Commit 7b103151 authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge tag 'for-linus-20190614' of git://git.kernel.dk/linux-block 

Pull block fixes from Jens Axboe:

 - Remove references to old schedulers for the scheduler switching and
   blkio controller documentation (Andreas)

 - Kill duplicate check for report zone for null_blk (Chaitanya)

 - Two bcache fixes (Coly)

 - Ensure that mq-deadline is selected if zoned block device is enabled,
   as we need that to support them (Damien)

 - Fix io_uring memory leak (Eric)

 - ps3vram fallout from LBDAF removal (Geert)

 - Redundant blk-mq debugfs debugfs_create return check cleanup (Greg)

 - Extend NOPLM quirk for ST1000LM024 drives (Hans)

 - Remove error path warning that can now trigger after the queue
   removal/addition fixes (Ming)

* tag 'for-linus-20190614' of git://git.kernel.dk/linux-block:
  block/ps3vram: Use %llu to format sector_t after LBDAF removal
  libata: Extend quirks for the ST1000LM024 drives with NOLPM quirk
  bcache: only set BCACHE_DEV_WB_RUNNING when cached device attached
  bcache: fix stack corruption by PRECEDING_KEY()
  blk-mq: remove WARN_ON(!q->elevator) from blk_mq_sched_free_requests
  blkio-controller.txt: Remove references to CFQ
  block/switching-sched.txt: Update to blk-mq schedulers
  null_blk: remove duplicate check for report zone
  blk-mq: no need to check return value of debugfs_create functions
  io_uring: fix memory leak of UNIX domain socket inode
  block: force select mq-deadline for zoned block devices
parents 5dcedf46 1d0c0651

Documentation/block/switching-sched.txt

+8 −10
Original line number Diff line number Diff line
@@ -13,11 +13,9 @@ you can do so by typing: 


# mount none /sys -t sysfs



As of the Linux 2.6.10 kernel, it is now possible to change the

IO scheduler for a given block device on the fly (thus making it possible,

for instance, to set the CFQ scheduler for the system default, but

set a specific device to use the deadline or noop schedulers - which

can improve that device's throughput).

It is possible to change the IO scheduler for a given block device on

the fly to select one of mq-deadline, none, bfq, or kyber schedulers -

which can improve that device's throughput.



To set a specific scheduler, simply do this:
@@ -30,8 +28,8 @@ The list of defined schedulers can be found by simply doing 
a "cat /sys/block/DEV/queue/scheduler" - the list of valid names

will be displayed, with the currently selected scheduler in brackets:



# cat /sys/block/hda/queue/scheduler

noop deadline [cfq]

# echo deadline > /sys/block/hda/queue/scheduler

# cat /sys/block/hda/queue/scheduler

noop [deadline] cfq
 
# cat /sys/block/sda/queue/scheduler

[mq-deadline] kyber bfq none

# echo none >/sys/block/sda/queue/scheduler

# cat /sys/block/sda/queue/scheduler

[none] mq-deadline kyber bfq

Documentation/cgroup-v1/blkio-controller.txt

+7 −89
Original line number Diff line number Diff line
@@ -8,61 +8,13 @@ both at leaf nodes as well as at intermediate nodes in a storage hierarchy. 
Plan is to use the same cgroup based management interface for blkio controller

and based on user options switch IO policies in the background.



Currently two IO control policies are implemented. First one is proportional

weight time based division of disk policy. It is implemented in CFQ. Hence

this policy takes effect only on leaf nodes when CFQ is being used. The second

one is throttling policy which can be used to specify upper IO rate limits

on devices. This policy is implemented in generic block layer and can be

used on leaf nodes as well as higher level logical devices like device mapper.

One IO control policy is throttling policy which can be used to

specify upper IO rate limits on devices. This policy is implemented in

generic block layer and can be used on leaf nodes as well as higher

level logical devices like device mapper.



HOWTO

=====

Proportional Weight division of bandwidth

-----------------------------------------

You can do a very simple testing of running two dd threads in two different

cgroups. Here is what you can do.



