<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/linux.git/drivers/md/dm-bufio.c, branch v7.1</title>
<subtitle>Linux kernel stable tree (mirror)</subtitle>
<id>https://git.radix-linux.su/kernel/linux.git/atom?h=v7.1</id>
<link rel='self' href='https://git.radix-linux.su/kernel/linux.git/atom?h=v7.1'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/'/>
<updated>2026-03-26T17:31:15+00:00</updated>
<entry>
<title>dm-bufio: use kzalloc_flex</title>
<updated>2026-03-26T17:31:15+00:00</updated>
<author>
<name>Rosen Penev</name>
<email>rosenp@gmail.com</email>
</author>
<published>2026-03-26T04:13:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=5387815aa821a8d6dd87ba3aa6869e3c9c709b2a'/>
<id>urn:sha1:5387815aa821a8d6dd87ba3aa6869e3c9c709b2a</id>
<content type='text'>
Avoid manual size calculations and use the proper helper.

Add __counted_by for extra runtime analysis.

Signed-off-by: Rosen Penev &lt;rosenp@gmail.com&gt;
Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm-bufio: avoid redundant buffer_tree lookups</title>
<updated>2026-01-19T14:39:21+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-01-11T21:38:21+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=be9badced98f89cf5c6f7690f7d9739a213c4502'/>
<id>urn:sha1:be9badced98f89cf5c6f7690f7d9739a213c4502</id>
<content type='text'>
dm-bufio's map from block number to buffer is organized as a hash table
of red-black trees.  It does far more lookups in this hash table than
necessary: typically one lookup to lock the tree, one lookup to search
the tree, and one lookup to unlock the tree.  Only one of those lookups
is needed.  Optimize it to do only the minimum number of lookups.

This improves performance.   It also reduces the object code size,
considering that the redundant hash table lookups were being inlined.
For example, the size of the text section of dm-bufio.o decreases from
15599 to 15070 bytes with gcc 15 and x86_64, or from 20652 to 20244
bytes with clang 21 and arm64.

Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm-bufio: merge cache_put() into cache_put_and_wake()</title>
<updated>2026-01-19T14:38:28+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-01-11T21:38:20+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=1bf7ba4ca342ada012e7ef88274fb306e88917ad'/>
<id>urn:sha1:1bf7ba4ca342ada012e7ef88274fb306e88917ad</id>
<content type='text'>
Merge cache_put() into its only caller, cache_put_and_wake().

Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm: add WQ_PERCPU to alloc_workqueue users</title>
<updated>2026-01-14T12:16:00+00:00</updated>
<author>
<name>Marco Crivellari</name>
<email>marco.crivellari@suse.com</email>
</author>
<published>2026-01-13T11:03:02+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=d4880868670198df321627a949e7b7f2d76cf54e'/>
<id>urn:sha1:d4880868670198df321627a949e7b7f2d76cf54e</id>
<content type='text'>
This continues the effort to refactor workqueue APIs, which began with
the introduction of new workqueues and a new alloc_workqueue flag in:

   commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
   commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")

The refactoring is going to alter the default behavior of
alloc_workqueue() to be unbound by default.

With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND),
any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND
must now use WQ_PERCPU. For more details see the Link tag below.

In order to keep alloc_workqueue() behavior identical, explicitly request
WQ_PERCPU.

Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/
Suggested-by: Tejun Heo &lt;tj@kernel.org&gt;
Signed-off-by: Marco Crivellari &lt;marco.crivellari@suse.com&gt;
Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm-bufio: align write boundary on physical block size</title>
<updated>2025-12-10T18:28:22+00:00</updated>
<author>
<name>Mikulas Patocka</name>
<email>mpatocka@redhat.com</email>
</author>
<published>2025-10-20T12:48:13+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=d0ac06ae53be0cdb61f5fe6b62d25d3317c51657'/>
<id>urn:sha1:d0ac06ae53be0cdb61f5fe6b62d25d3317c51657</id>
<content type='text'>
There may be devices with physical block size larger than 4k.

If dm-bufio sends I/O that is not aligned on physical block size,
performance is degraded.

The 4k minimum alignment limit is there because some SSDs report logical
and physical block size 512 despite having 4k internally - so dm-bufio
shouldn't send I/Os not aligned on 4k boundary, because they perform
badly (the SSD does read-modify-write for them).

Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
Reported-by: Uladzislau Rezki (Sony) &lt;urezki@gmail.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>Merge tag 'for-6.18/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm</title>
<updated>2025-10-04T01:48:02+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2025-10-04T01:48:02+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=7dbec0bbc3b468310be172f1ce6ddc9411c84952'/>
<id>urn:sha1:7dbec0bbc3b468310be172f1ce6ddc9411c84952</id>
<content type='text'>
Pull device mapper updates from Mikulas Patocka:

 - a new dm-pcache target for read/write caching on persistent memory

 - fix typos in docs

 - misc small refactoring

 - mark dm-error with DM_TARGET_PASSES_INTEGRITY

 - dm-request-based: fix NULL pointer dereference and quiesce_depth out of sync

 - dm-linear: optimize REQ_PREFLUSH

 - dm-vdo: return error on corrupted metadata

 - dm-integrity: support asynchronous hash interface

* tag 'for-6.18/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (27 commits)
  dm raid: use proper md_ro_state enumerators
  dm-integrity: prefer synchronous hash interface
  dm-integrity: enable asynchronous hash interface
  dm-integrity: rename internal_hash
  dm-integrity: add the "offset" argument
  dm-integrity: allocate the recalculate buffer with kmalloc
  dm-integrity: introduce integrity_kmap and integrity_kunmap
  dm-integrity: replace bvec_kmap_local with kmap_local_page
  dm-integrity: use internal variable for digestsize
  dm vdo: return error on corrupted metadata in start_restoring_volume functions
  dm vdo: Update code to use mem_is_zero
  dm: optimize REQ_PREFLUSH with data when using the linear target
  dm-pcache: use int type to store negative error codes
  dm: fix "writen"-&gt;"written"
  dm-pcache: cleanup: fix coding style report by checkpatch.pl
  dm-pcache: remove ctrl_lock for pcache_cache_segment
  dm: fix NULL pointer dereference in __dm_suspend()
  dm: fix queue start/stop imbalance under suspend/load/resume races
  dm-pcache: add persistent cache target in device-mapper
  dm error: mark as DM_TARGET_PASSES_INTEGRITY
  ...
</content>
</entry>
<entry>
<title>block: add a bio_init_inline helper</title>
<updated>2025-09-09T13:31:59+00:00</updated>
<author>
<name>Christoph Hellwig</name>
<email>hch@lst.de</email>
</author>
<published>2025-09-08T10:56:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=70a6f71b1a77decfc5b1db426ccbe914b58adb38'/>
<id>urn:sha1:70a6f71b1a77decfc5b1db426ccbe914b58adb38</id>
<content type='text'>
Just a simpler wrapper around bio_init for callers that want to
initialize a bio with inline bvecs.

Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: John Garry &lt;john.g.garry@oracle.com&gt;
Reviewed-by: Yu Kuai &lt;yukuai3@huawei.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@kernel.dk&gt;
</content>
</entry>
<entry>
<title>dm bufio: remove redundant __GFP_NOWARN</title>
<updated>2025-08-19T09:12:50+00:00</updated>
<author>
<name>Qianfeng Rong</name>
<email>rongqianfeng@vivo.com</email>
</author>
<published>2025-08-11T12:36:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=31a9403b0d5450aa2595e2fbb8f6091f097906ab'/>
<id>urn:sha1:31a9403b0d5450aa2595e2fbb8f6091f097906ab</id>
<content type='text'>
GFP_NOWAIT already includes __GFP_NOWARN, so let's remove the redundant
__GFP_NOWARN.  Also update comments to clarify the flag semantics.

Signed-off-by: Qianfeng Rong &lt;rongqianfeng@vivo.com&gt;
Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
</content>
</entry>
<entry>
<title>dm-bufio: fix sched in atomic context</title>
<updated>2025-07-10T14:48:50+00:00</updated>
<author>
<name>Sheng Yong</name>
<email>shengyong1@xiaomi.com</email>
</author>
<published>2025-07-10T06:48:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=b1bf1a782fdf5c482215c0c661b5da98b8e75773'/>
<id>urn:sha1:b1bf1a782fdf5c482215c0c661b5da98b8e75773</id>
<content type='text'>
If "try_verify_in_tasklet" is set for dm-verity, DM_BUFIO_CLIENT_NO_SLEEP
is enabled for dm-bufio. However, when bufio tries to evict buffers, there
is a chance to trigger scheduling in spin_lock_bh, the following warning
is hit:

