linux-stable.git/block, branch linux-5.14.y

blk-cgroup: synchronize blkg creation against policy deactivation

2021-11-17T10:04:15+00:00

[ Upstream commit 0c9d338c8443b06da8e8d3bfce824c5ea6d3488f ]

Our test reports a null pointer dereference:

[  168.534653] ==================================================================
[  168.535614] Disabling lock debugging due to kernel taint
[  168.536346] BUG: kernel NULL pointer dereference, address: 0000000000000008
[  168.537274] #PF: supervisor read access in kernel mode
[  168.537964] #PF: error_code(0x0000) - not-present page
[  168.538667] PGD 0 P4D 0
[  168.539025] Oops: 0000 [#1] PREEMPT SMP KASAN
[  168.539656] CPU: 13 PID: 759 Comm: bash Tainted: G    B             5.15.0-rc2-next-202100
[  168.540954] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20190727_0738364
[  168.542736] RIP: 0010:bfq_pd_init+0x88/0x1e0
[  168.543318] Code: 98 00 00 00 e8 c9 e4 5b ff 4c 8b 65 00 49 8d 7c 24 08 e8 bb e4 5b ff 4d0
[  168.545803] RSP: 0018:ffff88817095f9c0 EFLAGS: 00010002
[  168.546497] RAX: 0000000000000001 RBX: ffff888101a1c000 RCX: 0000000000000000
[  168.547438] RDX: 0000000000000003 RSI: 0000000000000002 RDI: ffff888106553428
[  168.548402] RBP: ffff888106553400 R08: ffffffff961bcaf4 R09: 0000000000000001
[  168.549365] R10: ffffffffa2e16c27 R11: fffffbfff45c2d84 R12: 0000000000000000
[  168.550291] R13: ffff888101a1c098 R14: ffff88810c7a08c8 R15: ffffffffa55541a0
[  168.551221] FS:  00007fac75227700(0000) GS:ffff88839ba80000(0000) knlGS:0000000000000000
[  168.552278] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  168.553040] CR2: 0000000000000008 CR3: 0000000165ce7000 CR4: 00000000000006e0
[  168.554000] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  168.554929] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  168.555888] Call Trace:
[  168.556221]  
[  168.556510]  blkg_create+0x1c0/0x8c0
[  168.556989]  blkg_conf_prep+0x574/0x650
[  168.557502]  ? stack_trace_save+0x99/0xd0
[  168.558033]  ? blkcg_conf_open_bdev+0x1b0/0x1b0
[  168.558629]  tg_set_conf.constprop.0+0xb9/0x280
[  168.559231]  ? kasan_set_track+0x29/0x40
[  168.559758]  ? kasan_set_free_info+0x30/0x60
[  168.560344]  ? tg_set_limit+0xae0/0xae0
[  168.560853]  ? do_sys_openat2+0x33b/0x640
[  168.561383]  ? do_sys_open+0xa2/0x100
[  168.561877]  ? __x64_sys_open+0x4e/0x60
[  168.562383]  ? __kasan_check_write+0x20/0x30
[  168.562951]  ? copyin+0x48/0x70
[  168.563390]  ? _copy_from_iter+0x234/0x9e0
[  168.563948]  tg_set_conf_u64+0x17/0x20
[  168.564467]  cgroup_file_write+0x1ad/0x380
[  168.565014]  ? cgroup_file_poll+0x80/0x80
[  168.565568]  ? __mutex_lock_slowpath+0x30/0x30
[  168.566165]  ? pgd_free+0x100/0x160
[  168.566649]  kernfs_fop_write_iter+0x21d/0x340
[  168.567246]  ? cgroup_file_poll+0x80/0x80
[  168.567796]  new_sync_write+0x29f/0x3c0
[  168.568314]  ? new_sync_read+0x410/0x410
[  168.568840]  ? __handle_mm_fault+0x1c97/0x2d80
[  168.569425]  ? copy_page_range+0x2b10/0x2b10
[  168.570007]  ? _raw_read_lock_bh+0xa0/0xa0
[  168.570622]  vfs_write+0x46e/0x630
[  168.571091]  ksys_write+0xcd/0x1e0
[  168.571563]  ? __x64_sys_read+0x60/0x60
[  168.572081]  ? __kasan_check_write+0x20/0x30
[  168.572659]  ? do_user_addr_fault+0x446/0xff0
[  168.573264]  __x64_sys_write+0x46/0x60
[  168.573774]  do_syscall_64+0x35/0x80
[  168.574264]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[  168.574960] RIP: 0033:0x7fac74915130
[  168.575456] Code: 73 01 c3 48 8b 0d 58 ed 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 0f 1f 444
[  168.577969] RSP: 002b:00007ffc3080e288 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
[  168.578986] RAX: ffffffffffffffda RBX: 0000000000000009 RCX: 00007fac74915130
[  168.579937] RDX: 0000000000000009 RSI: 000056007669f080 RDI: 0000000000000001
[  168.580884] RBP: 000056007669f080 R08: 000000000000000a R09: 00007fac75227700
[  168.581841] R10: 000056007655c8f0 R11: 0000000000000246 R12: 0000000000000009
[  168.582796] R13: 0000000000000001 R14: 00007fac74be55e0 R15: 00007fac74be08c0
[  168.583757]  
[  168.584063] Modules linked in:
[  168.584494] CR2: 0000000000000008
[  168.584964] ---[ end trace 2475611ad0f77a1a ]---