- Enable Block IO controller

	CONFIG_BLK_CGROUP=y



- Enable group scheduling in CFQ

	CONFIG_CFQ_GROUP_IOSCHED=y



- Compile and boot into kernel and mount IO controller (blkio); see

  cgroups.txt, Why are cgroups needed?.



	mount -t tmpfs cgroup_root /sys/fs/cgroup

	mkdir /sys/fs/cgroup/blkio

	mount -t cgroup -o blkio none /sys/fs/cgroup/blkio



- Create two cgroups

	mkdir -p /sys/fs/cgroup/blkio/test1/ /sys/fs/cgroup/blkio/test2



- Set weights of group test1 and test2

	echo 1000 > /sys/fs/cgroup/blkio/test1/blkio.weight

	echo 500 > /sys/fs/cgroup/blkio/test2/blkio.weight



- Create two same size files (say 512MB each) on same disk (file1, file2) and

  launch two dd threads in different cgroup to read those files.



	sync

	echo 3 > /proc/sys/vm/drop_caches



	dd if=/mnt/sdb/zerofile1 of=/dev/null &

	echo $! > /sys/fs/cgroup/blkio/test1/tasks

	cat /sys/fs/cgroup/blkio/test1/tasks



	dd if=/mnt/sdb/zerofile2 of=/dev/null &

	echo $! > /sys/fs/cgroup/blkio/test2/tasks

	cat /sys/fs/cgroup/blkio/test2/tasks



- At macro level, first dd should finish first. To get more precise data, keep

  on looking at (with the help of script), at blkio.disk_time and

  blkio.disk_sectors files of both test1 and test2 groups. This will tell how

  much disk time (in milliseconds), each group got and how many sectors each

  group dispatched to the disk. We provide fairness in terms of disk time, so

  ideally io.disk_time of cgroups should be in proportion to the weight.



Throttling/Upper Limit policy

-----------------------------

- Enable Block IO controller
@@ -94,7 +46,7 @@ Throttling/Upper Limit policy 
Hierarchical Cgroups

====================



Both CFQ and throttling implement hierarchy support; however,

Throttling implements hierarchy support; however,

throttling's hierarchy support is enabled iff "sane_behavior" is

enabled from cgroup side, which currently is a development option and

not publicly available.
@@ -107,9 +59,8 @@ If somebody created a hierarchy like as follows. 
			|

		     test3



CFQ by default and throttling with "sane_behavior" will handle the

hierarchy correctly.  For details on CFQ hierarchy support, refer to

Documentation/block/cfq-iosched.txt.  For throttling, all limits apply

Throttling with "sane_behavior" will handle the

hierarchy correctly. For throttling, all limits apply

to the whole subtree while all statistics are local to the IOs

directly generated by tasks in that cgroup.
@@ -130,10 +81,6 @@ CONFIG_DEBUG_BLK_CGROUP 
	- Debug help. Right now some additional stats file show up in cgroup

	  if this option is enabled.



CONFIG_CFQ_GROUP_IOSCHED

	- Enables group scheduling in CFQ. Currently only 1 level of group

	  creation is allowed.



CONFIG_BLK_DEV_THROTTLING

	- Enable block device throttling support in block layer.
@@ -344,32 +291,3 @@ Common files among various policies 
- blkio.reset_stats

	- Writing an int to this file will result in resetting all the stats

	  for that cgroup.
 


CFQ sysfs tunable

=================

/sys/block/<disk>/queue/iosched/slice_idle

------------------------------------------

On a faster hardware CFQ can be slow, especially with sequential workload.

This happens because CFQ idles on a single queue and single queue might not

drive deeper request queue depths to keep the storage busy. In such scenarios

one can try setting slice_idle=0 and that would switch CFQ to IOPS

(IO operations per second) mode on NCQ supporting hardware.



That means CFQ will not idle between cfq queues of a cfq group and hence be

able to driver higher queue depth and achieve better throughput. That also

means that cfq provides fairness among groups in terms of IOPS and not in

terms of disk time.



/sys/block/<disk>/queue/iosched/group_idle

------------------------------------------

If one disables idling on individual cfq queues and cfq service trees by

setting slice_idle=0, group_idle kicks in. That means CFQ will still idle

on the group in an attempt to provide fairness among groups.



By default group_idle is same as slice_idle and does not do anything if

slice_idle is enabled.



