linux-block.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats
@ 2019-11-06 21:58 Tejun Heo
  2019-11-06 21:58 ` [PATCH 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios Tejun Heo
                   ` (5 more replies)
  0 siblings, 6 replies; 9+ messages in thread
From: Tejun Heo @ 2019-11-06 21:58 UTC (permalink / raw)
  To: axboe; +Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team

Hello,

blk-cgroup IO stats currently use blkg_rwstat which unforutnately
requires walking all descendants recursively on read.  On systems with
a large number of cgroups (dead or alive), this can make each stat
read a substantially expensive operation.

This patch updates blk-cgroup to use cgroup rstat which makes stat
reading O(# descendants which have been active since last reading)
instead of O(# descendants).

 0001-bfq-iosched-stop-using-blkg-stat_bytes-and-stat_ios.patch
 0002-blk-throtl-stop-using-blkg-stat_bytes-and-stat_ios.patch
 0003-blk-cgroup-remove-now-unused-blkg_print_stat_-bytes-.patch
 0004-blk-cgroup-reimplement-basic-IO-stats-using-cgroup-r.patch
 0005-blk-cgroup-separate-out-blkg_rwstat-under-CONFIG_BLK.patch

0001-0002 make bfq-iosched and blk-throtl use their own blkg_rwstat to
track basic IO stats on cgroup1 instead of sharing blk-cgroup core's.
0003 is a follow-up cleanup.

0004 switches blk-cgroup to cgroup rstat.

0005 moves blkg_rwstat to its own files and gate it under a config
option as it's now only used by blk-throtl and bfq-iosched.

The patchset is on top of

  for-next/block 257855f86aa2 ("Merge branch 'for-5.4/block' into for-next")
+ [1] blkcg: make blkcg_print_stat() print stats only for online blkgs

and also available in the following git branch.

 git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git review-blkcg-rstat

Thanks.

 block/Kconfig              |    4 
 block/Kconfig.iosched      |    1 
 block/Makefile             |    1 
 block/bfq-cgroup.c         |   37 +++--
 block/bfq-iosched.c        |    4 
 block/bfq-iosched.h        |    6 
 block/blk-cgroup-rwstat.c  |  129 +++++++++++++++++++
 block/blk-cgroup-rwstat.h  |  149 ++++++++++++++++++++++
 block/blk-cgroup.c         |  304 ++++++++++++++-------------------------------
 block/blk-throttle.c       |   71 +++++++++-
 include/linux/blk-cgroup.h |  198 +++++------------------------
 11 files changed, 517 insertions(+), 387 deletions(-)

--
tejun

[1] http://lkml.kernel.org/r/20191105160951.GS3622521@devbig004.ftw2.facebook.com


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios
  2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
@ 2019-11-06 21:58 ` Tejun Heo
  2019-11-07 19:05   ` [PATCH v2 " Tejun Heo
  2019-11-06 21:58 ` [PATCH 2/5] blk-throtl: " Tejun Heo
                   ` (4 subsequent siblings)
  5 siblings, 1 reply; 9+ messages in thread
From: Tejun Heo @ 2019-11-06 21:58 UTC (permalink / raw)
  To: axboe
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team,
	Tejun Heo, Paolo Valente

When used on cgroup1, bfq uses the blkg->stat_bytes and ->stat_ios
from blk-cgroup core to populate six stat knobs.  blk-cgroup core is
moving away from blkg_rwstat to improve scalability and won't be able
to support this usage.

It isn't like the sharing gains all that much.  Let's break it out to
dedicated rwstat counters which are updated when on cgroup1.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Paolo Valente <paolo.valente@linaro.org>
---
 block/bfq-cgroup.c  | 37 ++++++++++++++++++++++++++-----------
 block/bfq-iosched.c |  4 ++++
 block/bfq-iosched.h |  4 ++++
 3 files changed, 34 insertions(+), 11 deletions(-)

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index 86a607cf19a1..c3093fde2e26 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -347,6 +347,14 @@ void bfqg_and_blkg_put(struct bfq_group *bfqg)
 	bfqg_put(bfqg);
 }
 
+void bfqg_stats_update_legacy_io(struct request_queue *q, struct request *rq)
+{
+	struct bfq_group *bfqg = blkg_to_bfqg(rq->bio->bi_blkg);
+
+	blkg_rwstat_add(&bfqg->stats.bytes, rq->cmd_flags, blk_rq_bytes(rq));
+	blkg_rwstat_add(&bfqg->stats.ios, rq->cmd_flags, 1);
+}
+
 /* @stats = 0 */
 static void bfqg_stats_reset(struct bfqg_stats *stats)
 {
@@ -431,6 +439,8 @@ void bfq_init_entity(struct bfq_entity *entity, struct bfq_group *bfqg)
 
 static void bfqg_stats_exit(struct bfqg_stats *stats)
 {
+	blkg_rwstat_exit(&stats->bytes);
+	blkg_rwstat_exit(&stats->ios);
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	blkg_rwstat_exit(&stats->merged);
 	blkg_rwstat_exit(&stats->service_time);
@@ -448,6 +458,10 @@ static void bfqg_stats_exit(struct bfqg_stats *stats)
 
 static int bfqg_stats_init(struct bfqg_stats *stats, gfp_t gfp)
 {
+	if (blkg_rwstat_init(&stats->bytes, gfp) ||
+	    blkg_rwstat_init(&stats->ios, gfp))
+		return -ENOMEM;
+
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	if (blkg_rwstat_init(&stats->merged, gfp) ||
 	    blkg_rwstat_init(&stats->service_time, gfp) ||
@@ -1125,7 +1139,8 @@ static int bfqg_print_rwstat_recursive(struct seq_file *sf, void *v)
 static u64 bfqg_prfill_sectors(struct seq_file *sf, struct blkg_policy_data *pd,
 			       int off)
 {
-	u64 sum = blkg_rwstat_total(&pd->blkg->stat_bytes);
+	struct bfq_group *bfqg = blkg_to_bfqg(pd->blkg);
+	u64 sum = blkg_rwstat_total(&bfqg->stats.bytes);
 
 	return __blkg_prfill_u64(sf, pd, sum >> 9);
 }
@@ -1142,8 +1157,8 @@ static u64 bfqg_prfill_sectors_recursive(struct seq_file *sf,
 {
 	struct blkg_rwstat_sample tmp;
 
-	blkg_rwstat_recursive_sum(pd->blkg, NULL,
-			offsetof(struct blkcg_gq, stat_bytes), &tmp);
+	blkg_rwstat_recursive_sum(pd->blkg, &blkcg_policy_bfq,
+			offsetof(struct bfq_group, stats.bytes), &tmp);
 
 	return __blkg_prfill_u64(sf, pd,
 		(tmp.cnt[BLKG_RWSTAT_READ] + tmp.cnt[BLKG_RWSTAT_WRITE]) >> 9);
@@ -1226,13 +1241,13 @@ struct cftype bfq_blkcg_legacy_files[] = {
 	/* statistics, covers only the tasks in the bfqg */
 	{
 		.name = "bfq.io_service_bytes",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_bytes,
+		.private = offsetof(struct bfq_group, stats.bytes),
+		.seq_show = bfqg_print_rwstat,
 	},
 	{
 		.name = "bfq.io_serviced",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_ios,
+		.private = offsetof(struct bfq_group, stats.ios),
+		.seq_show = bfqg_print_rwstat,
 	},
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	{
@@ -1269,13 +1284,13 @@ struct cftype bfq_blkcg_legacy_files[] = {
 	/* the same statistics which cover the bfqg and its descendants */
 	{
 		.name = "bfq.io_service_bytes_recursive",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_bytes_recursive,
+		.private = offsetof(struct bfq_group, stats.bytes),
+		.seq_show = bfqg_print_rwstat_recursive,
 	},
 	{
 		.name = "bfq.io_serviced_recursive",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_ios_recursive,
+		.private = offsetof(struct bfq_group, stats.ios),
+		.seq_show = bfqg_print_rwstat_recursive,
 	},
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	{
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index b33be928d164..4629f5580d46 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -5450,6 +5450,10 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 	bool idle_timer_disabled = false;
 	unsigned int cmd_flags;
 
+#ifdef CONFIG_BFQ_GROUP_IOSCHED
+	if (!cgroup_subsys_on_dfl(io_cgrp_subsys) && rq->bio)
+		bfqg_stats_update_legacy_io(q, rq);
+#endif
 	spin_lock_irq(&bfqd->lock);
 	if (blk_mq_sched_try_insert_merge(q, rq)) {
 		spin_unlock_irq(&bfqd->lock);
diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h
index 5d1a519640f6..2676c06218f1 100644
--- a/block/bfq-iosched.h
+++ b/block/bfq-iosched.h
@@ -809,6 +809,9 @@ struct bfq_stat {
 };
 
 struct bfqg_stats {
+	/* basic stats */
+	struct blkg_rwstat		bytes;
+	struct blkg_rwstat		ios;
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	/* number of ios merged */
 	struct blkg_rwstat		merged;
@@ -956,6 +959,7 @@ void bfq_put_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg);
 
 /* ---------------- cgroups-support interface ---------------- */
 
+void bfqg_stats_update_legacy_io(struct request_queue *q, struct request *rq);
 void bfqg_stats_update_io_add(struct bfq_group *bfqg, struct bfq_queue *bfqq,
 			      unsigned int op);
 void bfqg_stats_update_io_remove(struct bfq_group *bfqg, unsigned int op);
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 2/5] blk-throtl: stop using blkg->stat_bytes and ->stat_ios
  2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
  2019-11-06 21:58 ` [PATCH 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios Tejun Heo
@ 2019-11-06 21:58 ` Tejun Heo
  2019-11-06 21:58 ` [PATCH 3/5] blk-cgroup: remove now unused blkg_print_stat_{bytes|ios}_recursive() Tejun Heo
                   ` (3 subsequent siblings)
  5 siblings, 0 replies; 9+ messages in thread
From: Tejun Heo @ 2019-11-06 21:58 UTC (permalink / raw)
  To: axboe
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team,
	Tejun Heo

When used on cgroup1, blk-throtl uses the blkg->stat_bytes and
->stat_ios from blk-cgroup core to populate four stat knobs.
blk-cgroup core is moving away from blkg_rwstat to improve scalability
and won't be able to support this usage.