This is because blkg_alloc() is called from blkg_conf_prep() without
holding 'q->queue_lock', and elevator is exited before blkg_create():

thread 1                            thread 2
blkg_conf_prep
 spin_lock_irq(&q->queue_lock);
 blkg_lookup_check -> return NULL
 spin_unlock_irq(&q->queue_lock);

 blkg_alloc
  blkcg_policy_enabled -> true
  pd = ->pd_alloc_fn
  blkg->pd[i] = pd
                                   blk_mq_exit_sched
                                    bfq_exit_queue
                                     blkcg_deactivate_policy
                                      spin_lock_irq(&q->queue_lock);
                                      __clear_bit(pol->plid, q->blkcg_pols);
                                      spin_unlock_irq(&q->queue_lock);
                                    q->elevator = NULL;
  spin_lock_irq(&q->queue_lock);
   blkg_create
    if (blkg->pd[i])
     ->pd_init_fn -> q->elevator is NULL
  spin_unlock_irq(&q->queue_lock);

Because blkcg_deactivate_policy() requires queue to be frozen, we can
grab q_usage_counter to synchoronize blkg_conf_prep() against
blkcg_deactivate_policy().

Fixes: e21b7a0b9887 ("block, bfq: add full hierarchical scheduling and cgroups support")
Signed-off-by: Yu Kuai 
Acked-by: Tejun Heo 
Link: https://lore.kernel.org/r/20211020014036.2141723-1-yukuai3@huawei.com
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

block: remove inaccurate requeue check

2021-11-17T10:03:52+00:00

[ Upstream commit 037057a5a979c7eeb2ee5d12cf4c24b805192c75 ]

This check is meant to catch cases where a requeue is attempted on a
request that is still inserted. It's never really been useful to catch any
misuse, and now it's actively wrong. Outside of that, this should not be a
BUG_ON() to begin with.

Remove the check as it's now causing active harm, as requeue off the plug
path will trigger it even though the request state is just fine.

Reported-by: Yi Zhang 
Link: https://lore.kernel.org/linux-block/CAHj4cs80zAUc2grnCZ015-2Rvd-=gXRfB_dFKy=RTm+wRo09HQ@mail.gmail.com/
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

block: bump max plugged deferred size from 16 to 32

2021-11-17T10:03:51+00:00

[ Upstream commit ba0ffdd8ce48ad7f7e85191cd29f9674caca3745 ]

Particularly for NVMe with efficient deferred submission for many
requests, there are nice benefits to be seen by bumping the default max
plug count from 16 to 32. This is especially true for virtualized setups,
where the submit part is more expensive. But can be noticed even on
native hardware.

