This adds a new module parameter to null_blk, blocking. If set, null_blk
will set the BLK_MQ_F_BLOCKING flag, indicating that it sometimes/always
needs to block in its ->queue_rq() function.  The intent is to help find
regressions in blocking drivers, since not many of them exist.

If null_blk is loaded with submit_queues > 1 and blocking=1, this
shows the regression recently fixed by bf4907c0.
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Commit a4d907b6 unified the single and multi queue request handlers,
but in the process, it also screwed up the locking balance and calls
blk_mq_try_issue_directly() with the ctx preempt lock held. This is a
problem for drivers that have set BLK_MQ_F_BLOCKING, since now they
can't reliably sleep.

While in there, protect against similar issues in the future, by adding
a might_sleep() trigger in the BLOCKING path for direct issue or queue
run.
Reported-by: NJosef Bacik <josef@toxicpanda.com>
Tested-by: NJosef Bacik <josef@toxicpanda.com>
Fixes: a4d907b6 ("blk-mq: streamline blk_mq_make_request")
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

trivial fix to spelling mistake in pr_err error message
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

In blk_mq_alloc_request_hctx, blk_mq_sched_get_request doesn't
get sw context so we don't need to put the context with
blk_mq_put_ctx. Unless, we will see preempt counter underflow.

Cc: Omar Sandoval <osandov@fb.com>
Signed-off-by: NMinchan Kim <minchan@kernel.org>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NJens Axboe <axboe@fb.com>

Currently we return true in blk_mq_dispatch_rq_list() if we queued IO
successfully, but we really want to return whether or not the we made
progress. Progress includes if we got an error return.  If we don't,
this can lead to a hang in blk_mq_sched_dispatch_requests() when a
driver is draining IO by returning BLK_MQ_QUEUE_ERROR instead of
manually ending the IO in error and return BLK_MQ_QUEUE_OK.
Tested-by: NJosef Bacik <josef@toxicpanda.com>
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

When try to issue a request directly and we fail we will requeue the
request, but call blk_mq_end_request() as well.  This leads to the
completed request being on a queuelist and getting ended twice, which
causes list corruption in schedulers and other shenanigans.
Signed-off-by: NJosef Bacik <jbacik@fb.com>
Reviewed-by: NMing Lei <tom.leiming@gmail.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NJens Axboe <axboe@fb.com>

blkg_conf_prep() currently calls blkg_lookup_create() while holding
request queue spinlock. This means allocating memory for struct
blkcg_gq has to be made non-blocking. This causes occasional -ENOMEM
failures in call paths like below:

  pcpu_alloc+0x68f/0x710
  __alloc_percpu_gfp+0xd/0x10
  __percpu_counter_init+0x55/0xc0
  cfq_pd_alloc+0x3b2/0x4e0
  blkg_alloc+0x187/0x230
  blkg_create+0x489/0x670
  blkg_lookup_create+0x9a/0x230
  blkg_conf_prep+0x1fb/0x240
  __cfqg_set_weight_device.isra.105+0x5c/0x180
  cfq_set_weight_on_dfl+0x69/0xc0
  cgroup_file_write+0x39/0x1c0
  kernfs_fop_write+0x13f/0x1d0
  __vfs_write+0x23/0x120
  vfs_write+0xc2/0x1f0
  SyS_write+0x44/0xb0
  entry_SYSCALL_64_fastpath+0x18/0xad

In the code path above, percpu allocator cannot call vmalloc() due to
queue spinlock.

A failure in this call path gives grief to tools which are trying to
configure io weights. We see occasional failures happen shortly after
reboots even when system is not under any memory pressure. Machines
with a lot of cpus are more vulnerable to this condition.

Do struct blkcg_gq allocations outside the queue spinlock to allow
blocking during memory allocations.
Suggested-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NTahsin Erdogan <tahsin@google.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <axboe@fb.com>

I inadvertently applied the v5 version of this patch, whereas
the agreed upon version was v5. Revert this one so we can apply
the right one.

This reverts commit 7fc6b87a.

