REQ_SYNC is
+ * set, then it's WB_SYNC_ALL writeback. Bump the limit 4x for
+ * those, since someone is (or will be) waiting on that.
+ */
+ if ((rw & REQ_SYNC) || *rwb->bdp_wait)
+ limit = rwb->wb_max;
+ else if (time_before(jiffies, rwb->last_comp + HZ / 10)) {
+ /*
+ * If less than 100ms since we completed unrelated IO,
+ * limit us to half the depth for background writeback.
+ */
+ limit = rwb->wb_idle;
+ } else
+ limit = rwb->wb_normal;
+
+ return limit;
+}
+
+/*
+ * Block if we will exceed our limit, or if we are currently waiting for
+ * the timer to kick off queuing again.
+ */
+static void __blk_wb_wait(struct rq_wb *rwb, unsigned int rw, spinlock_t *lock)
+{
+ DEFINE_WAIT(wait);
+
+ if (!timer_pending(&rwb->timer) &&
+ atomic_inc_below(&rwb->inflight, get_limit(rwb, rw)))
+ return;
+
+ do {
+ prepare_to_wait_exclusive(&rwb->wait, &wait,
+ TASK_UNINTERRUPTIBLE);
+
+ if (!timer_pending(&rwb->timer) &&
+ atomic_inc_below(&rwb->inflight, get_limit(rwb, rw)))
+ break;
+
+ if (lock)
+ spin_unlock_irq(lock);
+
+ io_schedule();
+
+ if (lock)
+ spin_lock_irq(lock);
+ } while (1);
+
+ finish_wait(&rwb->wait, &wait);
+}
+
+/*
+ * 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.
+ */
+bool blk_wb_wait(struct rq_wb *rwb, struct bio *bio, spinlock_t *lock)
+{
+ /*
+ * If disabled, or not a WRITE (or a discard), do nothing
+ */
+ if (!rwb_enabled(rwb) || !(bio->bi_rw & REQ_WRITE) ||
+ (bio->bi_rw & REQ_DISCARD))
+ return false;
+
+ /*
+ * Don't throttle WRITE_ODIRECT
+ */
+ if ((bio->bi_rw & (REQ_SYNC | REQ_NOIDLE)) == REQ_SYNC)
+ return false;
+
+ __blk_wb_wait(rwb, bio->bi_rw, lock);
+ return true;
+}
+
+static void calc_wb_limits(struct rq_wb *rwb, unsigned int depth,
+ unsigned int perc)
+{
+ /*
+ * We'll use depth==64 as a reasonable max limit that should be able
+ * to achieve full device bandwidth anywhere.
+ */
+ depth = min(64U, depth);
+
+ /*
+ * Full perf writes are max 'perc' percentage of the depth
+ */
+ rwb->wb_max = (perc * depth + 1) / 100;
+ if (!rwb->wb_max && perc)
+ rwb->wb_max = 1;
+ rwb->wb_normal = (rwb->wb_max + 1) / 2;
+ rwb->wb_idle = (rwb->wb_max + 3) / 4;
+}
+
+void blk_wb_update_limits(struct rq_wb *rwb, unsigned int depth)
+{
+ calc_wb_limits(rwb, depth, rwb->perc);
+ wake_up_all(&rwb->wait);
+}
+
+static void blk_wb_timer(unsigned long data)
+{
+ struct rq_wb *rwb = (struct rq_wb *) data;
+
+ if (waitqueue_active(&rwb->wait))
+ wake_up_nr(&rwb->wait, 1);
+}
+
+#define DEF_WB_PERC 50
+#define DEF_WB_CACHE_DELAY 10000
+
+int blk_wb_init(struct request_queue *q)
+{
+ struct rq_wb *rwb;
+
+ rwb = kzalloc(sizeof(*rwb), GFP_KERNEL);
+ if (!rwb)
+ return -ENOMEM;
+
+ atomic_set(&rwb->inflight, 0);
+ init_waitqueue_head(&rwb->wait);
+ rwb->last_comp = jiffies;
+ rwb->bdp_wait = &q->backing_dev_info.wb.dirty_sleeping;
+ setup_timer(&rwb->timer, blk_wb_timer, (unsigned long) rwb);
+ rwb->perc = DEF_WB_PERC;
+ rwb->cache_delay_usecs = DEF_WB_CACHE_DELAY;
+ rwb->cache_delay = usecs_to_jiffies(rwb->cache_delay);
+ rwb->q = q;
+ blk_wb_update_limits(rwb, blk_queue_depth(q));
+ q->rq_wb = rwb;
+ return 0;
+}
+
+void blk_wb_exit(struct request_queue *q)
+{
+ if (q->rq_wb)
+ del_timer_sync(&q->rq_wb->timer);
+
+ kfree(q->rq_wb);
+ q->rq_wb = NULL;
+}
new file mode 100644
@@ -0,0 +1,33 @@
+#ifndef BLK_WB_H
+#define BLK_WB_H
+
+#include <linux/atomic.h>
+#include <linux/wait.h>
+
+struct rq_wb {
+ /*
+ * Settings that govern how we throttle
+ */
+ unsigned int perc; /* INPUT */
+ unsigned int wb_idle; /* idle writeback */
+ unsigned int wb_normal; /* normal writeback */
+ unsigned int wb_max; /* max throughput writeback */
+
+ unsigned int cache_delay;
+ unsigned int cache_delay_usecs;
+ unsigned long last_comp;
+ unsigned int *bdp_wait;
+ struct request_queue *q;