Reduce the multiple queue factor from 4 to 2, since we're changing the
default size.

While changing it, move the defines into the block layer private header.
These aren't values that anyone outside of the block layer uses, or
should use.

Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

block: schedule queue restart after BLK_STS_ZONE_RESOURCE

2021-11-17T10:03:31+00:00

[ Upstream commit 9586e67b911c95ba158fcc247b230e9c2d718623 ]

When dispatching a zone append write request to a SCSI zoned block device,
if the target zone of the request is already locked, the device driver will
return BLK_STS_ZONE_RESOURCE and the request will be pushed back to the
hctx dipatch queue. The queue will be marked as RESTART in
dd_finish_request() and restarted in __blk_mq_free_request(). However, this
restart applies to the hctx of the completed request. If the requeued
request is on a different hctx, dispatch will no be retried until another
request is submitted or the next periodic queue run triggers, leading to up
to 30 seconds latency for the requeued request.

Fix this problem by scheduling a queue restart similarly to the
BLK_STS_RESOURCE case or when we cannot get the budget.

Also, consolidate the checks into the "need_resource" variable to simplify
the condition.

Signed-off-by: Naohiro Aota 
Reviewed-by: Christoph Hellwig 
Cc: Niklas Cassel 
Link: https://lore.kernel.org/r/20211026165127.4151055-1-naohiro.aota@wdc.com
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

block: Fix partition check for host-aware zoned block devices

2021-11-02T18:50:52+00:00

commit e0c60d0102a5ad3475401e1a2faa3d3623eefce4 upstream.

Commit a33df75c6328 ("block: use an xarray for disk->part_tbl") modified
the method to check partition existence in host-aware zoned block
devices from disk_has_partitions() helper function call to empty check
of xarray disk->part_tbl. However, disk->part_tbl always has single
entry for disk->part0 and never becomes empty. This resulted in the
host-aware zoned devices always judged to have partitions, and it made
the sysfs queue/zoned attribute to be "none" instead of "host-aware"
regardless of partition existence in the devices.

This also caused DEBUG_LOCKS_WARN_ON(lock->magic != lock) for
sdkp->rev_mutex in scsi layer when the kernel detects host-aware zoned
device. Since block layer handled the host-aware zoned devices as non-
zoned devices, scsi layer did not have chance to initialize the mutex
for zone revalidation. Therefore, the warning was triggered.

To fix the issues, call the helper function disk_has_partitions() in
place of disk->part_tbl empty check. Since the function was removed with
the commit a33df75c6328, reimplement it to walk through entries in the
xarray disk->part_tbl.

Fixes: a33df75c6328 ("block: use an xarray for disk->part_tbl")
Signed-off-by: Shin'ichiro Kawasaki 
Cc: stable@vger.kernel.org # v5.14+
Reviewed-by: Johannes Thumshirn 
Reviewed-by: Christoph Hellwig 
Link: https://lore.kernel.org/r/20211026060115.753746-1-shinichiro.kawasaki@wdc.com
Signed-off-by: Jens Axboe 
Signed-off-by: Greg Kroah-Hartman

blk-cgroup: blk_cgroup_bio_start() should use irq-safe operations on blkg->iostat_cpu

2021-10-27T07:59:44+00:00

commit 5370b0f49078203acf3c064b634a09707167a864 upstream.

c3df5fb57fe8 ("cgroup: rstat: fix A-A deadlock on 32bit around
u64_stats_sync") made u64_stats updates irq-safe to avoid A-A deadlocks.
Unfortunately, the conversion missed one in blk_cgroup_bio_start(). Fix it.