Signed-off-by: NJens Axboe <axboe@fb.com>

Signed-off-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NJens Axboe <axboe@fb.com>

CONFIG_DEBUG_TEST_DRIVER_REMOVE found a possible leak of q->rq_wb when a
request queue is reregistered. This has been a problem since wbt was
introduced, but the WARN_ON(!list_empty(&stats->callbacks)) in the
blk-stat rework exposed it. Fix it by cleaning up wbt when we unregister
the queue.

Fixes: 87760e5e ("block: hook up writeback throttling")
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

blk_alloc_queue_node() already allocates q->stats, so
blk_mq_init_allocated_queue() is overwriting it with a new allocation.

Fixes: a83b576c ("block: fix stacked driver stats init and free")
Reviewed-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Now that the remaining drivers have been converted to one request queue
per gendisk, let's warn if a request queue gets registered more than
once. This will catch future drivers which might do it inadvertently or
any old drivers that I may have missed.
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Before commit 780db207(blk-mq: decouble blk-mq freezing
from generic bypassing), the dying flag is checked before
entering queue, and Tejun converts the checking into .mq_freeze_depth,
and assumes the counter is increased just after dying flag
is set. Unfortunately we doesn't do that in blk_set_queue_dying().

This patch calls blk_freeze_queue_start() in blk_set_queue_dying(),
so that we can block new I/O coming once the queue is set as dying.

Given blk_set_queue_dying() is always called in remove path
of block device, and queue will be cleaned up later, we don't
need to worry about undoing the counter.

Cc: Tejun Heo <tj@kernel.org>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

As the .q_usage_counter is used by both legacy and
mq path, we need to block new I/O if queue becomes
dead in blk_queue_enter().

So rename it and we can use this function in both
paths.
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

Without the barrier, reading DEAD flag of .q_usage_counter
and reading .mq_freeze_depth may be reordered, then the
following wait_event_interruptible() may never return.
Reviewed-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

This patch adds comment on two races related with
timeout handler:

- requeue from queue busy vs. timeout
- rq free & reallocation vs. timeout

Both the races themselves and current solution aren't
explicit enough, so add comments on them.

Cc: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

When iterating busy requests in timeout handler,
if the STARTED flag of one request isn't set, that means
the request is being processed in block layer or driver, and
isn't submitted to hardware yet.

In current implementation of blk_mq_check_expired(),
if the request queue becomes dying, un-started requests are
handled as being completed/freed immediately. This way is
wrong, and can cause rq corruption or double allocation[1][2],
when doing I/O and removing&resetting NVMe device at the sametime.

This patch fixes several issues reported by Yi Zhang.

[1]. oops log 1
[  581.789754] ------------[ cut here ]------------
[  581.789758] kernel BUG at block/blk-mq.c:374!
[  581.789760] invalid opcode: 0000 [#1] SMP
[  581.789761] Modules linked in: vfat fat ipmi_ssif intel_rapl sb_edac
edac_core x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm nvme
irqbypass crct10dif_pclmul nvme_core crc32_pclmul ghash_clmulni_intel
intel_cstate ipmi_si mei_me ipmi_devintf intel_uncore sg ipmi_msghandler
intel_rapl_perf iTCO_wdt mei iTCO_vendor_support mxm_wmi lpc_ich dcdbas shpchp
pcspkr acpi_power_meter wmi nfsd auth_rpcgss nfs_acl lockd dm_multipath grace
sunrpc ip_tables xfs libcrc32c sd_mod mgag200 i2c_algo_bit drm_kms_helper
syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm ahci libahci
crc32c_intel tg3 libata megaraid_sas i2c_core ptp fjes pps_core dm_mirror
dm_region_hash dm_log dm_mod
[  581.789796] CPU: 1 PID: 1617 Comm: kworker/1:1H Not tainted 4.10.0.bz1420297+ #4
[  581.789797] Hardware name: Dell Inc. PowerEdge R730xd/072T6D, BIOS 2.2.5 09/06/2016
[  581.789804] Workqueue: kblockd blk_mq_timeout_work
[  581.789806] task: ffff8804721c8000 task.stack: ffffc90006ee4000
[  581.789809] RIP: 0010:blk_mq_end_request+0x58/0x70
[  581.789810] RSP: 0018:ffffc90006ee7d50 EFLAGS: 00010202
[  581.789811] RAX: 0000000000000001 RBX: ffff8802e4195340 RCX: ffff88028e2f4b88
[  581.789812] RDX: 0000000000001000 RSI: 0000000000001000 RDI: 0000000000000000
[  581.789813] RBP: ffffc90006ee7d60 R08: 0000000000000003 R09: ffff88028e2f4b00
[  581.789814] R10: 0000000000001000 R11: 0000000000000001 R12: 00000000fffffffb
[  581.789815] R13: ffff88042abe5780 R14: 000000000000002d R15: ffff88046fbdff80
[  581.789817] FS:  0000000000000000(0000) GS:ffff88047fc00000(0000) knlGS:0000000000000000
[  581.789818] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  581.789819] CR2: 00007f64f403a008 CR3: 000000014d078000 CR4: 00000000001406e0
[  581.789820] Call Trace:
[  581.789825]  blk_mq_check_expired+0x76/0x80
[  581.789828]  bt_iter+0x45/0x50
[  581.789830]  blk_mq_queue_tag_busy_iter+0xdd/0x1f0
[  581.789832]  ? blk_mq_rq_timed_out+0x70/0x70
[  581.789833]  ? blk_mq_rq_timed_out+0x70/0x70
[  581.789840]  ? __switch_to+0x140/0x450
[  581.789841]  blk_mq_timeout_work+0x88/0x170
[  581.789845]  process_one_work+0x165/0x410
[  581.789847]  worker_thread+0x137/0x4c0
[  581.789851]  kthread+0x101/0x140
[  581.789853]  ? rescuer_thread+0x3b0/0x3b0
[  581.789855]  ? kthread_park+0x90/0x90
[  581.789860]  ret_from_fork+0x2c/0x40
[  581.789861] Code: 48 85 c0 74 0d 44 89 e6 48 89 df ff d0 5b 41 5c 5d c3 48
8b bb 70 01 00 00 48 85 ff 75 0f 48 89 df e8 7d f0 ff ff 5b 41 5c 5d c3 <0f>
0b e8 71 f0 ff ff 90 eb e9 0f 1f 40 00 66 2e 0f 1f 84 00 00
[  581.789882] RIP: blk_mq_end_request+0x58/0x70 RSP: ffffc90006ee7d50
[  581.789889] ---[ end trace bcaf03d9a14a0a70 ]---