+ atomic_t inflight;
+ wait_queue_head_t wait;
+ struct timer_list timer;
+};
+
+void __blk_wb_done(struct rq_wb *);
+void blk_wb_done(struct rq_wb *, struct request *);
+bool blk_wb_wait(struct rq_wb *, struct bio *, spinlock_t *);
+int blk_wb_init(struct request_queue *);
+void blk_wb_exit(struct request_queue *);
+void blk_wb_update_limits(struct rq_wb *, unsigned int);
+
+#endif
@@ -188,6 +188,7 @@ enum rq_flag_bits {
__REQ_PM, /* runtime pm request */
__REQ_HASHED, /* on IO scheduler merge hash */
__REQ_MQ_INFLIGHT, /* track inflight for MQ */
+ __REQ_BUF_INFLIGHT, /* track inflight for buffered */
__REQ_NR_BITS, /* stops here */
};
@@ -241,6 +242,7 @@ enum rq_flag_bits {
#define REQ_PM (1ULL << __REQ_PM)
#define REQ_HASHED (1ULL << __REQ_HASHED)
#define REQ_MQ_INFLIGHT (1ULL << __REQ_MQ_INFLIGHT)
+#define REQ_BUF_INFLIGHT (1ULL << __REQ_BUF_INFLIGHT)
typedef unsigned int blk_qc_t;
#define BLK_QC_T_NONE -1U
@@ -37,6 +37,7 @@ struct bsg_job;
struct blkcg_gq;
struct blk_flush_queue;
struct pr_ops;
+struct rq_wb;
#define BLKDEV_MIN_RQ 4
#define BLKDEV_MAX_RQ 128 /* Default maximum */
@@ -290,6 +291,8 @@ struct request_queue {
int nr_rqs[2]; /* # allocated [a]sync rqs */
int nr_rqs_elvpriv; /* # allocated rqs w/ elvpriv */
+ struct rq_wb *rq_wb;
+
/*
* If blkcg is not used, @q->root_rl serves all requests. If blkcg
* is used, root blkg allocates from @q->root_rl and all other
Test patch that throttles buffered writeback to make it a lot more smooth, and has way less impact on other system activity. Background writeback should be, by definition, background activity. The fact that we flush huge bundles of it at the time means that it potentially has heavy impacts on foreground workloads, which isn't ideal. We can't easily limit the sizes of writes that we do, since that would impact file system layout in the presence of delayed allocation. So just throttle back buffered writeback, unless someone is waiting for it. This is just a test patch, and as such, it registers a queue sysfs entry to both monitor the current state: $ cat /sys/block/nvme0n1/queue/wb_stats idle=16, normal=32, max=64, inflight=0, wait=0, timer=0, bdp_wait=0 'idle' denotes how many requests we will allow inflight for idle buffered writeback, 'normal' for higher priority writeback, and 'max' for when it's urgent we clean pages. The values are calculated based on the queue depth of the device, and the 'wb_percent' setting. If 'wb_percent' is set to zero, the functionality is turned off. 'inflight' shows how many requests are currently inflight for buffered writeback, 'wait' shows if anyone is currently waiting for access, 'timer' shows if we have processes being deferred in write back cache timeout, and bdp_wait shows if someone is currently throttled on this device in balance_dirty_pages(). Finally, if the device has write back caching, 'wb_cache_delay' delays by this amount of usecs when a write completes before allowing more. It'd be nice to auto-tune 'wb_percent' based on device response. Flash is less picky than rotating storage, but still needs throttling. For flash storage, a wb_percent setting of 50% gives good read latencies while still having good write bandwidth. For rotating storage, lower settings (like 10-15%) are more reasonable. Signed-off-by: Jens Axboe <axboe@fb.com> --- block/Makefile | 2 +- block/blk-core.c | 15 +++ block/blk-mq.c | 31 +++++- block/blk-settings.c | 3 + block/blk-sysfs.c | 89 +++++++++++++++++ block/blk-wb.c | 238 ++++++++++++++++++++++++++++++++++++++++++++++ block/blk-wb.h | 33 +++++++ include/linux/blk_types.h | 2 + include/linux/blkdev.h | 3 + 9 files changed, 413 insertions(+), 3 deletions(-) create mode 100644 block/blk-wb.c create mode 100644 block/blk-wb.h