One can experience an overall throughput drop if you have created multiple

groups and put applications in that group which are not driving enough

IO to keep disk busy. In that case set group_idle=0, and CFQ will not idle

on individual groups and throughput should improve.

block/Kconfig

+1 −0
Original line number Diff line number Diff line
@@ -73,6 +73,7 @@ config BLK_DEV_INTEGRITY 


config BLK_DEV_ZONED

	bool "Zoned block device support"

	select MQ_IOSCHED_DEADLINE

	---help---

	Block layer zoned block device support. This option enables

	support for ZAC/ZBC host-managed and host-aware zoned block devices.

block/blk-mq-debugfs.c

+34 −111
Original line number Diff line number Diff line
@@ -821,38 +821,28 @@ static const struct blk_mq_debugfs_attr blk_mq_debugfs_ctx_attrs[] = { 
	{},

};



static bool debugfs_create_files(struct dentry *parent, void *data,

static void debugfs_create_files(struct dentry *parent, void *data,

				 const struct blk_mq_debugfs_attr *attr)

{

	if (IS_ERR_OR_NULL(parent))

		return false;

		return;



	d_inode(parent)->i_private = data;



	for (; attr->name; attr++) {

		if (!debugfs_create_file(attr->name, attr->mode, parent,

					 (void *)attr, &blk_mq_debugfs_fops))

			return false;

	}

	return true;

	for (; attr->name; attr++)

		debugfs_create_file(attr->name, attr->mode, parent,

				    (void *)attr, &blk_mq_debugfs_fops);

}



int blk_mq_debugfs_register(struct request_queue *q)

void blk_mq_debugfs_register(struct request_queue *q)

{

	struct blk_mq_hw_ctx *hctx;

	int i;



	if (!blk_debugfs_root)

		return -ENOENT;



	q->debugfs_dir = debugfs_create_dir(kobject_name(q->kobj.parent),

					    blk_debugfs_root);

	if (!q->debugfs_dir)

		return -ENOMEM;



	if (!debugfs_create_files(q->debugfs_dir, q,

				  blk_mq_debugfs_queue_attrs))

		goto err;

	debugfs_create_files(q->debugfs_dir, q, blk_mq_debugfs_queue_attrs);



	/*

	 * blk_mq_init_sched() attempted to do this already, but q->debugfs_dir
@@ -864,11 +854,10 @@ int blk_mq_debugfs_register(struct request_queue *q) 


	/* Similarly, blk_mq_init_hctx() couldn't do this previously. */

	queue_for_each_hw_ctx(q, hctx, i) {

		if (!hctx->debugfs_dir && blk_mq_debugfs_register_hctx(q, hctx))

			goto err;

		if (q->elevator && !hctx->sched_debugfs_dir &&

		    blk_mq_debugfs_register_sched_hctx(q, hctx))

			goto err;

		if (!hctx->debugfs_dir)

			blk_mq_debugfs_register_hctx(q, hctx);

		if (q->elevator && !hctx->sched_debugfs_dir)

			blk_mq_debugfs_register_sched_hctx(q, hctx);

	}



	if (q->rq_qos) {
@@ -879,12 +868,6 @@ int blk_mq_debugfs_register(struct request_queue *q) 
			rqos = rqos->next;

		}

	}



	return 0;



err:

	blk_mq_debugfs_unregister(q);

	return -ENOMEM;

}



void blk_mq_debugfs_unregister(struct request_queue *q)
@@ -894,7 +877,7 @@ void blk_mq_debugfs_unregister(struct request_queue *q) 
	q->debugfs_dir = NULL;

}



static int blk_mq_debugfs_register_ctx(struct blk_mq_hw_ctx *hctx,

static void blk_mq_debugfs_register_ctx(struct blk_mq_hw_ctx *hctx,

					struct blk_mq_ctx *ctx)

{

	struct dentry *ctx_dir;
@@ -902,44 +885,24 @@ static int blk_mq_debugfs_register_ctx(struct blk_mq_hw_ctx *hctx, 


	snprintf(name, sizeof(name), "cpu%u", ctx->cpu);

	ctx_dir = debugfs_create_dir(name, hctx->debugfs_dir);

	if (!ctx_dir)

		return -ENOMEM;



	if (!debugfs_create_files(ctx_dir, ctx, blk_mq_debugfs_ctx_attrs))

		return -ENOMEM;



	return 0;

	debugfs_create_files(ctx_dir, ctx, blk_mq_debugfs_ctx_attrs);

}



int blk_mq_debugfs_register_hctx(struct request_queue *q,

void blk_mq_debugfs_register_hctx(struct request_queue *q,

				  struct blk_mq_hw_ctx *hctx)

{

	struct blk_mq_ctx *ctx;

	char name[20];

	int i;



	if (!q->debugfs_dir)

		return -ENOENT;



	snprintf(name, sizeof(name), "hctx%u", hctx->queue_num);

	hctx->debugfs_dir = debugfs_create_dir(name, q->debugfs_dir);

	if (!hctx->debugfs_dir)

		return -ENOMEM;



	if (!debugfs_create_files(hctx->debugfs_dir, hctx,

				  blk_mq_debugfs_hctx_attrs))

		goto err;

	debugfs_create_files(hctx->debugfs_dir, hctx, blk_mq_debugfs_hctx_attrs);



	hctx_for_each_ctx(hctx, ctx, i) {

		if (blk_mq_debugfs_register_ctx(hctx, ctx))

			goto err;

	}



	return 0;



err:

	blk_mq_debugfs_unregister_hctx(hctx);

	return -ENOMEM;

	hctx_for_each_ctx(hctx, ctx, i)

		blk_mq_debugfs_register_ctx(hctx, ctx);

}



void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx)
@@ -949,17 +912,13 @@ void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx) 
	hctx->debugfs_dir = NULL;

}



int blk_mq_debugfs_register_hctxs(struct request_queue *q)

void blk_mq_debugfs_register_hctxs(struct request_queue *q)

{

	struct blk_mq_hw_ctx *hctx;

	int i;



	queue_for_each_hw_ctx(q, hctx, i) {

		if (blk_mq_debugfs_register_hctx(q, hctx))

			return -ENOMEM;

	}



	return 0;

	queue_for_each_hw_ctx(q, hctx, i)

		blk_mq_debugfs_register_hctx(q, hctx);

}



void blk_mq_debugfs_unregister_hctxs(struct request_queue *q)
@@ -971,29 +930,16 @@ void blk_mq_debugfs_unregister_hctxs(struct request_queue *q) 
		blk_mq_debugfs_unregister_hctx(hctx);

}



int blk_mq_debugfs_register_sched(struct request_queue *q)

void blk_mq_debugfs_register_sched(struct request_queue *q)

{

	struct elevator_type *e = q->elevator->type;



	if (!q->debugfs_dir)

		return -ENOENT;



	if (!e->queue_debugfs_attrs)

		return 0;

		return;



	q->sched_debugfs_dir = debugfs_create_dir("sched", q->debugfs_dir);

	if (!q->sched_debugfs_dir)

		return -ENOMEM;



	if (!debugfs_create_files(q->sched_debugfs_dir, q,

				  e->queue_debugfs_attrs))

		goto err;



	return 0;



err:

	blk_mq_debugfs_unregister_sched(q);

	return -ENOMEM;

	debugfs_create_files(q->sched_debugfs_dir, q, e->queue_debugfs_attrs);

}



void blk_mq_debugfs_unregister_sched(struct request_queue *q)
@@ -1008,36 +954,22 @@ void blk_mq_debugfs_unregister_rqos(struct rq_qos *rqos) 
	rqos->debugfs_dir = NULL;

}



int blk_mq_debugfs_register_rqos(struct rq_qos *rqos)

void blk_mq_debugfs_register_rqos(struct rq_qos *rqos)

{

	struct request_queue *q = rqos->q;

	const char *dir_name = rq_qos_id_to_name(rqos->id);



	if (!q->debugfs_dir)

		return -ENOENT;



	if (rqos->debugfs_dir || !rqos->ops->debugfs_attrs)

		return 0;

		return;



	if (!q->rqos_debugfs_dir) {

	if (!q->rqos_debugfs_dir)

		q->rqos_debugfs_dir = debugfs_create_dir("rqos",

							 q->debugfs_dir);

		if (!q->rqos_debugfs_dir)

			return -ENOMEM;

	}



	rqos->debugfs_dir = debugfs_create_dir(dir_name,

					       rqos->q->rqos_debugfs_dir);

	if (!rqos->debugfs_dir)

		return -ENOMEM;



	if (!debugfs_create_files(rqos->debugfs_dir, rqos,

				  rqos->ops->debugfs_attrs))

		goto err;

	return 0;

 err:

	blk_mq_debugfs_unregister_rqos(rqos);

	return -ENOMEM;

	debugfs_create_files(rqos->debugfs_dir, rqos, rqos->ops->debugfs_attrs);

}



void blk_mq_debugfs_unregister_queue_rqos(struct request_queue *q)
@@ -1046,27 +978,18 @@ void blk_mq_debugfs_unregister_queue_rqos(struct request_queue *q) 
	q->rqos_debugfs_dir = NULL;

}



int blk_mq_debugfs_register_sched_hctx(struct request_queue *q,

void blk_mq_debugfs_register_sched_hctx(struct request_queue *q,

					struct blk_mq_hw_ctx *hctx)

{

	struct elevator_type *e = q->elevator->type;



	if (!hctx->debugfs_dir)

		return -ENOENT;



	if (!e->hctx_debugfs_attrs)

		return 0;

		return;



	hctx->sched_debugfs_dir = debugfs_create_dir("sched",

						     hctx->debugfs_dir);

	if (!hctx->sched_debugfs_dir)

		return -ENOMEM;



	if (!debugfs_create_files(hctx->sched_debugfs_dir, hctx,

				  e->hctx_debugfs_attrs))

		return -ENOMEM;



	return 0;

	debugfs_create_files(hctx->sched_debugfs_dir, hctx,

			     e->hctx_debugfs_attrs);

}



void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx)

block/blk-mq-debugfs.h

+15 −21
Original line number Diff line number Diff line
@@ -18,74 +18,68 @@ struct blk_mq_debugfs_attr { 
int __blk_mq_debugfs_rq_show(struct seq_file *m, struct request *rq);

int blk_mq_debugfs_rq_show(struct seq_file *m, void *v);



int blk_mq_debugfs_register(struct request_queue *q);

void blk_mq_debugfs_register(struct request_queue *q);

void blk_mq_debugfs_unregister(struct request_queue *q);

int blk_mq_debugfs_register_hctx(struct request_queue *q,

void blk_mq_debugfs_register_hctx(struct request_queue *q,

				  struct blk_mq_hw_ctx *hctx);

void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx);

int blk_mq_debugfs_register_hctxs(struct request_queue *q);

void blk_mq_debugfs_register_hctxs(struct request_queue *q);

void blk_mq_debugfs_unregister_hctxs(struct request_queue *q);



int blk_mq_debugfs_register_sched(struct request_queue *q);

void blk_mq_debugfs_register_sched(struct request_queue *q);

void blk_mq_debugfs_unregister_sched(struct request_queue *q);

int blk_mq_debugfs_register_sched_hctx(struct request_queue *q,

void blk_mq_debugfs_register_sched_hctx(struct request_queue *q,

				       struct blk_mq_hw_ctx *hctx);

void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx);



int blk_mq_debugfs_register_rqos(struct rq_qos *rqos);

void blk_mq_debugfs_register_rqos(struct rq_qos *rqos);

void blk_mq_debugfs_unregister_rqos(struct rq_qos *rqos);

void blk_mq_debugfs_unregister_queue_rqos(struct request_queue *q);

#else

static inline int blk_mq_debugfs_register(struct request_queue *q)

static inline void blk_mq_debugfs_register(struct request_queue *q)

{

	return 0;

}



static inline void blk_mq_debugfs_unregister(struct request_queue *q)

{

}



static inline int blk_mq_debugfs_register_hctx(struct request_queue *q,

static inline void blk_mq_debugfs_register_hctx(struct request_queue *q,

						struct blk_mq_hw_ctx *hctx)

{

	return 0;

}



static inline void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx)

{

}



static inline int blk_mq_debugfs_register_hctxs(struct request_queue *q)

static inline void blk_mq_debugfs_register_hctxs(struct request_queue *q)

{

	return 0;

}



static inline void blk_mq_debugfs_unregister_hctxs(struct request_queue *q)

{

}



static inline int blk_mq_debugfs_register_sched(struct request_queue *q)

static inline void blk_mq_debugfs_register_sched(struct request_queue *q)

{

	return 0;

}



static inline void blk_mq_debugfs_unregister_sched(struct request_queue *q)

{

}



static inline int blk_mq_debugfs_register_sched_hctx(struct request_queue *q,

static inline void blk_mq_debugfs_register_sched_hctx(struct request_queue *q,

						      struct blk_mq_hw_ctx *hctx)

{

	return 0;

}



static inline void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx)

{

}



static inline int blk_mq_debugfs_register_rqos(struct rq_qos *rqos)

static inline void blk_mq_debugfs_register_rqos(struct rq_qos *rqos)

{

	return 0;

}



static inline void blk_mq_debugfs_unregister_rqos(struct rq_qos *rqos)

CRA Git | Maintained and supported by SUSTech CRA and CCSE