It isn't like the sharing gains all that much.  Let's break them out
to dedicated rwstat counters which are updated when on cgroup1.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 block/blk-throttle.c | 70 ++++++++++++++++++++++++++++++++++++++------
 1 file changed, 61 insertions(+), 9 deletions(-)

diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 0445c998c377..6ac5b5d56170 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -176,6 +176,9 @@ struct throtl_grp {
 	unsigned int bio_cnt; /* total bios */
 	unsigned int bad_bio_cnt; /* bios exceeding latency threshold */
 	unsigned long bio_cnt_reset_time;
+
+	struct blkg_rwstat stat_bytes;
+	struct blkg_rwstat stat_ios;
 };
 
 /* We measure latency for request size from <= 4k to >= 1M */
@@ -489,6 +492,12 @@ static struct blkg_policy_data *throtl_pd_alloc(gfp_t gfp,
 	if (!tg)
 		return NULL;
 
+	if (blkg_rwstat_init(&tg->stat_bytes, gfp))
+		goto err_free_tg;
+
+	if (blkg_rwstat_init(&tg->stat_ios, gfp))
+		goto err_exit_stat_bytes;
+
 	throtl_service_queue_init(&tg->service_queue);
 
 	for (rw = READ; rw <= WRITE; rw++) {
@@ -513,6 +522,12 @@ static struct blkg_policy_data *throtl_pd_alloc(gfp_t gfp,
 	tg->idletime_threshold_conf = DFL_IDLE_THRESHOLD;
 
 	return &tg->pd;
+
+err_exit_stat_bytes:
+	blkg_rwstat_exit(&tg->stat_bytes);
+err_free_tg:
+	kfree(tg);
+	return NULL;
 }
 
 static void throtl_pd_init(struct blkg_policy_data *pd)
@@ -611,6 +626,8 @@ static void throtl_pd_free(struct blkg_policy_data *pd)
 	struct throtl_grp *tg = pd_to_tg(pd);
 
 	del_timer_sync(&tg->service_queue.pending_timer);
+	blkg_rwstat_exit(&tg->stat_bytes);
+	blkg_rwstat_exit(&tg->stat_ios);
 	kfree(tg);
 }
 
@@ -1464,6 +1481,32 @@ static ssize_t tg_set_conf_uint(struct kernfs_open_file *of,
 	return tg_set_conf(of, buf, nbytes, off, false);
 }
 
+static int tg_print_rwstat(struct seq_file *sf, void *v)
+{
+	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
+			  blkg_prfill_rwstat, &blkcg_policy_throtl,
+			  seq_cft(sf)->private, true);
+	return 0;
+}
+
+static u64 tg_prfill_rwstat_recursive(struct seq_file *sf,
+				      struct blkg_policy_data *pd, int off)
+{
+	struct blkg_rwstat_sample sum;
+
+	blkg_rwstat_recursive_sum(pd_to_blkg(pd), &blkcg_policy_throtl, off,
+				  &sum);
+	return __blkg_prfill_rwstat(sf, pd, &sum);
+}
+
+static int tg_print_rwstat_recursive(struct seq_file *sf, void *v)
+{
+	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
+			  tg_prfill_rwstat_recursive, &blkcg_policy_throtl,
+			  seq_cft(sf)->private, true);
+	return 0;
+}
+
 static struct cftype throtl_legacy_files[] = {
 	{
 		.name = "throttle.read_bps_device",
@@ -1491,23 +1534,23 @@ static struct cftype throtl_legacy_files[] = {
 	},
 	{
 		.name = "throttle.io_service_bytes",
-		.private = (unsigned long)&blkcg_policy_throtl,
-		.seq_show = blkg_print_stat_bytes,
+		.private = offsetof(struct throtl_grp, stat_bytes),
+		.seq_show = tg_print_rwstat,
 	},
 	{
 		.name = "throttle.io_service_bytes_recursive",
-		.private = (unsigned long)&blkcg_policy_throtl,
-		.seq_show = blkg_print_stat_bytes_recursive,
+		.private = offsetof(struct throtl_grp, stat_bytes),
+		.seq_show = tg_print_rwstat_recursive,
 	},
 	{
 		.name = "throttle.io_serviced",
-		.private = (unsigned long)&blkcg_policy_throtl,
-		.seq_show = blkg_print_stat_ios,
+		.private = offsetof(struct throtl_grp, stat_ios),
+		.seq_show = tg_print_rwstat,
 	},
 	{
 		.name = "throttle.io_serviced_recursive",
-		.private = (unsigned long)&blkcg_policy_throtl,
-		.seq_show = blkg_print_stat_ios_recursive,
+		.private = offsetof(struct throtl_grp, stat_ios),
+		.seq_show = tg_print_rwstat_recursive,
 	},
 	{ }	/* terminate */
 };
@@ -2127,7 +2170,16 @@ bool blk_throtl_bio(struct request_queue *q, struct blkcg_gq *blkg,
 	WARN_ON_ONCE(!rcu_read_lock_held());
 
 	/* see throtl_charge_bio() */
-	if (bio_flagged(bio, BIO_THROTTLED) || !tg->has_rules[rw])
+	if (bio_flagged(bio, BIO_THROTTLED))
+		goto out;
+
+	if (!cgroup_subsys_on_dfl(io_cgrp_subsys)) {
+		blkg_rwstat_add(&tg->stat_bytes, bio->bi_opf,
+				bio->bi_iter.bi_size);
+		blkg_rwstat_add(&tg->stat_ios, bio->bi_opf, 1);
+	}
+
+	if (!tg->has_rules[rw])
 		goto out;
 
 	spin_lock_irq(&q->queue_lock);
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 3/5] blk-cgroup: remove now unused blkg_print_stat_{bytes|ios}_recursive()
  2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
  2019-11-06 21:58 ` [PATCH 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios Tejun Heo
  2019-11-06 21:58 ` [PATCH 2/5] blk-throtl: " Tejun Heo
@ 2019-11-06 21:58 ` Tejun Heo
  2019-11-06 21:58 ` [PATCH 4/5] blk-cgroup: reimplement basic IO stats using cgroup rstat Tejun Heo
                   ` (2 subsequent siblings)
  5 siblings, 0 replies; 9+ messages in thread
From: Tejun Heo @ 2019-11-06 21:58 UTC (permalink / raw)
  To: axboe
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team,
	Tejun Heo

These don't have users anymore.  Remove them.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 block/blk-cgroup.c         | 83 --------------------------------------
 include/linux/blk-cgroup.h |  5 ---
 2 files changed, 88 deletions(-)

diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 238d99aa0a99..e23a21ba6d46 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -615,89 +615,6 @@ u64 blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
 }
 EXPORT_SYMBOL_GPL(blkg_prfill_rwstat);
 
-static u64 blkg_prfill_rwstat_field(struct seq_file *sf,
-				    struct blkg_policy_data *pd, int off)
-{
-	struct blkg_rwstat_sample rwstat = { };
-
-	blkg_rwstat_read((void *)pd->blkg + off, &rwstat);
-	return __blkg_prfill_rwstat(sf, pd, &rwstat);
-}
-
-/**
- * blkg_print_stat_bytes - seq_show callback for blkg->stat_bytes
- * @sf: seq_file to print to
- * @v: unused
- *
- * To be used as cftype->seq_show to print blkg->stat_bytes.
- * cftype->private must be set to the blkcg_policy.
- */
-int blkg_print_stat_bytes(struct seq_file *sf, void *v)
-{
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
-			  blkg_prfill_rwstat_field, (void *)seq_cft(sf)->private,
-			  offsetof(struct blkcg_gq, stat_bytes), true);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(blkg_print_stat_bytes);
-
-/**
- * blkg_print_stat_bytes - seq_show callback for blkg->stat_ios
- * @sf: seq_file to print to
- * @v: unused
- *
- * To be used as cftype->seq_show to print blkg->stat_ios.  cftype->private
- * must be set to the blkcg_policy.
- */
-int blkg_print_stat_ios(struct seq_file *sf, void *v)
-{
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
-			  blkg_prfill_rwstat_field, (void *)seq_cft(sf)->private,
-			  offsetof(struct blkcg_gq, stat_ios), true);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(blkg_print_stat_ios);
-
-static u64 blkg_prfill_rwstat_field_recursive(struct seq_file *sf,
-					      struct blkg_policy_data *pd,
-					      int off)
-{
-	struct blkg_rwstat_sample rwstat;
-
-	blkg_rwstat_recursive_sum(pd->blkg, NULL, off, &rwstat);
-	return __blkg_prfill_rwstat(sf, pd, &rwstat);
-}
-
-/**
- * blkg_print_stat_bytes_recursive - recursive version of blkg_print_stat_bytes
- * @sf: seq_file to print to
- * @v: unused
- */
-int blkg_print_stat_bytes_recursive(struct seq_file *sf, void *v)
-{
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
-			  blkg_prfill_rwstat_field_recursive,
-			  (void *)seq_cft(sf)->private,
-			  offsetof(struct blkcg_gq, stat_bytes), true);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(blkg_print_stat_bytes_recursive);
-
-/**
- * blkg_print_stat_ios_recursive - recursive version of blkg_print_stat_ios
- * @sf: seq_file to print to
- * @v: unused
- */
-int blkg_print_stat_ios_recursive(struct seq_file *sf, void *v)
-{
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
-			  blkg_prfill_rwstat_field_recursive,
-			  (void *)seq_cft(sf)->private,
-			  offsetof(struct blkcg_gq, stat_ios), true);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(blkg_print_stat_ios_recursive);
-
 /**
  * blkg_rwstat_recursive_sum - collect hierarchical blkg_rwstat
  * @blkg: blkg of interest
diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h
index bed9e43f9426..914ce55fa8c2 100644
--- a/include/linux/blk-cgroup.h
+++ b/include/linux/blk-cgroup.h
@@ -220,11 +220,6 @@ u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
 			 const struct blkg_rwstat_sample *rwstat);
 u64 blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
 		       int off);
-int blkg_print_stat_bytes(struct seq_file *sf, void *v);
-int blkg_print_stat_ios(struct seq_file *sf, void *v);
-int blkg_print_stat_bytes_recursive(struct seq_file *sf, void *v);
-int blkg_print_stat_ios_recursive(struct seq_file *sf, void *v);
-
 void blkg_rwstat_recursive_sum(struct blkcg_gq *blkg, struct blkcg_policy *pol,
 		int off, struct blkg_rwstat_sample *sum);
 