Fixes: 2d146aa3aa84 ("mm: memcontrol: switch to rstat")
Cc: stable@vger.kernel.org # v5.13+
Reported-by: syzbot+9738c8815b375ce482a1@syzkaller.appspotmail.com
Signed-off-by: Tejun Heo 
Link: https://lore.kernel.org/r/YWi7NrQdVlxD6J9W@slm.duckdns.org
Signed-off-by: Jens Axboe 
Signed-off-by: Greg Kroah-Hartman

block: decode QUEUE_FLAG_HCTX_ACTIVE in debugfs output

2021-10-27T07:59:31+00:00

[ Upstream commit 1dbdd99b511c966be9147ad72991a2856ac76f22 ]

While debugging an issue we've found that $DEBUGFS/block/$disk/state
doesn't decode QUEUE_FLAG_HCTX_ACTIVE but only displays its numerical
value.

Add QUEUE_FLAG(HCTX_ACTIVE) to the blk_queue_flag_name array so it'll get
decoded properly.

Signed-off-by: Johannes Thumshirn 
Link: https://lore.kernel.org/r/4351076388918075bd80ef07756f9d2ce63be12c.1633332053.git.johannes.thumshirn@wdc.com
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

block/mq-deadline: Move dd_queued() to fix defined but not used warning

2021-10-27T07:59:30+00:00

commit 55a51ea14094a1e7dd0d7f33237d246033dd39ab upstream.

If CONFIG_BLK_DEBUG_FS=n:

    block/mq-deadline.c:274:12: warning: ‘dd_queued’ defined but not used [-Wunused-function]
      274 | static u32 dd_queued(struct deadline_data *dd, enum dd_prio prio)
	  |            ^~~~~~~~~

Fix this by moving dd_queued() just before the sole function that calls
it.

Fixes: 7b05bf771084ff78 ("Revert "block/mq-deadline: Prioritize high-priority requests"")
Signed-off-by: Geert Uytterhoeven 
Fixes: 38ba64d12d4c ("block/mq-deadline: Track I/O statistics")
Reviewed-by: Bart Van Assche 
Link: https://lore.kernel.org/r/20210830091128.1854266-1-geert@linux-m68k.org
Signed-off-by: Jens Axboe 
Signed-off-by: Greg Kroah-Hartman

block: don't call rq_qos_ops->done_bio if the bio isn't tracked

2021-10-09T13:02:41+00:00

[ Upstream commit a647a524a46736786c95cdb553a070322ca096e3 ]

rq_qos framework is only applied on request based driver, so:

1) rq_qos_done_bio() needn't to be called for bio based driver

2) rq_qos_done_bio() needn't to be called for bio which isn't tracked,
such as bios ended from error handling code.

Especially in bio_endio():

1) request queue is referred via bio->bi_bdev->bd_disk->queue, which
may be gone since request queue refcount may not be held in above two
cases

2) q->rq_qos may be freed in blk_cleanup_queue() when calling into
__rq_qos_done_bio()

Fix the potential kernel panic by not calling rq_qos_ops->done_bio if
the bio isn't tracked. This way is safe because both ioc_rqos_done_bio()
and blkcg_iolatency_done_bio() are nop if the bio isn't tracked.

Reported-by: Yu Kuai 
Cc: tj@kernel.org
Signed-off-by: Ming Lei 
Reviewed-by: Christoph Hellwig 
Acked-by: Tejun Heo 
Link: https://lore.kernel.org/r/20210924110704.1541818-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

Revert "block, bfq: honor already-setup queue merges"

2021-10-07T05:53:14+00:00

[ Upstream commit ebc69e897e17373fbe1daaff1debaa77583a5284 ]

This reverts commit 2d52c58b9c9bdae0ca3df6a1eab5745ab3f7d80b.

We have had several folks complain that this causes hangs for them, which
is especially problematic as the commit has also hit stable already.

As no resolution seems to be forthcoming right now, revert the patch.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=214503
Fixes: 2d52c58b9c9b ("block, bfq: honor already-setup queue merges")
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin