Pull block updates from Jens Axboe:
- MD pull request via Yu:
- call del_gendisk synchronously (Xiao)
- cleanup unused variable (John)
- cleanup workqueue flags (Ryo)
- fix faulty rdev can't be removed during resync (Qixing)
- NVMe pull request via Christoph:
- try PCIe function level reset on init failure (Keith Busch)
- log TLS handshake failures at error level (Maurizio Lombardi)
- pci-epf: do not complete commands twice if nvmet_req_init()
fails (Rick Wertenbroek)
- misc cleanups (Alok Tiwari)
- Removal of the pktcdvd driver
This has been more than a decade coming at this point, and some
recently revealed breakages that had it causing issues even for cases
where it isn't required made me re-pull the trigger on this one. It's
known broken and nobody has stepped up to maintain the code
- Series for ublk supporting batch commands, enabling the use of
multishot where appropriate
- Speed up ublk exit handling
- Fix for the two-stage elevator fixing which could leak data
- Convert NVMe to use the new IOVA based API
- Increase default max transfer size to something more reasonable
- Series fixing write operations on zoned DM devices
- Add tracepoints for zoned block device operations
- Prep series working towards improving blk-mq queue management in the
presence of isolated CPUs
- Don't allow updating of the block size of a loop device that is
currently under exclusively ownership/open
- Set chunk sectors from stacked device stripe size and use it for the
atomic write size limit
- Switch to folios in bcache read_super()
- Fix for CD-ROM MRW exit flush handling
- Various tweaks, fixes, and cleanups
* tag 'for-6.17/block-20250728' of git://git.kernel.dk/linux: (94 commits)
block: restore two stage elevator switch while running nr_hw_queue update
cdrom: Call cdrom_mrw_exit from cdrom_release function
sunvdc: Balance device refcount in vdc_port_mpgroup_check
nvme-pci: try function level reset on init failure
dm: split write BIOs on zone boundaries when zone append is not emulated
block: use chunk_sectors when evaluating stacked atomic write limits
dm-stripe: limit chunk_sectors to the stripe size
md/raid10: set chunk_sectors limit
md/raid0: set chunk_sectors limit
block: sanitize chunk_sectors for atomic write limits
ilog2: add max_pow_of_two_factor()
nvmet: pci-epf: Do not complete commands twice if nvmet_req_init() fails
nvme-tcp: log TLS handshake failures at error level
docs: nvme: fix grammar in nvme-pci-endpoint-target.rst
nvme: fix typo in status code constant for self-test in progress
nvmet: remove redundant assignment of error code in nvmet_ns_enable()
nvme: fix incorrect variable in io cqes error message
nvme: fix multiple spelling and grammar issues in host drivers
block: fix blk_zone_append_update_request_bio() kernel-doc
md/raid10: fix set but not used variable in sync_request_write()
...
Pull vfs 'protection info' updates from Christian Brauner:
"This adds the new FS_IOC_GETLBMD_CAP ioctl() to query metadata and
protection info (PI) capabilities. This ioctl returns information
about the files integrity profile. This is useful for userspace
applications to understand a files end-to-end data protection support
and configure the I/O accordingly.
For now this interface is only supported by block devices. However the
design and placement of this ioctl in generic FS ioctl space allows us
to extend it to work over files as well. This maybe useful when
filesystems start supporting PI-aware layouts.
