linux-stable.git/drivers/md, branch v5.4.166

md: fix a lock order reversal in md_alloc

2021-09-30T08:09:24+00:00

[ Upstream commit 7df835a32a8bedf7ce88efcfa7c9b245b52ff139 ]

Commit b0140891a8cea3 ("md: Fix race when creating a new md device.")
not only moved assigning mddev->gendisk before calling add_disk, which
fixes the races described in the commit log, but also added a
mddev->open_mutex critical section over add_disk and creation of the
md kobj.  Adding a kobject after add_disk is racy vs deleting the gendisk
right after adding it, but md already prevents against that by holding
a mddev->active reference.

On the other hand taking this lock added a lock order reversal with what
is not disk->open_mutex (used to be bdev->bd_mutex when the commit was
added) for partition devices, which need that lock for the internal open
for the partition scan, and a recent commit also takes it for
non-partitioned devices, leading to further lockdep splatter.

Fixes: b0140891a8ce ("md: Fix race when creating a new md device.")
Fixes: d62633873590 ("block: support delayed holder registration")
Reported-by: syzbot+fadc0aaf497e6a493b9f@syzkaller.appspotmail.com
Signed-off-by: Christoph Hellwig 
Tested-by: syzbot+fadc0aaf497e6a493b9f@syzkaller.appspotmail.com
Reviewed-by: NeilBrown 
Signed-off-by: Song Liu 
Signed-off-by: Sasha Levin

dm crypt: Avoid percpu_counter spinlock contention in crypt_page_alloc()

2021-09-22T10:26:21+00:00

commit 528b16bfc3ae5f11638e71b3b63a81f9999df727 upstream.

On systems with many cores using dm-crypt, heavy spinlock contention in
percpu_counter_compare() can be observed when the page allocation limit
for a given device is reached or close to be reached. This is due
to percpu_counter_compare() taking a spinlock to compute an exact
result on potentially many CPUs at the same time.

Switch to non-exact comparison of allocated and allowed pages by using
the value returned by percpu_counter_read_positive() to avoid taking
the percpu_counter spinlock.

This may over/under estimate the actual number of allocated pages by at
most (batch-1) * num_online_cpus().

Currently, batch is bounded by 32. The system on which this issue was
first observed has 256 CPUs and 512GB of RAM. With a 4k page size, this
change may over/under estimate by 31MB. With ~10G (2%) allowed dm-crypt
allocations, this seems an acceptable error. Certainly preferred over
running into the spinlock contention.

This behavior was reproduced on an EC2 c5.24xlarge instance with 96 CPUs
and 192GB RAM as follows, but can be provoked on systems with less CPUs
as well.

 * Disable swap
 * Tune vm settings to promote regular writeback
     $ echo 50 > /proc/sys/vm/dirty_expire_centisecs
     $ echo 25 > /proc/sys/vm/dirty_writeback_centisecs
     $ echo $((128 * 1024 * 1024)) > /proc/sys/vm/dirty_background_bytes

 * Create 8 dmcrypt devices based on files on a tmpfs
 * Create and mount an ext4 filesystem on each crypt devices
 * Run stress-ng --hdd 8 within one of above filesystems

Total %system usage collected from sysstat goes to ~35%. Write throughput
on the underlying loop device is ~2GB/s. perf profiling an individual
kworker kcryptd thread shows the following profile, indicating spinlock
contention in percpu_counter_compare():

    99.98%     0.00%  kworker/u193:46  [kernel.kallsyms]  [k] ret_from_fork
      |
      --ret_from_fork
        kthread
        worker_thread
        |
         --99.92%--process_one_work
            |
            |--80.52%--kcryptd_crypt
            |    |
            |    |--62.58%--mempool_alloc
            |    |  |
            |    |   --62.24%--crypt_page_alloc
            |    |     |
            |    |      --61.51%--__percpu_counter_compare
            |    |        |
            |    |         --61.34%--__percpu_counter_sum
            |    |           |
            |    |           |--58.68%--_raw_spin_lock_irqsave
            |    |           |  |
            |    |           |   --58.30%--native_queued_spin_lock_slowpath
            |    |           |
            |    |            --0.69%--cpumask_next
            |    |                |
            |    |                 --0.51%--_find_next_bit
            |    |
            |    |--10.61%--crypt_convert
            |    |          |
            |    |          |--6.05%--xts_crypt
            ...

After applying this patch and running the same test, %system usage is
lowered to ~7% and write throughput on the loop device increases
to ~2.7GB/s. perf report shows mempool_alloc() as ~8% rather than ~62%
in the profile and not hitting the percpu_counter() spinlock anymore.

    |--8.15%--mempool_alloc
    |    |
    |    |--3.93%--crypt_page_alloc
    |    |    |
    |    |     --3.75%--__alloc_pages
    |    |         |
    |    |          --3.62%--get_page_from_freelist
    |    |              |
    |    |               --3.22%--rmqueue_bulk
    |    |                   |
    |    |                    --2.59%--_raw_spin_lock
    |    |                      |
    |    |                       --2.57%--native_queued_spin_lock_slowpath
    |    |
    |     --3.05%--_raw_spin_lock_irqsave
    |               |
    |                --2.49%--native_queued_spin_lock_slowpath

Suggested-by: DJ Gregor 
Reviewed-by: Mikulas Patocka 
Signed-off-by: Arne Welzel 
Fixes: 5059353df86e ("dm crypt: limit the number of allocated pages")
Cc: stable@vger.kernel.org
Signed-off-by: Mike Snitzer 
Signed-off-by: Greg Kroah-Hartman

bcache: add proper error unwinding in bcache_device_init

2021-09-15T07:47:27+00:00

[ Upstream commit 224b0683228c5f332f9cee615d85e75e9a347170 ]

Except for the IDA none of the allocations in bcache_device_init is
unwound on error, fix that.

Signed-off-by: Christoph Hellwig 
Acked-by: Coly Li 
Link: https://lore.kernel.org/r/20210809064028.1198327-7-hch@lst.de
Signed-off-by: Jens Axboe 
Signed-off-by: Sasha Levin

md/raid10: properly indicate failure when ending a failed write request

2021-08-12T11:21:03+00:00

commit 5ba03936c05584b6f6f79be5ebe7e5036c1dd252 upstream.

Similar to [1], this patch fixes the same bug in raid10. Also cleanup the
comments.

[1] commit 2417b9869b81 ("md/raid1: properly indicate failure when ending
                         a failed write request")
Cc: stable@vger.kernel.org
Fixes: 7cee6d4e6035 ("md/raid10: end bio when the device faulty")
Signed-off-by: Wei Shuyu 
Acked-by: Guoqing Jiang 
Signed-off-by: Song Liu 
Signed-off-by: Greg Kroah-Hartman

dm writecache: return the exact table values that were set

2021-07-25T12:35:14+00:00

commit 054bee16163df023e2589db09fd27d81f7ad9e72 upstream.

LVM doesn't like it when the target returns different values from what
was set in the constructor. Fix dm-writecache so that the returned
table values are exactly the same as requested values.

Signed-off-by: Mikulas Patocka 
Cc: stable@vger.kernel.org # v4.18+
Signed-off-by: Mike Snitzer 
Signed-off-by: Greg Kroah-Hartman

dm btree remove: assign new_root only when removal succeeds

2021-07-19T06:53:17+00:00

commit b6e58b5466b2959f83034bead2e2e1395cca8aeb upstream.

remove_raw() in dm_btree_remove() may fail due to IO read error
(e.g. read the content of origin block fails during shadowing),
and the value of shadow_spine::root is uninitialized, but
the uninitialized value is still assign to new_root in the
end of dm_btree_remove().

For dm-thin, the value of pmd->details_root or pmd->root will become
an uninitialized value, so if trying to read details_info tree again
out-of-bound memory may occur as showed below:

  general protection fault, probably for non-canonical address 0x3fdcb14c8d7520
  CPU: 4 PID: 515 Comm: dmsetup Not tainted 5.13.0-rc6
  Hardware name: QEMU Standard PC
  RIP: 0010:metadata_ll_load_ie+0x14/0x30
  Call Trace:
   sm_metadata_count_is_more_than_one+0xb9/0xe0
   dm_tm_shadow_block+0x52/0x1c0
   shadow_step+0x59/0xf0
   remove_raw+0xb2/0x170
   dm_btree_remove+0xf4/0x1c0
   dm_pool_delete_thin_device+0xc3/0x140
   pool_message+0x218/0x2b0
   target_message+0x251/0x290
   ctl_ioctl+0x1c4/0x4d0
   dm_ctl_ioctl+0xe/0x20
   __x64_sys_ioctl+0x7b/0xb0
   do_syscall_64+0x40/0xb0
   entry_SYSCALL_64_after_hwframe+0x44/0xae

Fixing it by only assign new_root when removal succeeds

Signed-off-by: Hou Tao 
Cc: stable@vger.kernel.org
Signed-off-by: Mike Snitzer 
Signed-off-by: Greg Kroah-Hartman

dm space maps: don't reset space map allocation cursor when committing

2021-07-19T06:53:09+00:00

[ Upstream commit 5faafc77f7de69147d1e818026b9a0cbf036a7b2 ]

Current commit code resets the place where the search for free blocks
will begin back to the start of the metadata device.  There are a couple
of repercussions to this:

- The first allocation after the commit is likely to take longer than
  normal as it searches for a free block in an area that is likely to
  have very few free blocks (if any).

- Any free blocks it finds will have been recently freed.  Reusing them
  means we have fewer old copies of the metadata to aid recovery from
  hardware error.

Fix these issues by leaving the cursor alone, only resetting when the
search hits the end of the metadata device.

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer 
Signed-off-by: Sasha Levin

dm verity: fix require_signatures module_param permissions

2021-06-16T09:59:37+00:00

[ Upstream commit 0c1f3193b1cdd21e7182f97dc9bca7d284d18a15 ]

The third parameter of module_param() is permissions for the sysfs node
but it looks like it is being used as the initial value of the parameter
here.  In fact, false here equates to omitting the file from sysfs and
does not affect the value of require_signatures.

Making the parameter writable is not simple because going from
false->true is fine but it should not be possible to remove the
requirement to verify a signature.  But it can be useful to inspect the
value of this parameter from userspace, so change the permissions to
make a read-only file in sysfs.

Signed-off-by: John Keeping 
Signed-off-by: Mike Snitzer 
Signed-off-by: Sasha Levin

dm snapshot: properly fix a crash when an origin has no snapshots

2021-06-03T06:59:02+00:00

commit 7e768532b2396bcb7fbf6f82384b85c0f1d2f197 upstream.

If an origin target has no snapshots, o->split_boundary is set to 0.
This causes BUG_ON(sectors <= 0) in block/bio.c:bio_split().

Fix this by initializing chunk_size, and in turn split_boundary, to
rounddown_pow_of_two(UINT_MAX) -- the largest power of two that fits
into "unsigned" type.

Signed-off-by: Mikulas Patocka 
Cc: stable@vger.kernel.org
Signed-off-by: Mike Snitzer 
Signed-off-by: Greg Kroah-Hartman

dm snapshot: fix crash with transient storage and zero chunk size

2021-05-26T10:05:18+00:00

commit c699a0db2d62e3bbb7f0bf35c87edbc8d23e3062 upstream.

The following commands will crash the kernel:

modprobe brd rd_size=1048576
dmsetup create o --table "0 `blockdev --getsize /dev/ram0` snapshot-origin /dev/ram0"
dmsetup create s --table "0 `blockdev --getsize /dev/ram0` snapshot /dev/ram0 /dev/ram1 N 0"

The reason is that when we test for zero chunk size, we jump to the label
bad_read_metadata without setting the "r" variable. The function
snapshot_ctr destroys all the structures and then exits with "r == 0". The
kernel then crashes because it falsely believes that snapshot_ctr
succeeded.

In order to fix the bug, we set the variable "r" to -EINVAL.

Signed-off-by: Mikulas Patocka 
Cc: stable@vger.kernel.org
Signed-off-by: Mike Snitzer 
Signed-off-by: Greg Kroah-Hartman