[2]. oops log2
[ 6984.857362] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
[ 6984.857372] IP: nvme_queue_rq+0x6e6/0x8cd [nvme]
[ 6984.857373] PGD 0
[ 6984.857374]
[ 6984.857376] Oops: 0000 [#1] SMP
[ 6984.857379] Modules linked in: ipmi_ssif vfat fat intel_rapl sb_edac
edac_core x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm
irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel ipmi_si iTCO_wdt
iTCO_vendor_support mxm_wmi ipmi_devintf intel_cstate sg dcdbas intel_uncore
mei_me intel_rapl_perf mei pcspkr lpc_ich ipmi_msghandler shpchp
acpi_power_meter wmi nfsd auth_rpcgss dm_multipath nfs_acl lockd grace sunrpc
ip_tables xfs libcrc32c sd_mod mgag200 i2c_algo_bit drm_kms_helper syscopyarea
sysfillrect crc32c_intel sysimgblt fb_sys_fops ttm nvme drm nvme_core ahci
libahci i2c_core tg3 libata ptp megaraid_sas pps_core fjes dm_mirror
dm_region_hash dm_log dm_mod
[ 6984.857416] CPU: 7 PID: 1635 Comm: kworker/7:1H Not tainted
4.10.0-2.el7.bz1420297.x86_64 #1
[ 6984.857417] Hardware name: Dell Inc. PowerEdge R730xd/072T6D, BIOS 2.2.5 09/06/2016
[ 6984.857427] Workqueue: kblockd blk_mq_run_work_fn
[ 6984.857429] task: ffff880476e3da00 task.stack: ffffc90002e90000
[ 6984.857432] RIP: 0010:nvme_queue_rq+0x6e6/0x8cd [nvme]
[ 6984.857433] RSP: 0018:ffffc90002e93c50 EFLAGS: 00010246
[ 6984.857434] RAX: 0000000000000000 RBX: ffff880275646600 RCX: 0000000000001000
[ 6984.857435] RDX: 0000000000000fff RSI: 00000002fba2a000 RDI: ffff8804734e6950
[ 6984.857436] RBP: ffffc90002e93d30 R08: 0000000000002000 R09: 0000000000001000
[ 6984.857437] R10: 0000000000001000 R11: 0000000000000000 R12: ffff8804741d8000
[ 6984.857438] R13: 0000000000000040 R14: ffff880475649f80 R15: ffff8804734e6780
[ 6984.857439] FS:  0000000000000000(0000) GS:ffff88047fcc0000(0000) knlGS:0000000000000000
[ 6984.857440] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 6984.857442] CR2: 0000000000000010 CR3: 0000000001c09000 CR4: 00000000001406e0
[ 6984.857443] Call Trace:
[ 6984.857451]  ? mempool_free+0x2b/0x80
[ 6984.857455]  ? bio_free+0x4e/0x60
[ 6984.857459]  blk_mq_dispatch_rq_list+0xf5/0x230
[ 6984.857462]  blk_mq_process_rq_list+0x133/0x170
[ 6984.857465]  __blk_mq_run_hw_queue+0x8c/0xa0
[ 6984.857467]  blk_mq_run_work_fn+0x12/0x20
[ 6984.857473]  process_one_work+0x165/0x410
[ 6984.857475]  worker_thread+0x137/0x4c0
[ 6984.857478]  kthread+0x101/0x140
[ 6984.857480]  ? rescuer_thread+0x3b0/0x3b0
[ 6984.857481]  ? kthread_park+0x90/0x90
[ 6984.857489]  ret_from_fork+0x2c/0x40
[ 6984.857490] Code: 8b bd 70 ff ff ff 89 95 50 ff ff ff 89 8d 58 ff ff ff 44
89 95 60 ff ff ff e8 b7 dd 12 e1 8b 95 50 ff ff ff 48 89 85 68 ff ff ff <4c>
8b 48 10 44 8b 58 18 8b 8d 58 ff ff ff 44 8b 95 60 ff ff ff
[ 6984.857511] RIP: nvme_queue_rq+0x6e6/0x8cd [nvme] RSP: ffffc90002e93c50
[ 6984.857512] CR2: 0000000000000010
[ 6984.895359] ---[ end trace 2d7ceb528432bf83 ]---

Cc: stable@vger.kernel.org
Reported-by: NYi Zhang <yizhan@redhat.com>
Tested-by: NYi Zhang <yizhan@redhat.com>
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

blkg_conf_prep() currently calls blkg_lookup_create() while holding
request queue spinlock. This means allocating memory for struct
blkcg_gq has to be made non-blocking. This causes occasional -ENOMEM
failures in call paths like below:

  pcpu_alloc+0x68f/0x710
  __alloc_percpu_gfp+0xd/0x10
  __percpu_counter_init+0x55/0xc0
  cfq_pd_alloc+0x3b2/0x4e0
  blkg_alloc+0x187/0x230
  blkg_create+0x489/0x670
  blkg_lookup_create+0x9a/0x230
  blkg_conf_prep+0x1fb/0x240
  __cfqg_set_weight_device.isra.105+0x5c/0x180
  cfq_set_weight_on_dfl+0x69/0xc0
  cgroup_file_write+0x39/0x1c0
  kernfs_fop_write+0x13f/0x1d0
  __vfs_write+0x23/0x120
  vfs_write+0xc2/0x1f0
  SyS_write+0x44/0xb0
  entry_SYSCALL_64_fastpath+0x18/0xad

In the code path above, percpu allocator cannot call vmalloc() due to
queue spinlock.

A failure in this call path gives grief to tools which are trying to
configure io weights. We see occasional failures happen shortly after
reboots even when system is not under any memory pressure. Machines
with a lot of cpus are more vulnerable to this condition.

Update blkg_create() function to temporarily drop the rcu and queue
locks when it is allowed by gfp mask.
Suggested-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NTahsin Erdogan <tahsin@google.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <axboe@fb.com>

Compile-tested only (by hacking it to compile on x86).

Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Compile-tested only (by hacking it to compile on x86).
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Compile-tested only.

Cc: Tim Waugh <tim@cyberelk.net>
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Compile-tested only.

Cc: Tim Waugh <tim@cyberelk.net>
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Compile-tested only.

Cc: Tim Waugh <tim@cyberelk.net>
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Compile-tested only.
Signed-off-by: NOmar Sandoval <osandov@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

One hard problem adding .low limit is to detect idle cgroup. If one
cgroup doesn't dispatch enough IO against its low limit, we must have a
mechanism to determine if other cgroups dispatch more IO. We added the
think time detection mechanism before, but it doesn't work for all
workloads. Here we add a latency based approach.

We already have mechanism to calculate latency threshold for each IO
size. For every IO dispatched from a cgorup, we compare its latency
against its threshold and record the info. If most IO latency is below
threshold (in the code I use 75%), the cgroup could be treated idle and
other cgroups can dispatch more IO.

Currently this latency target check is only for SSD as we can't
calcualte the latency target for hard disk. And this is only for cgroup
leaf node so far.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

User configures latency target, but the latency threshold for each
request size isn't fixed. For a SSD, the IO latency highly depends on
request size. To calculate latency threshold, we sample some data, eg,
average latency for request size 4k, 8k, 16k, 32k .. 1M. The latency
threshold of each request size will be the sample latency (I'll call it
base latency) plus latency target. For example, the base latency for
request size 4k is 80us and user configures latency target 60us. The 4k
latency threshold will be 80 + 60 = 140us.

To sample data, we calculate the order base 2 of rounded up IO sectors.
If the IO size is bigger than 1M, it will be accounted as 1M. Since the
calculation does round up, the base latency will be slightly smaller
than actual value. Also if there isn't any IO dispatched for a specific
IO size, we will use the base latency of smaller IO size for this IO
size.

But we shouldn't sample data at any time. The base latency is supposed
to be latency where disk isn't congested, because we use latency
threshold to schedule IOs between cgroups. If disk is congested, the
latency is higher, using it for scheduling is meaningless. Hence we only
do the sampling when block throttling is in the LOW limit, with
assumption disk isn't congested in such state. If the assumption isn't
true, eg, low limit is too high, calculated latency threshold will be
higher.

Hard disk is completely different. Latency depends on spindle seek
instead of request size. Currently this feature is SSD only, we probably
can use a fixed threshold like 4ms for hard disk though.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Currently there is no way to know the request size when the request is
finished. Next patch will need this info. We could add extra field to
record the size, but blk_issue_stat has enough space to record it, so
this patch just overloads blk_issue_stat. With this, we will have 49bits
to track time, which still is very long time.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Here we introduce per-cgroup latency target. The target determines how a
cgroup can afford latency increasement. We will use the target latency
to calculate a threshold and use it to schedule IO for cgroups. If a
cgroup's bandwidth is below its low limit but its average latency is
below the threshold, other cgroups can safely dispatch more IO even
their bandwidth is higher than their low limits. On the other hand, if
the first cgroup's latency is higher than the threshold, other cgroups
are throttled to their low limits. So the target latency determines how
we efficiently utilize free disk resource without sacifice of worload's
IO latency.

For example, assume 4k IO average latency is 50us when disk isn't
congested. A cgroup sets the target latency to 30us. Then the cgroup can
accept 50+30=80us IO latency. If the cgroupt's average IO latency is
90us and its bandwidth is below low limit, other cgroups are throttled
to their low limit. If the cgroup's average IO latency is 60us, other
cgroups are allowed to dispatch more IO. When other cgroups dispatch
more IO, the first cgroup's IO latency will increase. If it increases to
81us, we then throttle other cgroups.

User will configure the interface in this way:
echo "8:16 rbps=2097152 wbps=max latency=100 idle=200" > io.low

latency is in microsecond unit

By default, latency target is 0, which means to guarantee IO latency.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Last patch introduces a way to detect idle cgroup. We use it to make
upgrade/downgrade decision. And the new algorithm can detect completely
idle cgroup too, so we can delete the corresponding code.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

Add interface to configure the threshold. The io.low interface will
like:
echo "8:16 rbps=2097152 wbps=max idle=2000" > io.low

idle is in microsecond unit.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

A cgroup gets assigned a low limit, but the cgroup could never dispatch
enough IO to cross the low limit. In such case, the queue state machine
will remain in LIMIT_LOW state and all other cgroups will be throttled
according to low limit. This is unfair for other cgroups. We should
treat the cgroup idle and upgrade the state machine to lower state.

We also have a downgrade logic. If the state machine upgrades because of
cgroup idle (real idle), the state machine will downgrade soon as the
cgroup is below its low limit. This isn't what we want. A more
complicated case is cgroup isn't idle when queue is in LIMIT_LOW. But
when queue gets upgraded to lower state, other cgroups could dispatch
more IO and this cgroup can't dispatch enough IO, so the cgroup is below
its low limit and looks like idle (fake idle). In this case, the queue
should downgrade soon. The key to determine if we should do downgrade is
to detect if cgroup is truely idle.

Unfortunately it's very hard to determine if a cgroup is real idle. This
patch uses the 'think time check' idea from CFQ for the purpose. Please
note, the idea doesn't work for all workloads. For example, a workload
with io depth 8 has disk utilization 100%, hence think time is 0, eg,
not idle. But the workload can run higher bandwidth with io depth 16.
Compared to io depth 16, the io depth 8 workload is idle. We use the
idea to roughly determine if a cgroup is idle.

We treat a cgroup idle if its think time is above a threshold (by
default 1ms for SSD and 100ms for HD). The idea is think time above the
threshold will start to harm performance. HD is much slower so a longer
think time is ok.

The patch (and the latter patches) uses 'unsigned long' to track time.
We convert 'ns' to 'us' with 'ns >> 10'. This is fast but loses
precision, should not a big deal.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

When cgroups all reach low limit, cgroups can dispatch more IO. This
could make some cgroups dispatch more IO but others not, and even some
cgroups could dispatch less IO than their low limit. For example, cg1
low limit 10MB/s, cg2 limit 80MB/s, assume disk maximum bandwidth is
120M/s for the workload. Their bps could something like this:

cg1/cg2 bps: T1: 10/80 -> T2: 60/60 -> T3: 10/80

At T1, all cgroups reach low limit, so they can dispatch more IO later.
Then cg1 dispatch more IO and cg2 has no room to dispatch enough IO. At
T2, cg2 only dispatches 60M/s. Since We detect cg2 dispatches less IO
than its low limit 80M/s, we downgrade the queue from LIMIT_MAX to
LIMIT_LOW, then all cgroups are throttled to their low limit (T3). cg2
will have bandwidth below its low limit at most time.

The big problem here is we don't know the maximum bandwidth of the
workload, so we can't make smart decision to avoid the situation. This
patch makes cgroup bandwidth change smooth. After disk upgrades from
LIMIT_LOW to LIMIT_MAX, we don't allow cgroups use all bandwidth upto
their max limit immediately. Their bandwidth limit will be increased
gradually to avoid above situation. So above example will became
something like:

cg1/cg2 bps: 10/80 -> 15/105 -> 20/100 -> 25/95 -> 30/90 -> 35/85 -> 40/80
-> 45/75 -> 22/98

In this way cgroups bandwidth will be above their limit in majority
time, this still doesn't fully utilize disk bandwidth, but that's
something we pay for sharing.

Scale up is linear. The limit scales up 1/2 .low limit every
throtl_slice after upgrade. The scale up will stop if the adjusted limit
hits .max limit. Scale down is exponential. We cut the scale value half
if a cgroup doesn't hit its .low limit. If the scale becomes 0, we then
fully downgrade the queue to LIMIT_LOW state.

Note this doesn't completely avoid cgroup running under its low limit.
The best way to guarantee cgroup doesn't run under its limit is to set
max limit. For example, if we set cg1 max limit to 40, cg2 will never
run under its low limit.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

cgroup could be assigned a limit, but doesn't dispatch enough IO, eg the
cgroup is idle. When this happens, the cgroup doesn't hit its limit, so
we can't move the state machine to higher level and all cgroups will be
throttled to their lower limit, so we waste bandwidth. Detecting idle
cgroup is hard. This patch handles a simple case, a cgroup doesn't
dispatch any IO. We ignore such cgroup's limit, so other cgroups can use
the bandwidth.

Please note this will be replaced with a more sophisticated algorithm
later, but this demonstrates the idea how we handle idle cgroups, so I
leave it here.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

The throtl_slice is 100ms by default. This is a long time for SSD, a lot
of IO can run. To make cgroups have smoother throughput, we choose a
small value (20ms) for SSD.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

throtl_slice is important for blk-throttling. It's called slice
internally but it really is a time window blk-throttling samples data.
blk-throttling will make decision based on the samplings. An example is
bandwidth measurement. A cgroup's bandwidth is measured in the time
interval of throtl_slice.

A small throtl_slice meanse cgroups have smoother throughput but burn
more CPUs. It has 100ms default value, which is not appropriate for all
disks. A fast SSD can dispatch a lot of IOs in 100ms. This patch makes
it tunable.

Since throtl_slice isn't a time slice, the sysfs name
'throttle_sample_time' reflects its character better.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

cgroup could be throttled to a limit but when all cgroups cross high
limit, queue enters a higher state and so the group should be throttled
to a higher limit. It's possible the cgroup is sleeping because of
throttle and other cgroups don't dispatch IO any more. In this case,
nobody can trigger current downgrade/upgrade logic. To fix this issue,
we could either set up a timer to wakeup the cgroup if other cgroups are
idle or make sure this cgroup doesn't sleep too long. Setting up a timer
means we must change the timer very frequently. This patch chooses the
latter. Making cgroup sleep time not too big wouldn't change cgroup
bps/iops, but could make it wakeup more frequently, which isn't a big
issue because throtl_slice * 8 is already quite big.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

When queue state machine is in LIMIT_MAX state, but a cgroup is below
its low limit for some time, the queue should be downgraded to lower
state as one cgroup's low limit isn't met.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

When queue is in LIMIT_LOW state and all cgroups with low limit cross
the bps/iops limitation, we will upgrade queue's state to
LIMIT_MAX. To determine if a cgroup exceeds its limitation, we check if
the cgroup has pending request. Since cgroup is throttled according to
the limit, pending request means the cgroup reaches the limit.

If a cgroup has limit set for both read and write, we consider the
combination of them for upgrade. The reason is read IO and write IO can
interfere with each other. If we do the upgrade based in one direction
IO, the other direction IO could be severly harmed.

For a cgroup hierarchy, there are two cases. Children has lower low
limit than parent. Parent's low limit is meaningless. If children's
bps/iops cross low limit, we can upgrade queue state. The other case is
children has higher low limit than parent. Children's low limit is
meaningless. As long as parent's bps/iops (which is a sum of childrens
bps/iops) cross low limit, we can upgrade queue state.
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>

each queue will have a state machine. Initially queue is in LIMIT_LOW
state, which means all cgroups will be throttled according to their low
limit. After all cgroups with low limit cross the limit, the queue state
gets upgraded to LIMIT_MAX state.
For max limit, cgroup will use the limit configured by user.
For low limit, cgroup will use the minimal value between low limit and
max limit configured by user. If the minimal value is 0, which means the
cgroup doesn't configure low limit, we will use max limit to throttle
the cgroup and the cgroup is ready to upgrade to LIMIT_MAX
Signed-off-by: NShaohua Li <shli@fb.com>
Signed-off-by: NJens Axboe <axboe@fb.com>