A new structure struct logical_block_metadata_cap is introduced, which
contains the following fields:
- lbmd_flags:
bitmask of logical block metadata capability flags
- lbmd_interval:
the amount of data described by each unit of logical block metadata
- lbmd_size:
size in bytes of the logical block metadata associated with each
interval
- lbmd_opaque_size:
size in bytes of the opaque block tag associated with each interval
- lbmd_opaque_offset:
offset in bytes of the opaque block tag within the logical block
metadata
- lbmd_pi_size:
size in bytes of the T10 PI tuple associated with each interval
- lbmd_pi_offset:
offset in bytes of T10 PI tuple within the logical block metadata
- lbmd_pi_guard_tag_type:
T10 PI guard tag type
- lbmd_pi_app_tag_size:
size in bytes of the T10 PI application tag
- lbmd_pi_ref_tag_size:
size in bytes of the T10 PI reference tag
- lbmd_pi_storage_tag_size:
size in bytes of the T10 PI storage tag
The internal logic to fetch the capability is encapsulated in a helper
function blk_get_meta_cap(), which uses the blk_integrity profile
associated with the device. The ioctl returns -EOPNOTSUPP, if
CONFIG_BLK_DEV_INTEGRITY is not enabled"
* tag 'vfs-6.17-rc1.integrity' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
block: fix lbmd_guard_tag_type assignment in FS_IOC_GETLBMD_CAP
block: fix FS_IOC_GETLBMD_CAP parsing in blkdev_common_ioctl()
fs: add ioctl to query metadata and protection info capabilities
nvme: set pi_offset only when checksum type is not BLK_INTEGRITY_CSUM_NONE
block: introduce pi_tuple_size field in blk_integrity
block: rename tuple_size field in blk_integrity to metadata_size
Pull fallocate updates from Christian Brauner:
"fallocate() currently supports creating preallocated files
efficiently. However, on most filesystems fallocate() will preallocate
blocks in an unwriten state even if FALLOC_FL_ZERO_RANGE is specified.
The extent state must later be converted to a written state when the
user writes data into this range, which can trigger numerous metadata
changes and journal I/O. This may leads to significant write
amplification and performance degradation in synchronous write mode.
At the moment, the only method to avoid this is to create an empty
file and write zero data into it (for example, using 'dd' with a large
block size). However, this method is slow and consumes a considerable
amount of disk bandwidth.
Now that more and more flash-based storage devices are available it is
possible to efficiently write zeros to SSDs using the unmap write
zeroes command if the devices do not write physical zeroes to the
media.
For example, if SCSI SSDs support the UMMAP bit or NVMe SSDs support
the DEAC bit[1], the write zeroes command does not write actual data
to the device, instead, NVMe converts the zeroed range to a
deallocated state, which works fast and consumes almost no disk write
bandwidth.
This series implements the BLK_FEAT_WRITE_ZEROES_UNMAP feature and
BLK_FLAG_WRITE_ZEROES_UNMAP_DISABLED flag for SCSI, NVMe and
device-mapper drivers, and add the FALLOC_FL_WRITE_ZEROES and
STATX_ATTR_WRITE_ZEROES_UNMAP support for ext4 and raw bdev devices.
fallocate() is subsequently extended with the FALLOC_FL_WRITE_ZEROES
flag. FALLOC_FL_WRITE_ZEROES zeroes a specified file range in such a
way that subsequent writes to that range do not require further
changes to the file mapping metadata. This flag is beneficial for
subsequent pure overwriting within this range, as it can save on block
allocation and, consequently, significant metadata changes"
* tag 'vfs-6.17-rc1.fallocate' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
ext4: add FALLOC_FL_WRITE_ZEROES support
block: add FALLOC_FL_WRITE_ZEROES support
block: factor out common part in blkdev_fallocate()
fs: introduce FALLOC_FL_WRITE_ZEROES to fallocate
dm: clear unmap write zeroes limits when disabling write zeroes
scsi: sd: set max_hw_wzeroes_unmap_sectors if device supports SD_ZERO_*_UNMAP
nvmet: set WZDS and DRB if device enables unmap write zeroes operation
nvme: set max_hw_wzeroes_unmap_sectors if device supports DEAC bit
block: introduce max_{hw|user}_wzeroes_unmap_sectors to queue limits
Pull MD updates from Yu:
"- call del_gendisk synchronously, from Xiao
- cleanup unused variable, from John
- cleanup workqueue flags, from Ryo
- fix faulty rdev can't be removed during resync, from Qixing"
* tag 'md-6.17-20250722' of https://git.kernel.org/pub/scm/linux/kernel/git/mdraid/linux:
md/raid10: fix set but not used variable in sync_request_write()
md: allow removing faulty rdev during resync
md/raid5: unset WQ_CPU_INTENSIVE for raid5 unbound workqueue
md: remove/add redundancy group only in level change
md: Don't clear MD_CLOSING until mddev is freed
md: call del_gendisk in control path
Commit 2df7168717 ("dm: Always split write BIOs to zoned device
limits") updates the device-mapper driver to perform splits for the
write BIOs. However, it did not address the cases where DM targets do
not emulate zone append, such as in the cases of dm-linear or dm-flakey.
For these targets, when the write BIOs span across zone boundaries, they
trigger WARN_ON_ONCE(bio_straddles_zones(bio)) in
blk_zone_wplug_handle_write(). This results in I/O errors. The errors
are reproduced by running blktests test case zbd/004 using zoned
dm-linear or dm-flakey devices.
To avoid the I/O errors, handle the write BIOs regardless whether DM
targets emulate zone append or not, so that all write BIOs are split at
zone boundaries. For that purpose, drop the check for zone append
emulation in dm_zone_bio_needs_split(). Its argument 'md' is no longer
used then drop it also.
Fixes: 2df7168717 ("dm: Always split write BIOs to zoned device limits")
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>
Link: https://lore.kernel.org/r/20250717103539.37279-1-shinichiro.kawasaki@wdc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently we use min io size as the chunk size when deciding on the
atomic write size limits - see blk_stack_atomic_writes_head().
The limit min_io size is not a reliable value to store the chunk size, as
this may be mutated by the block stacking code. Such an example would be
for the min io size less than the physical block size, and the min io size
is raised to the physical block size - see blk_stack_limits().
The block stacking limits will rely on chunk_sectors in future,
so set this value (to the chunk size).
Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: John Garry <john.g.garry@oracle.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20250711105258.3135198-4-john.g.garry@oracle.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Building with W=1 reports the following:
drivers/md/raid10.c: In function ‘sync_request_write’:
drivers/md/raid10.c:2441:21: error: variable ‘d’ set but not used [-Werror=unused-but-set-variable]
2441 | int d;
| ^
cc1: all warnings being treated as errors
Remove the usage of that variable.
Fixes: 752d0464b7 ("md: clean up accounting for issued sync IO")
Signed-off-by: John Garry <john.g.garry@oracle.com>
Link: https://lore.kernel.org/linux-raid/20250709104814.2307276-1-john.g.garry@oracle.com
Signed-off-by: Yu Kuai <yukuai@kernel.org>
Pull device mapper fix from Mikulas Patocka:
- dm-bufio: fix scheduling in atomic
* tag 'for-6.16/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm-bufio: fix sched in atomic context
During RAID resync, faulty rdev cannot be removed and will result in
"Device or resource busy" error when attempting hot removal.
Reproduction steps:
mdadm -Cv /dev/md0 -l1 -n3 -e1.2 /dev/sd{b..d}
mdadm /dev/md0 -f /dev/sdb
mdadm /dev/md0 -r /dev/sdb
-> mdadm: hot remove failed for /dev/sdb: Device or resource busy
After commit 4b10a3bc67 ("md: ensure resync is prioritized over
recovery"), when a device becomes faulty during resync, the
md_choose_sync_action() function returns early without calling
remove_and_add_spares(), preventing faulty device removal.
This patch extracts a helper function remove_spares() to support
removing faulty devices during RAID resync operations.
Fixes: 4b10a3bc67 ("md: ensure resync is prioritized over recovery")
Signed-off-by: Zheng Qixing <zhengqixing@huawei.com>
Reviewed-by: Li Nan <linan122@huawei.com>
Link: https://lore.kernel.org/linux-raid/20250707075412.150301-1-zhengqixing@huaweicloud.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
When specified with WQ_CPU_INTENSIVE, the workqueue doesn't
participate in concurrency management. This behaviour is already
accounted for WQ_UNBOUND workqueues given that they are assigned
to their own worker threads.
Unset WQ_CPU_INTENSIVE as the use of flag has no effect when
used with WQ_UNBOUND.
Signed-off-by: Ryo Takakura <ryotkkr98@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/linux-raid/20250601013702.64640-1-ryotkkr98@gmail.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
UNTIL_STOP is used to avoid mddev is freed on the last close before adding
disks to mddev. And it should be cleared when stopping an array which is
mentioned in commit efeb53c0e5 ("md: Allow md devices to be created by
name."). So reset ->hold_active to 0 in md_clean.
And MD_CLOSING should be kept until mddev is freed to avoid reopen.
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Xiao Ni <xni@redhat.com>
Link: https://lore.kernel.org/linux-raid/20250611073108.25463-3-xni@redhat.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Now del_gendisk and put_disk are called asynchronously in workqueue work.
The asynchronous way has a problem that the device node can still exist
after mdadm --stop command returns in a short window. So udev rule can
open this device node and create the struct mddev in kernel again. So put
del_gendisk in control path and still leave put_disk in md_kobj_release
to avoid uaf of gendisk.
Function del_gendisk can't be called with reconfig_mutex. If it's called
with reconfig mutex, a deadlock can happen. del_gendisk waits all sysfs
files access to finish and sysfs file access waits reconfig mutex. So
put del_gendisk after releasing reconfig mutex.
But there is still a window that sysfs can be accessed between mddev_unlock
and del_gendisk. So some actions (add disk, change level, .e.g) can happen
which lead unexpected results. MD_DELETED is used to resolve this problem.
MD_DELETED is set before releasing reconfig mutex and it should be checked
for these sysfs access which need reconfig mutex. For sysfs access which
don't need reconfig mutex, del_gendisk will wait them to finish.
But it doesn't need to do this in function mddev_lock_nointr. There are
ten places that call it.
* Five of them are in dm raid which we don't need to care. MD_DELETED is
only used for md raid.
* stop_sync_thread, md_do_sync and md_start_sync are related sync request,
and it needs to wait sync thread to finish before stopping an array.
* md_ioctl: md_open is called before md_ioctl, so ->openers is added. It
will fail to stop the array. So it doesn't need to check MD_DELETED here
* md_set_readonly:
It needs to call mddev_set_closing_and_sync_blockdev when setting readonly
or read_auto. So it will fail to stop the array too because MD_CLOSING is
already set.
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Xiao Ni <xni@redhat.com>
Link: https://lore.kernel.org/linux-raid/20250611073108.25463-2-xni@redhat.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Pull block fixes from Jens Axboe:
- MD changes via Yu:
- fix UAF due to stack memory used for bio mempool (Jinchao)
- fix raid10/raid1 nowait IO error path (Nigel and Qixing)
- fix kernel crash from reading bitmap sysfs entry (Håkon)
- Fix for a UAF in the nbd connect error path
- Fix for blocksize being bigger than pagesize, if THP isn't enabled
* tag 'block-6.16-20250710' of git://git.kernel.dk/linux:
block: reject bs > ps block devices when THP is disabled
nbd: fix uaf in nbd_genl_connect() error path
md/md-bitmap: fix GPF in bitmap_get_stats()
md/raid1,raid10: strip REQ_NOWAIT from member bios
raid10: cleanup memleak at raid10_make_request
md/raid1: Fix stack memory use after return in raid1_reshape
The commit message of commit 6ec1f02394 ("md/md-bitmap: fix stats
collection for external bitmaps") states:
Remove the external bitmap check as the statistics should be
available regardless of bitmap storage location.
Return -EINVAL only for invalid bitmap with no storage (neither in
superblock nor in external file).
But, the code does not adhere to the above, as it does only check for
a valid super-block for "internal" bitmaps. Hence, we observe:
Oops: GPF, probably for non-canonical address 0x1cd66f1f40000028
RIP: 0010:bitmap_get_stats+0x45/0xd0
Call Trace:
seq_read_iter+0x2b9/0x46a
seq_read+0x12f/0x180
proc_reg_read+0x57/0xb0
vfs_read+0xf6/0x380
ksys_read+0x6d/0xf0
do_syscall_64+0x8c/0x1b0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
We fix this by checking the existence of a super-block for both the
internal and external case.
Fixes: 6ec1f02394 ("md/md-bitmap: fix stats collection for external bitmaps")
Cc: stable@vger.kernel.org
Reported-by: Gerald Gibson <gerald.gibson@oracle.com>
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Link: https://lore.kernel.org/linux-raid/20250702091035.2061312-1-haakon.bugge@oracle.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
RAID layers don't implement proper non-blocking semantics for
REQ_NOWAIT, making the flag potentially misleading when propagated
to member disks.
This patch clear REQ_NOWAIT from cloned bios in raid1/raid10. Retain
original bio's REQ_NOWAIT flag for upper layer error handling.
Maybe we can implement non-blocking I/O handling mechanisms within
RAID in future work.
Fixes: 9f346f7d4e ("md/raid1,raid10: don't handle IO error for
REQ_RAHEAD and REQ_NOWAIT")
Signed-off-by: Zheng Qixing <zhengqixing@huawei.com>
Link: https://lore.kernel.org/linux-raid/20250702102341.1969154-1-zhengqixing@huaweicloud.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
In the raid1_reshape function, newpool is
allocated on the stack and assigned to conf->r1bio_pool.
This results in conf->r1bio_pool.wait.head pointing
to a stack address.
Accessing this address later can lead to a kernel panic.
Example access path:
raid1_reshape()
{
// newpool is on the stack
mempool_t newpool, oldpool;
// initialize newpool.wait.head to stack address
mempool_init(&newpool, ...);
conf->r1bio_pool = newpool;
}
raid1_read_request() or raid1_write_request()
{
alloc_r1bio()
{
mempool_alloc()
{
// if pool->alloc fails
remove_element()
{
--pool->curr_nr;
}
}
}
}
mempool_free()
{
if (pool->curr_nr < pool->min_nr) {
// pool->wait.head is a stack address
// wake_up() will try to access this invalid address
// which leads to a kernel panic
return;
wake_up(&pool->wait);
}
}
Fix:
reinit conf->r1bio_pool.wait after assigning newpool.
Fixes: afeee514ce ("md: convert to bioset_init()/mempool_init()")
Signed-off-by: Wang Jinchao <wangjinchao600@gmail.com>
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/linux-raid/20250612112901.3023950-1-wangjinchao600@gmail.com
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
DM targets must not split zone append and write operations using
dm_accept_partial_bio() as doing so is forbidden for zone append BIOs,
breaks zone append emulation using regular write BIOs and potentially
creates deadlock situations with queue freeze operations.
Modify dm_accept_partial_bio() to add missing BUG_ON() checks for all
these cases, that is, check that the BIO is a write or write zeroes
operation. This change packs all the zone related checks together under
a static_branch_unlikely(&zoned_enabled) and done only if the target is
a zoned device.
Fixes: f211268ed1 ("dm: Use the block layer zone append emulation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>
Link: https://lore.kernel.org/r/20250625093327.548866-6-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Read and write operations issued to a dm-crypt target may be split
according to the dm-crypt internal limits defined by the max_read_size
and max_write_size module parameters (default is 128 KB). The intent is
to improve processing time of large BIOs by splitting them into smaller
operations that can be parallelized on different CPUs.
For zoned dm-crypt targets, this BIO splitting is still done but without
the parallel execution to ensure that the issuing order of write
operations to the underlying devices remains sequential. However, the
splitting itself causes other problems:
1) Since dm-crypt relies on the block layer zone write plugging to
handle zone append emulation using regular write operations, the
reminder of a split write BIO will always be plugged into the target
zone write plugged. Once the on-going write BIO finishes, this
reminder BIO is unplugged and issued from the zone write plug work.
If this reminder BIO itself needs to be split, the reminder will be
re-issued and plugged again, but that causes a call to a
blk_queue_enter(), which may block if a queue freeze operation was
initiated. This results in a deadlock as DM submission still holds
BIOs that the queue freeze side is waiting for.
2) dm-crypt relies on the emulation done by the block layer using
regular write operations for processing zone append operations. This
still requires to properly return the written sector as the BIO
sector of the original BIO. However, this can be done correctly only
and only if there is a single clone BIO used for processing the
original zone append operation issued by the user. If the size of a
zone append operation is larger than dm-crypt max_write_size, then
the orginal BIO will be split and processed as a chain of regular
write operations. Such chaining result in an incorrect written sector
being returned to the zone append issuer using the original BIO
sector. This in turn results in file system data corruptions using
xfs or btrfs.
Fix this by modifying get_max_request_size() to always return the size
of the BIO to avoid it being split with dm_accpet_partial_bio() in
crypt_map(). get_max_request_size() is renamed to
get_max_request_sectors() to clarify the unit of the value returned
and its interface is changed to take a struct dm_target pointer and a
pointer to the struct bio being processed. In addition to this change,
to ensure that crypt_alloc_buffer() works correctly, set the dm-crypt
device max_hw_sectors limit to be at most
BIO_MAX_VECS << PAGE_SECTORS_SHIFT (1 MB with a 4KB page architecture).
This forces DM core to split write BIOs before passing them to
crypt_map(), and thus guaranteeing that dm-crypt can always accept an
entire write BIO without needing to split it.