BUG: sleeping function called from invalid context at drivers/md/dm-bufio.c:2745
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 123, name: kworker/2:2
preempt_count: 201, expected: 0
RCU nest depth: 0, expected: 0
4 locks held by kworker/2:2/123:
 #0: ffff88800a2d1548 ((wq_completion)dm_bufio_cache){....}-{0:0}, at: process_one_work+0xe46/0x1970
 #1: ffffc90000d97d20 ((work_completion)(&amp;dm_bufio_replacement_work)){....}-{0:0}, at: process_one_work+0x763/0x1970
 #2: ffffffff8555b528 (dm_bufio_clients_lock){....}-{3:3}, at: do_global_cleanup+0x1ce/0x710
 #3: ffff88801d5820b8 (&amp;c-&gt;spinlock){....}-{2:2}, at: do_global_cleanup+0x2a5/0x710
Preemption disabled at:
[&lt;0000000000000000&gt;] 0x0
CPU: 2 UID: 0 PID: 123 Comm: kworker/2:2 Not tainted 6.16.0-rc3-g90548c634bd0 #305 PREEMPT(voluntary)
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014
Workqueue: dm_bufio_cache do_global_cleanup
Call Trace:
 &lt;TASK&gt;
 dump_stack_lvl+0x53/0x70
 __might_resched+0x360/0x4e0
 do_global_cleanup+0x2f5/0x710
 process_one_work+0x7db/0x1970
 worker_thread+0x518/0xea0
 kthread+0x359/0x690
 ret_from_fork+0xf3/0x1b0
 ret_from_fork_asm+0x1a/0x30
 &lt;/TASK&gt;

That can be reproduced by:

  veritysetup format --data-block-size=4096 --hash-block-size=4096 /dev/vda /dev/vdb
  SIZE=$(blockdev --getsz /dev/vda)
  dmsetup create myverity -r --table "0 $SIZE verity 1 /dev/vda /dev/vdb 4096 4096 &lt;data_blocks&gt; 1 sha256 &lt;root_hash&gt; &lt;salt&gt; 1 try_verify_in_tasklet"
  mount /dev/dm-0 /mnt -o ro
  echo 102400 &gt; /sys/module/dm_bufio/parameters/max_cache_size_bytes
  [read files in /mnt]

Cc: stable@vger.kernel.org	# v6.4+
Fixes: 450e8dee51aa ("dm bufio: improve concurrent IO performance")
Signed-off-by: Wang Shuai &lt;wangshuai12@xiaomi.com&gt;
Signed-off-by: Sheng Yong &lt;shengyong1@xiaomi.com&gt;
Signed-off-by: Mikulas Patocka &lt;mpatocka@redhat.com&gt;
</content>
</entry>
<entry>
<title>Merge tag 'for-6.16/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm</title>
<updated>2025-06-03T22:54:46+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2025-06-03T22:54:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=3c727285f1c99b4be5458ba68329748b207584ce'/>
<id>urn:sha1:3c727285f1c99b4be5458ba68329748b207584ce</id>
<content type='text'>
Pull device mapper updates from Mikulas Patocka:

 - better error handling when reloading a table

 - use use generic disable_* functions instead of open coding them

 - lock queue limits when reading them

 - remove unneeded kvfree from alloc_targets

 - fix BLK_FEAT_ATOMIC_WRITES

 - pass through operations on wrapped inline crypto keys

 - dm-verity:
     - use softirq context only when !need_resched()
     - fix a memory leak if some arguments are specified multiple times

 - dm-mpath:
    - interface for explicit probing of active paths
    - replace spin_lock_irqsave with spin_lock_irq

 - dm-delay: don't busy-wait in kthread

 - dm-bufio: remove maximum age based eviction

 - dm-flakey: various fixes

 - vdo indexer: don't read request structure after enqueuing

 - dm-zone: Use bdev_*() helper functions where applicable

 - dm-mirror: fix a tiny race condition

 - dm-stripe: small code cleanup

* tag 'for-6.16/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (29 commits)
  dm-stripe: small code cleanup
  dm-verity: fix a memory leak if some arguments are specified multiple times
  dm-mirror: fix a tiny race condition
  dm-table: check BLK_FEAT_ATOMIC_WRITES inside limits_lock
  dm mpath: replace spin_lock_irqsave with spin_lock_irq
  dm-mpath: Don't grab work_mutex while probing paths
  dm-zone: Use bdev_*() helper functions where applicable
  dm vdo indexer: don't read request structure after enqueuing
  dm: pass through operations on wrapped inline crypto keys
  blk-crypto: export wrapped key functions
  dm-table: Set BLK_FEAT_ATOMIC_WRITES for target queue limits
  dm mpath: Interface for explicit probing of active paths
  dm: Allow .prepare_ioctl to handle ioctls directly
  dm-flakey: make corrupting read bios work
  dm-flakey: remove useless ERROR_READS check in flakey_end_io
  dm-flakey: error all IOs when num_features is absent
  dm-flakey: Clean up parsing messages
  dm: remove unneeded kvfree from alloc_targets
  dm-bufio: remove maximum age based eviction
  dm-verity: use softirq context only when !need_resched()
  ...
</content>
</entry>
</feed>