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 4/5] blk-cgroup: reimplement basic IO stats using cgroup rstat
  2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
                   ` (2 preceding siblings ...)
  2019-11-06 21:58 ` [PATCH 3/5] blk-cgroup: remove now unused blkg_print_stat_{bytes|ios}_recursive() Tejun Heo
@ 2019-11-06 21:58 ` Tejun Heo
  2019-11-06 21:58 ` [PATCH 5/5] blk-cgroup: separate out blkg_rwstat under CONFIG_BLK_CGROUP_RWSTAT Tejun Heo
  2019-11-07 19:04 ` [PATCH 0.5/5] bfq-iosched: relocate bfqg_*rwstat*() helpers Tejun Heo
  5 siblings, 0 replies; 9+ messages in thread
From: Tejun Heo @ 2019-11-06 21:58 UTC (permalink / raw)
  To: axboe
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team,
	Tejun Heo, Dan Schatzberg, Daniel Xu

blk-cgroup has been using blkg_rwstat to track basic IO stats.
Unfortunately, reading recursive stats scales badly as itinvolves
walking all descendants.  On systems with a huge number of cgroups
(dead or alive), this can lead to substantial CPU cost when reading IO
stats.

This patch reimplements basic IO stats using cgroup rstat which uses
more memory but makes recursive stat reading O(# descendants which
have been active since last reading) instead of O(# descendants).

* blk-cgroup core no longer uses sync/async stats.  Introduce new stat
  enums - BLKG_IOSTAT_{READ|WRITE|DISCARD}.

* Add blkg_iostat[_set] which encapsulates byte and io stats, last
  values for propagation delta calculation and u64_stats_sync for
  correctness on 32bit archs.

* Update the new percpu stat counters directly and implement
  blkcg_rstat_flush() to implement propagation.

* blkg_print_stat() can now bring the stats up to date by calling
  cgroup_rstat_flush() and print them instead of directly summing up
  all descendants.

* It now allocates 96 bytes per cpu.  It used to be 40 bytes.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Dan Schatzberg <dschatzberg@fb.com>
Cc: Daniel Xu <dlxu@fb.com>
---
 block/blk-cgroup.c         | 124 +++++++++++++++++++++++++++++--------
 include/linux/blk-cgroup.h |  48 ++++++++++++--
 2 files changed, 142 insertions(+), 30 deletions(-)

diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index e23a21ba6d46..0b349f6a6a69 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -80,8 +80,7 @@ static void blkg_free(struct blkcg_gq *blkg)
 		if (blkg->pd[i])
 			blkcg_policy[i]->pd_free_fn(blkg->pd[i]);
 
-	blkg_rwstat_exit(&blkg->stat_ios);
-	blkg_rwstat_exit(&blkg->stat_bytes);
+	free_percpu(blkg->iostat_cpu);
 	percpu_ref_exit(&blkg->refcnt);
 	kfree(blkg);
 }
@@ -146,7 +145,7 @@ static struct blkcg_gq *blkg_alloc(struct blkcg *blkcg, struct request_queue *q,
 				   gfp_t gfp_mask)
 {
 	struct blkcg_gq *blkg;
-	int i;
+	int i, cpu;
 
 	/* alloc and init base part */
 	blkg = kzalloc_node(sizeof(*blkg), gfp_mask, q->node);
@@ -156,8 +155,8 @@ static struct blkcg_gq *blkg_alloc(struct blkcg *blkcg, struct request_queue *q,
 	if (percpu_ref_init(&blkg->refcnt, blkg_release, 0, gfp_mask))
 		goto err_free;
 
-	if (blkg_rwstat_init(&blkg->stat_bytes, gfp_mask) ||
-	    blkg_rwstat_init(&blkg->stat_ios, gfp_mask))
+	blkg->iostat_cpu = alloc_percpu_gfp(struct blkg_iostat_set, gfp_mask);
+	if (!blkg->iostat_cpu)
 		goto err_free;
 
 	blkg->q = q;
@@ -167,6 +166,10 @@ static struct blkcg_gq *blkg_alloc(struct blkcg *blkcg, struct request_queue *q,
 	INIT_WORK(&blkg->async_bio_work, blkg_async_bio_workfn);
 	blkg->blkcg = blkcg;
 
+	u64_stats_init(&blkg->iostat.sync);
+	for_each_possible_cpu(cpu)
+		u64_stats_init(&per_cpu_ptr(blkg->iostat_cpu, cpu)->sync);
+
 	for (i = 0; i < BLKCG_MAX_POLS; i++) {
 		struct blkcg_policy *pol = blkcg_policy[i];
 		struct blkg_policy_data *pd;
@@ -393,7 +396,6 @@ struct blkcg_gq *blkg_lookup_create(struct blkcg *blkcg,
 static void blkg_destroy(struct blkcg_gq *blkg)
 {
 	struct blkcg *blkcg = blkg->blkcg;
-	struct blkcg_gq *parent = blkg->parent;
 	int i;
 
 	lockdep_assert_held(&blkg->q->queue_lock);
@@ -410,11 +412,6 @@ static void blkg_destroy(struct blkcg_gq *blkg)
 			pol->pd_offline_fn(blkg->pd[i]);
 	}
 
-	if (parent) {
-		blkg_rwstat_add_aux(&parent->stat_bytes, &blkg->stat_bytes);
-		blkg_rwstat_add_aux(&parent->stat_ios, &blkg->stat_ios);
-	}
-
 	blkg->online = false;
 
 	radix_tree_delete(&blkcg->blkg_tree, blkg->q->id);
@@ -464,7 +461,7 @@ static int blkcg_reset_stats(struct cgroup_subsys_state *css,
 {
 	struct blkcg *blkcg = css_to_blkcg(css);
 	struct blkcg_gq *blkg;
-	int i;
+	int i, cpu;
 
 	mutex_lock(&blkcg_pol_mutex);
 	spin_lock_irq(&blkcg->lock);
@@ -475,8 +472,12 @@ static int blkcg_reset_stats(struct cgroup_subsys_state *css,
 	 * anyway.  If you get hit by a race, retry.
 	 */
 	hlist_for_each_entry(blkg, &blkcg->blkg_list, blkcg_node) {
-		blkg_rwstat_reset(&blkg->stat_bytes);
-		blkg_rwstat_reset(&blkg->stat_ios);
+		for_each_possible_cpu(cpu) {
+			struct blkg_iostat_set *bis =
+				per_cpu_ptr(blkg->iostat_cpu, cpu);
+			memset(bis, 0, sizeof(*bis));
+		}
+		memset(&blkg->iostat, 0, sizeof(blkg->iostat));
 
 		for (i = 0; i < BLKCG_MAX_POLS; i++) {
 			struct blkcg_policy *pol = blkcg_policy[i];
@@ -838,16 +839,18 @@ static int blkcg_print_stat(struct seq_file *sf, void *v)
 	struct blkcg *blkcg = css_to_blkcg(seq_css(sf));
 	struct blkcg_gq *blkg;
 
+	cgroup_rstat_flush(blkcg->css.cgroup);
 	rcu_read_lock();
 
 	hlist_for_each_entry_rcu(blkg, &blkcg->blkg_list, blkcg_node) {
+		struct blkg_iostat_set *bis = &blkg->iostat;
 		const char *dname;
 		char *buf;
-		struct blkg_rwstat_sample rwstat;
 		u64 rbytes, wbytes, rios, wios, dbytes, dios;
 		size_t size = seq_get_buf(sf, &buf), off = 0;
 		int i;
 		bool has_stats = false;
+		unsigned seq;
 
 		spin_lock_irq(&blkg->q->queue_lock);
 
@@ -866,17 +869,16 @@ static int blkcg_print_stat(struct seq_file *sf, void *v)
 		 */
 		off += scnprintf(buf+off, size-off, "%s ", dname);
 
-		blkg_rwstat_recursive_sum(blkg, NULL,
-				offsetof(struct blkcg_gq, stat_bytes), &rwstat);
-		rbytes = rwstat.cnt[BLKG_RWSTAT_READ];
-		wbytes = rwstat.cnt[BLKG_RWSTAT_WRITE];
-		dbytes = rwstat.cnt[BLKG_RWSTAT_DISCARD];
+		do {
+			seq = u64_stats_fetch_begin(&bis->sync);
 
-		blkg_rwstat_recursive_sum(blkg, NULL,
-					offsetof(struct blkcg_gq, stat_ios), &rwstat);
-		rios = rwstat.cnt[BLKG_RWSTAT_READ];
-		wios = rwstat.cnt[BLKG_RWSTAT_WRITE];
-		dios = rwstat.cnt[BLKG_RWSTAT_DISCARD];
+			rbytes = bis->cur.bytes[BLKG_IOSTAT_READ];
+			wbytes = bis->cur.bytes[BLKG_IOSTAT_WRITE];
+			dbytes = bis->cur.bytes[BLKG_IOSTAT_DISCARD];
+			rios = bis->cur.ios[BLKG_IOSTAT_READ];
+			wios = bis->cur.ios[BLKG_IOSTAT_WRITE];
+			dios = bis->cur.ios[BLKG_IOSTAT_DISCARD];
+		} while (u64_stats_fetch_retry(&bis->sync, seq));
 
 		if (rbytes || wbytes || rios || wios) {
 			has_stats = true;
@@ -1212,6 +1214,77 @@ static int blkcg_can_attach(struct cgroup_taskset *tset)
 	return ret;
 }
 
+static void blkg_iostat_set(struct blkg_iostat *dst, struct blkg_iostat *src)
+{
+	int i;
+
+	for (i = 0; i < BLKG_IOSTAT_NR; i++) {
+		dst->bytes[i] = src->bytes[i];
+		dst->ios[i] = src->ios[i];
+	}
+}
+
+static void blkg_iostat_add(struct blkg_iostat *dst, struct blkg_iostat *src)
+{
+	int i;
+
+	for (i = 0; i < BLKG_IOSTAT_NR; i++) {
+		dst->bytes[i] += src->bytes[i];
+		dst->ios[i] += src->ios[i];
+	}
+}
+
+static void blkg_iostat_sub(struct blkg_iostat *dst, struct blkg_iostat *src)
+{
+	int i;
+
+	for (i = 0; i < BLKG_IOSTAT_NR; i++) {
+		dst->bytes[i] -= src->bytes[i];
+		dst->ios[i] -= src->ios[i];
+	}
+}
+
+static void blkcg_rstat_flush(struct cgroup_subsys_state *css, int cpu)
+{
+	struct blkcg *blkcg = css_to_blkcg(css);
+	struct blkcg_gq *blkg;
+
+	rcu_read_lock();
+
+	hlist_for_each_entry_rcu(blkg, &blkcg->blkg_list, blkcg_node) {
+		struct blkcg_gq *parent = blkg->parent;
+		struct blkg_iostat_set *bisc = per_cpu_ptr(blkg->iostat_cpu, cpu);
+		struct blkg_iostat cur, delta;
+		unsigned seq;
+
+		/* fetch the current per-cpu values */
+		do {
+			seq = u64_stats_fetch_begin(&bisc->sync);
+			blkg_iostat_set(&cur, &bisc->cur);
+		} while (u64_stats_fetch_retry(&bisc->sync, seq));
+
+		/* propagate percpu delta to global */
+		u64_stats_update_begin(&blkg->iostat.sync);
+		blkg_iostat_set(&delta, &cur);
+		blkg_iostat_sub(&delta, &bisc->last);
+		blkg_iostat_add(&blkg->iostat.cur, &delta);
+		blkg_iostat_add(&bisc->last, &delta);
+		u64_stats_update_end(&blkg->iostat.sync);
+
+		/* propagate global delta to parent */
+		if (parent) {
+			u64_stats_update_begin(&parent->iostat.sync);
+			blkg_iostat_set(&delta, &blkg->iostat.cur);
+			blkg_iostat_sub(&delta, &blkg->iostat.last);
+			blkg_iostat_add(&parent->iostat.cur, &delta);
+			blkg_iostat_add(&blkg->iostat.last, &delta);
+			u64_stats_update_end(&parent->iostat.sync);
+		}
+	}
+
+	rcu_read_unlock();
+}
+
 static void blkcg_bind(struct cgroup_subsys_state *root_css)
 {
 	int i;
@@ -1244,6 +1317,7 @@ struct cgroup_subsys io_cgrp_subsys = {
 	.css_offline = blkcg_css_offline,
 	.css_free = blkcg_css_free,
 	.can_attach = blkcg_can_attach,
+	.css_rstat_flush = blkcg_rstat_flush,
 	.bind = blkcg_bind,
 	.dfl_cftypes = blkcg_files,
 	.legacy_cftypes = blkcg_legacy_files,
diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h
index 914ce55fa8c2..867ab391e409 100644
--- a/include/linux/blk-cgroup.h
+++ b/include/linux/blk-cgroup.h
@@ -15,7 +15,9 @@
  */
 
 #include <linux/cgroup.h>
+#include <linux/percpu.h>
 #include <linux/percpu_counter.h>
+#include <linux/u64_stats_sync.h>
 #include <linux/seq_file.h>
 #include <linux/radix-tree.h>
 #include <linux/blkdev.h>
@@ -31,6 +33,14 @@
 
 #ifdef CONFIG_BLK_CGROUP
 
+enum blkg_iostat_type {
+	BLKG_IOSTAT_READ,
+	BLKG_IOSTAT_WRITE,
+	BLKG_IOSTAT_DISCARD,
+
+	BLKG_IOSTAT_NR,
+};
+
 enum blkg_rwstat_type {
 	BLKG_RWSTAT_READ,
 	BLKG_RWSTAT_WRITE,
@@ -61,6 +71,17 @@ struct blkcg {
 #endif
 };
 
+struct blkg_iostat {
+	u64				bytes[BLKG_IOSTAT_NR];
+	u64				ios[BLKG_IOSTAT_NR];
+};
+
+struct blkg_iostat_set {
+	struct u64_stats_sync		sync;
+	struct blkg_iostat		cur;
+	struct blkg_iostat		last;
+};
+
 /*
  * blkg_[rw]stat->aux_cnt is excluded for local stats but included for
  * recursive.  Used to carry stats of dead children.
@@ -127,8 +148,8 @@ struct blkcg_gq {
 	/* is this blkg online? protected by both blkcg and q locks */
 	bool				online;
 
-	struct blkg_rwstat		stat_bytes;
-	struct blkg_rwstat		stat_ios;
+	struct blkg_iostat_set __percpu	*iostat_cpu;
+	struct blkg_iostat_set		iostat;
 
 	struct blkg_policy_data		*pd[BLKCG_MAX_POLS];
 
@@ -740,15 +761,32 @@ static inline bool blkcg_bio_issue_check(struct request_queue *q,
 	throtl = blk_throtl_bio(q, blkg, bio);
 
 	if (!throtl) {
+		struct blkg_iostat_set *bis;
+		int rwd, cpu;
+
+		if (op_is_discard(bio->bi_opf))
+			rwd = BLKG_IOSTAT_DISCARD;
+		else if (op_is_write(bio->bi_opf))
+			rwd = BLKG_IOSTAT_WRITE;
+		else
+			rwd = BLKG_IOSTAT_READ;
+
+		cpu = get_cpu();
+		bis = per_cpu_ptr(blkg->iostat_cpu, cpu);
+		u64_stats_update_begin(&bis->sync);
+
 		/*
 		 * If the bio is flagged with BIO_QUEUE_ENTERED it means this
 		 * is a split bio and we would have already accounted for the
 		 * size of the bio.
 		 */
 		if (!bio_flagged(bio, BIO_QUEUE_ENTERED))
-			blkg_rwstat_add(&blkg->stat_bytes, bio->bi_opf,
-					bio->bi_iter.bi_size);
-		blkg_rwstat_add(&blkg->stat_ios, bio->bi_opf, 1);
+			bis->cur.bytes[rwd] += bio->bi_iter.bi_size;
+		bis->cur.ios[rwd]++;
+
+		u64_stats_update_end(&bis->sync);
+		cgroup_rstat_updated(blkg->blkcg->css.cgroup, cpu);
+		put_cpu();
 	}
 
 	blkcg_bio_issue_init(bio);
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 5/5] blk-cgroup: separate out blkg_rwstat under CONFIG_BLK_CGROUP_RWSTAT
  2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
                   ` (3 preceding siblings ...)
  2019-11-06 21:58 ` [PATCH 4/5] blk-cgroup: reimplement basic IO stats using cgroup rstat Tejun Heo
@ 2019-11-06 21:58 ` Tejun Heo
  2019-11-07 19:04 ` [PATCH 0.5/5] bfq-iosched: relocate bfqg_*rwstat*() helpers Tejun Heo
  5 siblings, 0 replies; 9+ messages in thread
From: Tejun Heo @ 2019-11-06 21:58 UTC (permalink / raw)
  To: axboe
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team,
	Tejun Heo

blkg_rwstat is now only used by bfq-iosched and blk-throtl when on
cgroup1.  Let's move it into its own files and gate it behind a config
option.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 block/Kconfig              |   4 +
 block/Kconfig.iosched      |   1 +
 block/Makefile             |   1 +
 block/bfq-iosched.h        |   2 +
 block/blk-cgroup-rwstat.c  | 129 ++++++++++++++++++++++++++++++
 block/blk-cgroup-rwstat.h  | 149 ++++++++++++++++++++++++++++++++++
 block/blk-cgroup.c         |  97 ----------------------
 block/blk-throttle.c       |   1 +
 include/linux/blk-cgroup.h | 159 -------------------------------------
 9 files changed, 287 insertions(+), 256 deletions(-)
 create mode 100644 block/blk-cgroup-rwstat.c
 create mode 100644 block/blk-cgroup-rwstat.h

diff --git a/block/Kconfig b/block/Kconfig
index 41c0917ce622..c23094a14a2b 100644
--- a/block/Kconfig
+++ b/block/Kconfig
@@ -32,6 +32,9 @@ config BLK_RQ_ALLOC_TIME
 config BLK_SCSI_REQUEST
 	bool
 