This change does not have any effect on the read path of dm-crypt. Read
operations can still be split and the BIO fragments processed in
parallel. There is also no impact on the performance of the write path
given that all zone write BIOs were already processed inline instead of
in parallel.
This change also does not affect in any way regular dm-crypt block
devices.
Fixes: f211268ed1 ("dm: Use the block layer zone append emulation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>
Link: https://lore.kernel.org/r/20250625093327.548866-5-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Any zoned DM target that requires zone append emulation will use the
block layer zone write plugging. In such case, DM target drivers must
not split BIOs using dm_accept_partial_bio() as doing so can potentially
lead to deadlocks with queue freeze operations. Regular write operations
used to emulate zone append operations also cannot be split by the
target driver as that would result in an invalid writen sector value
return using the BIO sector.
In order for zoned DM target drivers to avoid such incorrect BIO
splitting, we must ensure that large BIOs are split before being passed
to the map() function of the target, thus guaranteeing that the
limits for the mapped device are not exceeded.
dm-crypt and dm-flakey are the only target drivers supporting zoned
devices and using dm_accept_partial_bio().
In the case of dm-crypt, this function is used to split BIOs to the
internal max_write_size limit (which will be suppressed in a different
patch). However, since crypt_alloc_buffer() uses a bioset allowing only
up to BIO_MAX_VECS (256) vectors in a BIO. The dm-crypt device
max_segments limit, which is not set and so default to BLK_MAX_SEGMENTS
(128), must thus be respected and write BIOs split accordingly.
In the case of dm-flakey, since zone append emulation is not required,
the block layer zone write plugging is not used and no splitting of BIOs
required.
Modify the function dm_zone_bio_needs_split() to use the block layer
helper function bio_needs_zone_write_plugging() to force a call to
bio_split_to_limits() in dm_split_and_process_bio(). This allows DM
target drivers to avoid using dm_accept_partial_bio() for write
operations on zoned DM devices.
Fixes: f211268ed1 ("dm: Use the block layer zone append emulation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20250625093327.548866-4-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for fixing device mapper zone write handling, introduce
the inline helper function bio_needs_zone_write_plugging() to test if a
BIO requires handling through zone write plugging using the function
blk_zone_plug_bio(). This function returns true for any write
(op_is_write(bio) == true) operation directed at a zoned block device
using zone write plugging, that is, a block device with a disk that has
a zone write plug hash table.
This helper allows simplifying the check on entry to blk_zone_plug_bio()
and used in to protect calls to it for blk-mq devices and DM devices.
Fixes: f211268ed1 ("dm: Use the block layer zone append emulation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20250625093327.548866-3-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull device mapper fixes from Mikulas Patocka:
- dm-crypt: fix a crash on 32-bit machines
- dm-raid: replace "rdev" with correct loop variable name "r"
* tag 'for-6.16/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm-raid: fix variable in journal device check
dm-crypt: Extend state buffer size in crypt_iv_lmk_one
This reverts commit 866898efbb.
The generic bottom-up min_heap implementation causes performance
regression in invalidate_buckets_lru(), a hot path in bcache. Before the
cache is fully populated, new_bucket_prio() often returns zero, leading to
many equal comparisons. In such cases, bottom-up sift_down performs up to
2 * log2(n) comparisons, while the original top-down approach completes
with just O() comparisons, resulting in a measurable performance gap.
The performance degradation is further worsened by the non-inlined
min_heap API functions introduced in commit 92a8b224b8 ("lib/min_heap:
introduce non-inline versions of min heap API functions"), adding function
call overhead to this critical path.
As reported by Robert, bcache now suffers from latency spikes, with P100
(max) latency increasing from 600 ms to 2.4 seconds every 5 minutes.
These regressions degrade bcache's effectiveness as a low-latency cache
layer and lead to frequent timeouts and application stalls in production
environments.
This revert aims to restore bcache's original low-latency behavior.
Link: https://lore.kernel.org/lkml/CAJhEC05+0S69z+3+FB2Cd0hD+pCRyWTKLEOsc8BOmH73p1m+KQ@mail.gmail.com
Link: https://lkml.kernel.org/r/20250614202353.1632957-3-visitorckw@gmail.com
Fixes: 866898efbb ("bcache: remove heap-related macros and switch to generic min_heap")
Fixes: 92a8b224b8 ("lib/min_heap: introduce non-inline versions of min heap API functions")
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Reported-by: Robert Pang <robertpang@google.com>
Closes: https://lore.kernel.org/linux-bcache/CAJhEC06F_AtrPgw2-7CvCqZgeStgCtitbD-ryuPpXQA-JG5XXw@mail.gmail.com
Acked-by: Coly Li <colyli@kernel.org>
Cc: Ching-Chun (Jim) Huang <jserv@ccns.ncku.edu.tw>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "bcache: Revert min_heap migration due to performance
regression".
This patch series reverts the migration of bcache from its original heap
implementation to the generic min_heap library. While the original change
aimed to simplify the code and improve maintainability, it introduced a
severe performance regression in real-world scenarios.
As reported by Robert, systems using bcache now suffer from periodic
latency spikes, with P100 (max) latency increasing from 600 ms to 2.4
seconds every 5 minutes. This degrades bcache's value as a low-latency
caching layer, and leads to frequent timeouts and application stalls in
production environments.
The primary cause of this regression is the behavior of the generic
min_heap implementation's bottom-up sift_down, which performs up to 2 *
log2(n) comparisons when many elements are equal. The original top-down
variant used by bcache only required O(1) comparisons in such cases. The
issue was further exacerbated by commit 92a8b224b8 ("lib/min_heap:
introduce non-inline versions of min heap API functions"), which
introduced non-inlined versions of the min_heap API, adding function call
overhead to a performance-critical hot path.
This patch (of 3):
This reverts commit 3d8a9a1c35.
Although removing the custom swap function simplified the code, this
change is part of a broader migration to the generic min_heap API that
introduced significant performance regressions in bcache.
As reported by Robert, bcache now suffers from latency spikes, with P100
(max) latency increasing from 600 ms to 2.4 seconds every 5 minutes.
These regressions degrade bcache's effectiveness as a low-latency cache
layer and lead to frequent timeouts and application stalls in production
environments.
This revert is part of a series of changes to restore previous performance
by undoing the min_heap transition.
Link: https://lkml.kernel.org/r/20250614202353.1632957-1-visitorckw@gmail.com
Link: https://lore.kernel.org/lkml/CAJhEC05+0S69z+3+FB2Cd0hD+pCRyWTKLEOsc8BOmH73p1m+KQ@mail.gmail.com
Link: https://lkml.kernel.org/r/20250614202353.1632957-2-visitorckw@gmail.com
Fixes: 866898efbb ("bcache: remove heap-related macros and switch to generic min_heap")
Fixes: 92a8b224b8 ("lib/min_heap: introduce non-inline versions of min heap API functions")
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Reported-by: Robert Pang <robertpang@google.com>
Closes: https://lore.kernel.org/linux-bcache/CAJhEC06F_AtrPgw2-7CvCqZgeStgCtitbD-ryuPpXQA-JG5XXw@mail.gmail.com
Acked-by: Coly Li <colyli@kernel.org>
Cc: Ching-Chun (Jim) Huang <jserv@ccns.ncku.edu.tw>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pull more block updates from Jens Axboe:
- NVMe pull request via Christoph:
- TCP error handling fix (Shin'ichiro Kawasaki)
- TCP I/O stall handling fixes (Hannes Reinecke)
- fix command limits status code (Keith Busch)
- support vectored buffers also for passthrough (Pavel Begunkov)
- spelling fixes (Yi Zhang)
- MD pull request via Yu:
- fix REQ_RAHEAD and REQ_NOWAIT IO err handling for raid1/10
- fix max_write_behind setting for dm-raid
- some minor cleanups
- Integrity data direction fix and cleanup
- bcache NULL pointer fix
- Fix for loop missing write start/end handling
- Decouple hardware queues and IO threads in ublk
- Slew of ublk selftests additions and updates
* tag 'block-6.16-20250606' of git://git.kernel.dk/linux: (29 commits)
nvme: spelling fixes
nvme-tcp: fix I/O stalls on congested sockets
nvme-tcp: sanitize request list handling
nvme-tcp: remove tag set when second admin queue config fails
nvme: enable vectored registered bufs for passthrough cmds
nvme: fix implicit bool to flags conversion
nvme: fix command limits status code
selftests: ublk: kublk: improve behavior on init failure
block: flip iter directions in blk_rq_integrity_map_user()
block: drop direction param from bio_integrity_copy_user()
selftests: ublk: cover PER_IO_DAEMON in more stress tests
Documentation: ublk: document UBLK_F_PER_IO_DAEMON
selftests: ublk: add stress test for per io daemons
selftests: ublk: add functional test for per io daemons
selftests: ublk: kublk: decouple ublk_queues from ublk server threads
selftests: ublk: kublk: move per-thread data out of ublk_queue
selftests: ublk: kublk: lift queue initialization out of thread
selftests: ublk: kublk: tie sqe allocation to io instead of queue
selftests: ublk: kublk: plumb q_id in io_uring user_data
ublk: have a per-io daemon instead of a per-queue daemon
...