+config BLK_CGROUP_RWSTAT
+	bool
+
 config BLK_DEV_BSG
 	bool "Block layer SG support v4"
 	default y
@@ -86,6 +89,7 @@ config BLK_DEV_ZONED
 config BLK_DEV_THROTTLING
 	bool "Block layer bio throttling support"
 	depends on BLK_CGROUP=y
+	select BLK_CGROUP_RWSTAT
 	---help---
 	Block layer bio throttling support. It can be used to limit
 	the IO rate to a device. IO rate policies are per cgroup and
diff --git a/block/Kconfig.iosched b/block/Kconfig.iosched
index b89310a022ad..7df14133adc8 100644
--- a/block/Kconfig.iosched
+++ b/block/Kconfig.iosched
@@ -31,6 +31,7 @@ config IOSCHED_BFQ
 config BFQ_GROUP_IOSCHED
        bool "BFQ hierarchical scheduling support"
        depends on IOSCHED_BFQ && BLK_CGROUP
+       select BLK_CGROUP_RWSTAT
        ---help---
 
        Enable hierarchical scheduling in BFQ, using the blkio
diff --git a/block/Makefile b/block/Makefile
index 9ef57ace90d4..205a5f2fef17 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -16,6 +16,7 @@ obj-$(CONFIG_BLK_SCSI_REQUEST)	+= scsi_ioctl.o
 obj-$(CONFIG_BLK_DEV_BSG)	+= bsg.o
 obj-$(CONFIG_BLK_DEV_BSGLIB)	+= bsg-lib.o
 obj-$(CONFIG_BLK_CGROUP)	+= blk-cgroup.o
+obj-$(CONFIG_BLK_CGROUP_RWSTAT)	+= blk-cgroup-rwstat.o
 obj-$(CONFIG_BLK_DEV_THROTTLING)	+= blk-throttle.o
 obj-$(CONFIG_BLK_CGROUP_IOLATENCY)	+= blk-iolatency.o
 obj-$(CONFIG_BLK_CGROUP_IOCOST)	+= blk-iocost.o
diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h
index 2676c06218f1..9c82c1f35716 100644
--- a/block/bfq-iosched.h
+++ b/block/bfq-iosched.h
@@ -10,6 +10,8 @@
 #include <linux/hrtimer.h>
 #include <linux/blk-cgroup.h>
 
+#include "blk-cgroup-rwstat.h"
+
 #define BFQ_IOPRIO_CLASSES	3
 #define BFQ_CL_IDLE_TIMEOUT	(HZ/5)
 
diff --git a/block/blk-cgroup-rwstat.c b/block/blk-cgroup-rwstat.c
new file mode 100644
index 000000000000..85d5790ac49b
--- /dev/null
+++ b/block/blk-cgroup-rwstat.c
@@ -0,0 +1,129 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * Legacy blkg rwstat helpers enabled by CONFIG_BLK_CGROUP_RWSTAT.
+ * Do not use in new code.
+ */
+#include "blk-cgroup-rwstat.h"
+
+int blkg_rwstat_init(struct blkg_rwstat *rwstat, gfp_t gfp)
+{
+	int i, ret;
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++) {
+		ret = percpu_counter_init(&rwstat->cpu_cnt[i], 0, gfp);
+		if (ret) {
+			while (--i >= 0)
+				percpu_counter_destroy(&rwstat->cpu_cnt[i]);
+			return ret;
+		}
+		atomic64_set(&rwstat->aux_cnt[i], 0);
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(blkg_rwstat_init);
+
+void blkg_rwstat_exit(struct blkg_rwstat *rwstat)
+{
+	int i;
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++)
+		percpu_counter_destroy(&rwstat->cpu_cnt[i]);
+}
+EXPORT_SYMBOL_GPL(blkg_rwstat_exit);
+
+/**
+ * __blkg_prfill_rwstat - prfill helper for a blkg_rwstat
+ * @sf: seq_file to print to
+ * @pd: policy private data of interest
+ * @rwstat: rwstat to print
+ *
+ * Print @rwstat to @sf for the device assocaited with @pd.
+ */
+u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
+			 const struct blkg_rwstat_sample *rwstat)
+{
+	static const char *rwstr[] = {
+		[BLKG_RWSTAT_READ]	= "Read",
+		[BLKG_RWSTAT_WRITE]	= "Write",
+		[BLKG_RWSTAT_SYNC]	= "Sync",
+		[BLKG_RWSTAT_ASYNC]	= "Async",
+		[BLKG_RWSTAT_DISCARD]	= "Discard",
+	};
+	const char *dname = blkg_dev_name(pd->blkg);
+	u64 v;
+	int i;
+
+	if (!dname)
+		return 0;
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++)
+		seq_printf(sf, "%s %s %llu\n", dname, rwstr[i],
+			   rwstat->cnt[i]);
+
+	v = rwstat->cnt[BLKG_RWSTAT_READ] +
+		rwstat->cnt[BLKG_RWSTAT_WRITE] +
+		rwstat->cnt[BLKG_RWSTAT_DISCARD];
+	seq_printf(sf, "%s Total %llu\n", dname, v);
+	return v;
+}
+EXPORT_SYMBOL_GPL(__blkg_prfill_rwstat);
+
+/**
+ * blkg_prfill_rwstat - prfill callback for blkg_rwstat
+ * @sf: seq_file to print to
+ * @pd: policy private data of interest
+ * @off: offset to the blkg_rwstat in @pd
+ *
+ * prfill callback for printing a blkg_rwstat.
+ */
+u64 blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
+		       int off)
+{
+	struct blkg_rwstat_sample rwstat = { };
+
+	blkg_rwstat_read((void *)pd + off, &rwstat);
+	return __blkg_prfill_rwstat(sf, pd, &rwstat);
+}
+EXPORT_SYMBOL_GPL(blkg_prfill_rwstat);
+
+/**
+ * blkg_rwstat_recursive_sum - collect hierarchical blkg_rwstat
+ * @blkg: blkg of interest
+ * @pol: blkcg_policy which contains the blkg_rwstat
+ * @off: offset to the blkg_rwstat in blkg_policy_data or @blkg
+ * @sum: blkg_rwstat_sample structure containing the results
+ *
+ * Collect the blkg_rwstat specified by @blkg, @pol and @off and all its
+ * online descendants and their aux counts.  The caller must be holding the
+ * queue lock for online tests.
+ *
+ * If @pol is NULL, blkg_rwstat is at @off bytes into @blkg; otherwise, it
+ * is at @off bytes into @blkg's blkg_policy_data of the policy.
+ */
+void blkg_rwstat_recursive_sum(struct blkcg_gq *blkg, struct blkcg_policy *pol,
+		int off, struct blkg_rwstat_sample *sum)
+{
+	struct blkcg_gq *pos_blkg;
+	struct cgroup_subsys_state *pos_css;
+	unsigned int i;
+
+	lockdep_assert_held(&blkg->q->queue_lock);
+
+	rcu_read_lock();
+	blkg_for_each_descendant_pre(pos_blkg, pos_css, blkg) {
+		struct blkg_rwstat *rwstat;
+
+		if (!pos_blkg->online)
+			continue;
+
+		if (pol)
+			rwstat = (void *)blkg_to_pd(pos_blkg, pol) + off;
+		else
+			rwstat = (void *)pos_blkg + off;
+
+		for (i = 0; i < BLKG_RWSTAT_NR; i++)
+			sum->cnt[i] = blkg_rwstat_read_counter(rwstat, i);
+	}
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL_GPL(blkg_rwstat_recursive_sum);
diff --git a/block/blk-cgroup-rwstat.h b/block/blk-cgroup-rwstat.h
new file mode 100644
index 000000000000..ee746919c41f
--- /dev/null
+++ b/block/blk-cgroup-rwstat.h
@@ -0,0 +1,149 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * Legacy blkg rwstat helpers enabled by CONFIG_BLK_CGROUP_RWSTAT.
+ * Do not use in new code.
+ */
+#ifndef _BLK_CGROUP_RWSTAT_H
+#define _BLK_CGROUP_RWSTAT_H
+
+#include <linux/blk-cgroup.h>
+
+enum blkg_rwstat_type {
+	BLKG_RWSTAT_READ,
+	BLKG_RWSTAT_WRITE,
+	BLKG_RWSTAT_SYNC,
+	BLKG_RWSTAT_ASYNC,
+	BLKG_RWSTAT_DISCARD,
+
+	BLKG_RWSTAT_NR,
+	BLKG_RWSTAT_TOTAL = BLKG_RWSTAT_NR,
+};
+
+/*
+ * blkg_[rw]stat->aux_cnt is excluded for local stats but included for
+ * recursive.  Used to carry stats of dead children.
+ */
+struct blkg_rwstat {
+	struct percpu_counter		cpu_cnt[BLKG_RWSTAT_NR];
+	atomic64_t			aux_cnt[BLKG_RWSTAT_NR];
+};
+
+struct blkg_rwstat_sample {
+	u64				cnt[BLKG_RWSTAT_NR];
+};
+
+static inline u64 blkg_rwstat_read_counter(struct blkg_rwstat *rwstat,
+		unsigned int idx)
+{
+	return atomic64_read(&rwstat->aux_cnt[idx]) +
+		percpu_counter_sum_positive(&rwstat->cpu_cnt[idx]);
+}
+
+int blkg_rwstat_init(struct blkg_rwstat *rwstat, gfp_t gfp);
+void blkg_rwstat_exit(struct blkg_rwstat *rwstat);
+u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
+			 const struct blkg_rwstat_sample *rwstat);
+u64 blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
+		       int off);
+void blkg_rwstat_recursive_sum(struct blkcg_gq *blkg, struct blkcg_policy *pol,
+		int off, struct blkg_rwstat_sample *sum);
+
+
+/**
+ * blkg_rwstat_add - add a value to a blkg_rwstat
+ * @rwstat: target blkg_rwstat
+ * @op: REQ_OP and flags
+ * @val: value to add
+ *
+ * Add @val to @rwstat.  The counters are chosen according to @rw.  The
+ * caller is responsible for synchronizing calls to this function.
+ */
+static inline void blkg_rwstat_add(struct blkg_rwstat *rwstat,
+				   unsigned int op, uint64_t val)
+{
+	struct percpu_counter *cnt;
+
+	if (op_is_discard(op))
+		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_DISCARD];
+	else if (op_is_write(op))
+		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_WRITE];
+	else
+		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_READ];
+
+	percpu_counter_add_batch(cnt, val, BLKG_STAT_CPU_BATCH);
+
+	if (op_is_sync(op))
+		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_SYNC];
+	else
+		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_ASYNC];
+
+	percpu_counter_add_batch(cnt, val, BLKG_STAT_CPU_BATCH);
+}
+
+/**
+ * blkg_rwstat_read - read the current values of a blkg_rwstat
+ * @rwstat: blkg_rwstat to read
+ *
+ * Read the current snapshot of @rwstat and return it in the aux counts.
+ */
+static inline void blkg_rwstat_read(struct blkg_rwstat *rwstat,
+		struct blkg_rwstat_sample *result)
+{
+	int i;
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++)
+		result->cnt[i] =
+			percpu_counter_sum_positive(&rwstat->cpu_cnt[i]);
+}
+
+/**
+ * blkg_rwstat_total - read the total count of a blkg_rwstat
+ * @rwstat: blkg_rwstat to read
+ *
+ * Return the total count of @rwstat regardless of the IO direction.  This
+ * function can be called without synchronization and takes care of u64
+ * atomicity.
+ */
+static inline uint64_t blkg_rwstat_total(struct blkg_rwstat *rwstat)
+{
+	struct blkg_rwstat_sample tmp = { };
+
+	blkg_rwstat_read(rwstat, &tmp);
+	return tmp.cnt[BLKG_RWSTAT_READ] + tmp.cnt[BLKG_RWSTAT_WRITE];
+}
+
+/**
+ * blkg_rwstat_reset - reset a blkg_rwstat
+ * @rwstat: blkg_rwstat to reset
+ */
+static inline void blkg_rwstat_reset(struct blkg_rwstat *rwstat)
+{
+	int i;
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++) {
+		percpu_counter_set(&rwstat->cpu_cnt[i], 0);
+		atomic64_set(&rwstat->aux_cnt[i], 0);
+	}
+}
+
+/**
+ * blkg_rwstat_add_aux - add a blkg_rwstat into another's aux count
+ * @to: the destination blkg_rwstat
+ * @from: the source
+ *
+ * Add @from's count including the aux one to @to's aux count.
+ */
+static inline void blkg_rwstat_add_aux(struct blkg_rwstat *to,
+				       struct blkg_rwstat *from)
+{
+	u64 sum[BLKG_RWSTAT_NR];
+	int i;
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++)
+		sum[i] = percpu_counter_sum_positive(&from->cpu_cnt[i]);
+
+	for (i = 0; i < BLKG_RWSTAT_NR; i++)
+		atomic64_add(sum[i] + atomic64_read(&from->aux_cnt[i]),
+			     &to->aux_cnt[i]);
+}
+#endif	/* _BLK_CGROUP_RWSTAT_H */
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 0b349f6a6a69..d0b9bfa563eb 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -561,103 +561,6 @@ u64 __blkg_prfill_u64(struct seq_file *sf, struct blkg_policy_data *pd, u64 v)
 }
 EXPORT_SYMBOL_GPL(__blkg_prfill_u64);
 
-/**
- * __blkg_prfill_rwstat - prfill helper for a blkg_rwstat
- * @sf: seq_file to print to
- * @pd: policy private data of interest
- * @rwstat: rwstat to print
- *
- * Print @rwstat to @sf for the device assocaited with @pd.
- */
-u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
-			 const struct blkg_rwstat_sample *rwstat)
-{
-	static const char *rwstr[] = {
-		[BLKG_RWSTAT_READ]	= "Read",
-		[BLKG_RWSTAT_WRITE]	= "Write",
-		[BLKG_RWSTAT_SYNC]	= "Sync",
-		[BLKG_RWSTAT_ASYNC]	= "Async",
-		[BLKG_RWSTAT_DISCARD]	= "Discard",
-	};
-	const char *dname = blkg_dev_name(pd->blkg);
-	u64 v;
-	int i;
-
-	if (!dname)
-		return 0;
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++)
-		seq_printf(sf, "%s %s %llu\n", dname, rwstr[i],
-			   rwstat->cnt[i]);
-
-	v = rwstat->cnt[BLKG_RWSTAT_READ] +
-		rwstat->cnt[BLKG_RWSTAT_WRITE] +
-		rwstat->cnt[BLKG_RWSTAT_DISCARD];
-	seq_printf(sf, "%s Total %llu\n", dname, v);
-	return v;
-}
-EXPORT_SYMBOL_GPL(__blkg_prfill_rwstat);
-
-/**
- * blkg_prfill_rwstat - prfill callback for blkg_rwstat
- * @sf: seq_file to print to
- * @pd: policy private data of interest
- * @off: offset to the blkg_rwstat in @pd
- *
- * prfill callback for printing a blkg_rwstat.
- */
-u64 blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
-		       int off)
-{
-	struct blkg_rwstat_sample rwstat = { };
-
-	blkg_rwstat_read((void *)pd + off, &rwstat);
-	return __blkg_prfill_rwstat(sf, pd, &rwstat);
-}
-EXPORT_SYMBOL_GPL(blkg_prfill_rwstat);
-
-/**
- * blkg_rwstat_recursive_sum - collect hierarchical blkg_rwstat
- * @blkg: blkg of interest
- * @pol: blkcg_policy which contains the blkg_rwstat
- * @off: offset to the blkg_rwstat in blkg_policy_data or @blkg
- * @sum: blkg_rwstat_sample structure containing the results
- *
- * Collect the blkg_rwstat specified by @blkg, @pol and @off and all its
- * online descendants and their aux counts.  The caller must be holding the
- * queue lock for online tests.
- *
- * If @pol is NULL, blkg_rwstat is at @off bytes into @blkg; otherwise, it
- * is at @off bytes into @blkg's blkg_policy_data of the policy.
- */
-void blkg_rwstat_recursive_sum(struct blkcg_gq *blkg, struct blkcg_policy *pol,
-		int off, struct blkg_rwstat_sample *sum)
-{
-	struct blkcg_gq *pos_blkg;
-	struct cgroup_subsys_state *pos_css;
-	unsigned int i;
-
-	lockdep_assert_held(&blkg->q->queue_lock);
-
-	rcu_read_lock();
-	blkg_for_each_descendant_pre(pos_blkg, pos_css, blkg) {
-		struct blkg_rwstat *rwstat;
-
-		if (!pos_blkg->online)
-			continue;
-
-		if (pol)
-			rwstat = (void *)blkg_to_pd(pos_blkg, pol) + off;
-		else
-			rwstat = (void *)pos_blkg + off;
-
-		for (i = 0; i < BLKG_RWSTAT_NR; i++)
-			sum->cnt[i] = blkg_rwstat_read_counter(rwstat, i);
-	}
-	rcu_read_unlock();
-}
-EXPORT_SYMBOL_GPL(blkg_rwstat_recursive_sum);
-
 /* Performs queue bypass and policy enabled checks then looks up blkg. */
 static struct blkcg_gq *blkg_lookup_check(struct blkcg *blkcg,
 					  const struct blkcg_policy *pol,
diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 6ac5b5d56170..ae1f73a92e12 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -12,6 +12,7 @@
 #include <linux/blktrace_api.h>
 #include <linux/blk-cgroup.h>
 #include "blk.h"
+#include "blk-cgroup-rwstat.h"
 
 /* Max dispatch from a group in 1 round */
 static int throtl_grp_quantum = 8;
diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h
index 867ab391e409..48a66738143d 100644
--- a/include/linux/blk-cgroup.h
+++ b/include/linux/blk-cgroup.h
@@ -41,17 +41,6 @@ enum blkg_iostat_type {
 	BLKG_IOSTAT_NR,
 };
 
-enum blkg_rwstat_type {
-	BLKG_RWSTAT_READ,
-	BLKG_RWSTAT_WRITE,
-	BLKG_RWSTAT_SYNC,
-	BLKG_RWSTAT_ASYNC,
-	BLKG_RWSTAT_DISCARD,
-
-	BLKG_RWSTAT_NR,
-	BLKG_RWSTAT_TOTAL = BLKG_RWSTAT_NR,
-};
-
 struct blkcg_gq;
 
 struct blkcg {
@@ -82,19 +71,6 @@ struct blkg_iostat_set {
 	struct blkg_iostat		last;
 };
 
-/*
- * blkg_[rw]stat->aux_cnt is excluded for local stats but included for
- * recursive.  Used to carry stats of dead children.
- */
-struct blkg_rwstat {
-	struct percpu_counter		cpu_cnt[BLKG_RWSTAT_NR];
-	atomic64_t			aux_cnt[BLKG_RWSTAT_NR];
-};
-
-struct blkg_rwstat_sample {
-	u64				cnt[BLKG_RWSTAT_NR];
-};
-
 /*
  * A blkcg_gq (blkg) is association between a block cgroup (blkcg) and a
  * request_queue (q).  This is used by blkcg policies which need to track
@@ -223,13 +199,6 @@ int blkcg_activate_policy(struct request_queue *q,
 void blkcg_deactivate_policy(struct request_queue *q,
 			     const struct blkcg_policy *pol);
 
-static inline u64 blkg_rwstat_read_counter(struct blkg_rwstat *rwstat,
-		unsigned int idx)
-{
-	return atomic64_read(&rwstat->aux_cnt[idx]) +
-		percpu_counter_sum_positive(&rwstat->cpu_cnt[idx]);
-}
-
 const char *blkg_dev_name(struct blkcg_gq *blkg);
 void blkcg_print_blkgs(struct seq_file *sf, struct blkcg *blkcg,
 		       u64 (*prfill)(struct seq_file *,
@@ -237,12 +206,6 @@ void blkcg_print_blkgs(struct seq_file *sf, struct blkcg *blkcg,
 		       const struct blkcg_policy *pol, int data,
 		       bool show_total);
 u64 __blkg_prfill_u64(struct seq_file *sf, struct blkg_policy_data *pd, u64 v);
-u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
-			 const struct blkg_rwstat_sample *rwstat);
-u64 blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd,
-		       int off);
-void blkg_rwstat_recursive_sum(struct blkcg_gq *blkg, struct blkcg_policy *pol,
-		int off, struct blkg_rwstat_sample *sum);
 
 struct blkg_conf_ctx {
 	struct gendisk			*disk;
@@ -594,128 +557,6 @@ static inline void blkg_put(struct blkcg_gq *blkg)
 		if (((d_blkg) = __blkg_lookup(css_to_blkcg(pos_css),	\
 					      (p_blkg)->q, false)))
 
-static inline int blkg_rwstat_init(struct blkg_rwstat *rwstat, gfp_t gfp)
-{
-	int i, ret;
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++) {
-		ret = percpu_counter_init(&rwstat->cpu_cnt[i], 0, gfp);
-		if (ret) {
-			while (--i >= 0)
-				percpu_counter_destroy(&rwstat->cpu_cnt[i]);
-			return ret;
-		}
-		atomic64_set(&rwstat->aux_cnt[i], 0);
-	}
-	return 0;
-}
-
-static inline void blkg_rwstat_exit(struct blkg_rwstat *rwstat)
-{
-	int i;
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++)
-		percpu_counter_destroy(&rwstat->cpu_cnt[i]);
-}
-
-/**
- * blkg_rwstat_add - add a value to a blkg_rwstat
- * @rwstat: target blkg_rwstat
- * @op: REQ_OP and flags
- * @val: value to add
- *
- * Add @val to @rwstat.  The counters are chosen according to @rw.  The
- * caller is responsible for synchronizing calls to this function.
- */
-static inline void blkg_rwstat_add(struct blkg_rwstat *rwstat,
-				   unsigned int op, uint64_t val)
-{
-	struct percpu_counter *cnt;
-
-	if (op_is_discard(op))
-		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_DISCARD];
-	else if (op_is_write(op))
-		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_WRITE];
-	else
-		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_READ];
-
-	percpu_counter_add_batch(cnt, val, BLKG_STAT_CPU_BATCH);
-
-	if (op_is_sync(op))
-		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_SYNC];
-	else
-		cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_ASYNC];
-
-	percpu_counter_add_batch(cnt, val, BLKG_STAT_CPU_BATCH);
-}
-
-/**
- * blkg_rwstat_read - read the current values of a blkg_rwstat
- * @rwstat: blkg_rwstat to read
- *
- * Read the current snapshot of @rwstat and return it in the aux counts.
- */
-static inline void blkg_rwstat_read(struct blkg_rwstat *rwstat,
-		struct blkg_rwstat_sample *result)
-{
-	int i;
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++)
-		result->cnt[i] =
-			percpu_counter_sum_positive(&rwstat->cpu_cnt[i]);
-}
-
-/**
- * blkg_rwstat_total - read the total count of a blkg_rwstat
- * @rwstat: blkg_rwstat to read
- *
- * Return the total count of @rwstat regardless of the IO direction.  This
- * function can be called without synchronization and takes care of u64
- * atomicity.
- */
-static inline uint64_t blkg_rwstat_total(struct blkg_rwstat *rwstat)
-{
-	struct blkg_rwstat_sample tmp = { };
-
-	blkg_rwstat_read(rwstat, &tmp);
-	return tmp.cnt[BLKG_RWSTAT_READ] + tmp.cnt[BLKG_RWSTAT_WRITE];
-}
-
-/**
- * blkg_rwstat_reset - reset a blkg_rwstat
- * @rwstat: blkg_rwstat to reset
- */
-static inline void blkg_rwstat_reset(struct blkg_rwstat *rwstat)
-{
-	int i;
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++) {
-		percpu_counter_set(&rwstat->cpu_cnt[i], 0);
-		atomic64_set(&rwstat->aux_cnt[i], 0);
-	}
-}
-
-/**
- * blkg_rwstat_add_aux - add a blkg_rwstat into another's aux count
- * @to: the destination blkg_rwstat
- * @from: the source
- *
- * Add @from's count including the aux one to @to's aux count.
- */
-static inline void blkg_rwstat_add_aux(struct blkg_rwstat *to,
-				       struct blkg_rwstat *from)
-{
-	u64 sum[BLKG_RWSTAT_NR];
-	int i;
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++)
-		sum[i] = percpu_counter_sum_positive(&from->cpu_cnt[i]);
-
-	for (i = 0; i < BLKG_RWSTAT_NR; i++)
-		atomic64_add(sum[i] + atomic64_read(&from->aux_cnt[i]),
-			     &to->aux_cnt[i]);
-}
-
 #ifdef CONFIG_BLK_DEV_THROTTLING
 extern bool blk_throtl_bio(struct request_queue *q, struct blkcg_gq *blkg,
 			   struct bio *bio);
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH 0.5/5] bfq-iosched: relocate bfqg_*rwstat*() helpers
  2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
                   ` (4 preceding siblings ...)
  2019-11-06 21:58 ` [PATCH 5/5] blk-cgroup: separate out blkg_rwstat under CONFIG_BLK_CGROUP_RWSTAT Tejun Heo
@ 2019-11-07 19:04 ` Tejun Heo
  2019-11-07 19:08   ` Jens Axboe
  5 siblings, 1 reply; 9+ messages in thread
From: Tejun Heo @ 2019-11-07 19:04 UTC (permalink / raw)
  To: axboe; +Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team

From a7b9100d72d75a1472bed130b6edd271b734776f Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 7 Nov 2019 11:01:19 -0800

Collect them right under #ifdef CONFIG_BFQ_CGROUP_DEBUG.  The next
patch will use them from !DEBUG path and this makes it easy to move
them out of the ifdef block.

This is pure code reorganization.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
Jens, this is an extra patch to help fixing the build bug when
!CONFIG_BFQ_CGROUP_DEBUG.  The git tree has been updated accordingly.
Please let me know if you want the whole series reposted.

Thanks.

 block/bfq-cgroup.c | 46 +++++++++++++++++++++++-----------------------
 1 file changed, 23 insertions(+), 23 deletions(-)

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index 86a607cf19a1..d4755d4ad009 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -1058,17 +1058,34 @@ static ssize_t bfq_io_set_weight(struct kernfs_open_file *of,
 }
 
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
-static int bfqg_print_stat(struct seq_file *sf, void *v)
+static int bfqg_print_rwstat(struct seq_file *sf, void *v)
 {
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), blkg_prfill_stat,
-			  &blkcg_policy_bfq, seq_cft(sf)->private, false);
+	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), blkg_prfill_rwstat,
+			  &blkcg_policy_bfq, seq_cft(sf)->private, true);
 	return 0;
 }
 
-static int bfqg_print_rwstat(struct seq_file *sf, void *v)
+static u64 bfqg_prfill_rwstat_recursive(struct seq_file *sf,
+					struct blkg_policy_data *pd, int off)
 {
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), blkg_prfill_rwstat,
-			  &blkcg_policy_bfq, seq_cft(sf)->private, true);
+	struct blkg_rwstat_sample sum;
+
+	blkg_rwstat_recursive_sum(pd_to_blkg(pd), &blkcg_policy_bfq, off, &sum);
+	return __blkg_prfill_rwstat(sf, pd, &sum);
+}
+
+static int bfqg_print_rwstat_recursive(struct seq_file *sf, void *v)
+{
+	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
+			  bfqg_prfill_rwstat_recursive, &blkcg_policy_bfq,
+			  seq_cft(sf)->private, true);
+	return 0;
+}
+
+static int bfqg_print_stat(struct seq_file *sf, void *v)
+{
+	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), blkg_prfill_stat,
+			  &blkcg_policy_bfq, seq_cft(sf)->private, false);
 	return 0;
 }
 
@@ -1097,15 +1114,6 @@ static u64 bfqg_prfill_stat_recursive(struct seq_file *sf,
 	return __blkg_prfill_u64(sf, pd, sum);
 }
 
-static u64 bfqg_prfill_rwstat_recursive(struct seq_file *sf,
-					struct blkg_policy_data *pd, int off)
-{
-	struct blkg_rwstat_sample sum;
-
-	blkg_rwstat_recursive_sum(pd_to_blkg(pd), &blkcg_policy_bfq, off, &sum);
-	return __blkg_prfill_rwstat(sf, pd, &sum);
-}
-
 static int bfqg_print_stat_recursive(struct seq_file *sf, void *v)
 {
 	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
@@ -1114,14 +1122,6 @@ static int bfqg_print_stat_recursive(struct seq_file *sf, void *v)
 	return 0;
 }
 
-static int bfqg_print_rwstat_recursive(struct seq_file *sf, void *v)
-{
-	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)),
-			  bfqg_prfill_rwstat_recursive, &blkcg_policy_bfq,
-			  seq_cft(sf)->private, true);
-	return 0;
-}
-
 static u64 bfqg_prfill_sectors(struct seq_file *sf, struct blkg_policy_data *pd,
 			       int off)
 {
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 9+ messages in thread

* [PATCH v2 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios
  2019-11-06 21:58 ` [PATCH 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios Tejun Heo
@ 2019-11-07 19:05   ` Tejun Heo
  0 siblings, 0 replies; 9+ messages in thread
From: Tejun Heo @ 2019-11-07 19:05 UTC (permalink / raw)
  To: axboe
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team,
	Paolo Valente

From 4f711dd784f8666ec41fa5319f16838235557a30 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 7 Nov 2019 11:01:19 -0800

When used on cgroup1, bfq uses the blkg->stat_bytes and ->stat_ios
from blk-cgroup core to populate six stat knobs.  blk-cgroup core is
moving away from blkg_rwstat to improve scalability and won't be able
to support this usage.

It isn't like the sharing gains all that much.  Let's break it out to
dedicated rwstat counters which are updated when on cgroup1.  This
makes use of bfqg_*rwstat*() helpers outside of
CONFIG_BFQ_CGROUP_DEBUG.  Move them out.

v2: Compile fix when !CONFIG_BFQ_CGROUP_DEBUG.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Paolo Valente <paolo.valente@linaro.org>
---
 block/bfq-cgroup.c  | 39 +++++++++++++++++++++++++++------------
 block/bfq-iosched.c |  4 ++++
 block/bfq-iosched.h |  4 ++++
 3 files changed, 35 insertions(+), 12 deletions(-)

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index d4755d4ad009..cea0ae12f937 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -347,6 +347,14 @@ void bfqg_and_blkg_put(struct bfq_group *bfqg)
 	bfqg_put(bfqg);
 }
 
+void bfqg_stats_update_legacy_io(struct request_queue *q, struct request *rq)
+{
+	struct bfq_group *bfqg = blkg_to_bfqg(rq->bio->bi_blkg);
+
+	blkg_rwstat_add(&bfqg->stats.bytes, rq->cmd_flags, blk_rq_bytes(rq));
+	blkg_rwstat_add(&bfqg->stats.ios, rq->cmd_flags, 1);
+}
+
 /* @stats = 0 */
 static void bfqg_stats_reset(struct bfqg_stats *stats)
 {
@@ -431,6 +439,8 @@ void bfq_init_entity(struct bfq_entity *entity, struct bfq_group *bfqg)
 
 static void bfqg_stats_exit(struct bfqg_stats *stats)
 {
+	blkg_rwstat_exit(&stats->bytes);
+	blkg_rwstat_exit(&stats->ios);
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	blkg_rwstat_exit(&stats->merged);
 	blkg_rwstat_exit(&stats->service_time);
@@ -448,6 +458,10 @@ static void bfqg_stats_exit(struct bfqg_stats *stats)
 
 static int bfqg_stats_init(struct bfqg_stats *stats, gfp_t gfp)
 {
+	if (blkg_rwstat_init(&stats->bytes, gfp) ||
+	    blkg_rwstat_init(&stats->ios, gfp))
+		return -ENOMEM;
+
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	if (blkg_rwstat_init(&stats->merged, gfp) ||
 	    blkg_rwstat_init(&stats->service_time, gfp) ||
@@ -1057,7 +1071,6 @@ static ssize_t bfq_io_set_weight(struct kernfs_open_file *of,
 	return bfq_io_set_device_weight(of, buf, nbytes, off);
 }
 
-#ifdef CONFIG_BFQ_CGROUP_DEBUG
 static int bfqg_print_rwstat(struct seq_file *sf, void *v)
 {
 	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), blkg_prfill_rwstat,
@@ -1082,6 +1095,7 @@ static int bfqg_print_rwstat_recursive(struct seq_file *sf, void *v)
 	return 0;
 }
 
+#ifdef CONFIG_BFQ_CGROUP_DEBUG
 static int bfqg_print_stat(struct seq_file *sf, void *v)
 {
 	blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), blkg_prfill_stat,
@@ -1125,7 +1139,8 @@ static int bfqg_print_stat_recursive(struct seq_file *sf, void *v)
 static u64 bfqg_prfill_sectors(struct seq_file *sf, struct blkg_policy_data *pd,
 			       int off)
 {
-	u64 sum = blkg_rwstat_total(&pd->blkg->stat_bytes);
+	struct bfq_group *bfqg = blkg_to_bfqg(pd->blkg);
+	u64 sum = blkg_rwstat_total(&bfqg->stats.bytes);
 
 	return __blkg_prfill_u64(sf, pd, sum >> 9);
 }
@@ -1142,8 +1157,8 @@ static u64 bfqg_prfill_sectors_recursive(struct seq_file *sf,
 {
 	struct blkg_rwstat_sample tmp;
 
-	blkg_rwstat_recursive_sum(pd->blkg, NULL,
-			offsetof(struct blkcg_gq, stat_bytes), &tmp);
+	blkg_rwstat_recursive_sum(pd->blkg, &blkcg_policy_bfq,
+			offsetof(struct bfq_group, stats.bytes), &tmp);
 
 	return __blkg_prfill_u64(sf, pd,
 		(tmp.cnt[BLKG_RWSTAT_READ] + tmp.cnt[BLKG_RWSTAT_WRITE]) >> 9);
@@ -1226,13 +1241,13 @@ struct cftype bfq_blkcg_legacy_files[] = {
 	/* statistics, covers only the tasks in the bfqg */
 	{
 		.name = "bfq.io_service_bytes",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_bytes,
+		.private = offsetof(struct bfq_group, stats.bytes),
+		.seq_show = bfqg_print_rwstat,
 	},
 	{
 		.name = "bfq.io_serviced",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_ios,
+		.private = offsetof(struct bfq_group, stats.ios),
+		.seq_show = bfqg_print_rwstat,
 	},
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	{
@@ -1269,13 +1284,13 @@ struct cftype bfq_blkcg_legacy_files[] = {
 	/* the same statistics which cover the bfqg and its descendants */
 	{
 		.name = "bfq.io_service_bytes_recursive",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_bytes_recursive,
+		.private = offsetof(struct bfq_group, stats.bytes),
+		.seq_show = bfqg_print_rwstat_recursive,
 	},
 	{
 		.name = "bfq.io_serviced_recursive",
-		.private = (unsigned long)&blkcg_policy_bfq,
-		.seq_show = blkg_print_stat_ios_recursive,
+		.private = offsetof(struct bfq_group, stats.ios),
+		.seq_show = bfqg_print_rwstat_recursive,
 	},
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	{
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index b33be928d164..4629f5580d46 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -5450,6 +5450,10 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 	bool idle_timer_disabled = false;
 	unsigned int cmd_flags;
 
+#ifdef CONFIG_BFQ_GROUP_IOSCHED
+	if (!cgroup_subsys_on_dfl(io_cgrp_subsys) && rq->bio)
+		bfqg_stats_update_legacy_io(q, rq);
+#endif
 	spin_lock_irq(&bfqd->lock);
 	if (blk_mq_sched_try_insert_merge(q, rq)) {
 		spin_unlock_irq(&bfqd->lock);
diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h
index 5d1a519640f6..2676c06218f1 100644
--- a/block/bfq-iosched.h
+++ b/block/bfq-iosched.h
@@ -809,6 +809,9 @@ struct bfq_stat {
 };
 
 struct bfqg_stats {
+	/* basic stats */
+	struct blkg_rwstat		bytes;
+	struct blkg_rwstat		ios;
 #ifdef CONFIG_BFQ_CGROUP_DEBUG
 	/* number of ios merged */
 	struct blkg_rwstat		merged;
@@ -956,6 +959,7 @@ void bfq_put_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg);
 
 /* ---------------- cgroups-support interface ---------------- */
 
+void bfqg_stats_update_legacy_io(struct request_queue *q, struct request *rq);
 void bfqg_stats_update_io_add(struct bfq_group *bfqg, struct bfq_queue *bfqq,
 			      unsigned int op);
 void bfqg_stats_update_io_remove(struct bfq_group *bfqg, unsigned int op);
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 9+ messages in thread

* Re: [PATCH 0.5/5] bfq-iosched: relocate bfqg_*rwstat*() helpers
  2019-11-07 19:04 ` [PATCH 0.5/5] bfq-iosched: relocate bfqg_*rwstat*() helpers Tejun Heo
@ 2019-11-07 19:08   ` Jens Axboe
  0 siblings, 0 replies; 9+ messages in thread
From: Jens Axboe @ 2019-11-07 19:08 UTC (permalink / raw)
  To: Tejun Heo
  Cc: cgroups, linux-block, linux-kernel, lizefan, hannes, kernel-team

On 11/7/19 12:04 PM, Tejun Heo wrote:
> Jens, this is an extra patch to help fixing the build bug when
> !CONFIG_BFQ_CGROUP_DEBUG.  The git tree has been updated accordingly.
> Please let me know if you want the whole series reposted.

I seem to run into issues applying the series with the 0.5 added first
and with the updated patch 1. So maybe a resend isn't a bad idea :-)

-- 
Jens Axboe


^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2019-11-07 19:08 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-11-06 21:58 [PATCHSET block/for-next] blk-cgroup: use cgroup rstat for IO stats Tejun Heo
2019-11-06 21:58 ` [PATCH 1/5] bfq-iosched: stop using blkg->stat_bytes and ->stat_ios Tejun Heo
2019-11-07 19:05   ` [PATCH v2 " Tejun Heo
2019-11-06 21:58 ` [PATCH 2/5] blk-throtl: " Tejun Heo
2019-11-06 21:58 ` [PATCH 3/5] blk-cgroup: remove now unused blkg_print_stat_{bytes|ios}_recursive() Tejun Heo
2019-11-06 21:58 ` [PATCH 4/5] blk-cgroup: reimplement basic IO stats using cgroup rstat Tejun Heo
2019-11-06 21:58 ` [PATCH 5/5] blk-cgroup: separate out blkg_rwstat under CONFIG_BLK_CGROUP_RWSTAT Tejun Heo
2019-11-07 19:04 ` [PATCH 0.5/5] bfq-iosched: relocate bfqg_*rwstat*() helpers Tejun Heo
2019-11-07 19:08   ` Jens Axboe

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).