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()
...
This commit doesn't fix any bug, it is just code cleanup. Use the
function format_dev_t instead of sprintf, because format_dev_t does the
same thing.
Remove the useless memset call.
An unsigned integer can take at most 10 digits, so extend the array size
to 22. (note that because the range of minor and major numbers is limited,
the size 16 could not be exceeded, thus this function couldn't write
beyond string end)
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
If some of the arguments "check_at_most_once", "ignore_zero_blocks",
"use_fec_from_device", "root_hash_sig_key_desc" were specified more than
once on the target line, a memory leak would happen.
This commit fixes the memory leak. It also fixes error handling in
verity_verify_sig_parse_opt_args.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: stable@vger.kernel.org
There's a tiny race condition in dm-mirror. The functions queue_bio and
write_callback grab a spinlock, add a bio to the list, drop the spinlock
and wake up the mirrord thread that processes bios in the list.
It may be possible that the mirrord thread processes the bio just after
spin_unlock_irqrestore is called, before wakeup_mirrord. This spurious
wake-up is normally harmless, however if the device mapper device is
unloaded just after the bio was processed, it may be possible that
wakeup_mirrord(ms) uses invalid "ms" pointer.
Fix this bug by moving wakeup_mirrord inside the spinlock.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: stable@vger.kernel.org
dm_set_device_limits() should check q->limits.features for
BLK_FEAT_ATOMIC_WRITES while holding q->limits_lock, like it does for
the rest of the queue limits.
Fixes: b7c18b17a1 ("dm-table: Set BLK_FEAT_ATOMIC_WRITES for target queue limits")
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Pull non-MM updates from Andrew Morton:
- "hung_task: extend blocking task stacktrace dump to semaphore" from
Lance Yang enhances the hung task detector.
The detector presently dumps the blocking tasks's stack when it is
blocked on a mutex. Lance's series extends this to semaphores
- "nilfs2: improve sanity checks in dirty state propagation" from
Wentao Liang addresses a couple of minor flaws in nilfs2
- "scripts/gdb: Fixes related to lx_per_cpu()" from Illia Ostapyshyn
fixes a couple of issues in the gdb scripts
- "Support kdump with LUKS encryption by reusing LUKS volume keys" from
Coiby Xu addresses a usability problem with kdump.
When the dump device is LUKS-encrypted, the kdump kernel may not have
the keys to the encrypted filesystem. A full writeup of this is in
the series [0/N] cover letter
- "sysfs: add counters for lockups and stalls" from Max Kellermann adds
/sys/kernel/hardlockup_count and /sys/kernel/hardlockup_count and
/sys/kernel/rcu_stall_count
- "fork: Page operation cleanups in the fork code" from Pasha Tatashin
implements a number of code cleanups in fork.c
- "scripts/gdb/symbols: determine KASLR offset on s390 during early
boot" from Ilya Leoshkevich fixes some s390 issues in the gdb
scripts
* tag 'mm-nonmm-stable-2025-05-31-15-28' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (67 commits)
llist: make llist_add_batch() a static inline
delayacct: remove redundant code and adjust indentation
squashfs: add optional full compressed block caching
crash_dump, nvme: select CONFIGFS_FS as built-in
scripts/gdb/symbols: determine KASLR offset on s390 during early boot
scripts/gdb/symbols: factor out pagination_off()
scripts/gdb/symbols: factor out get_vmlinux()
kernel/panic.c: format kernel-doc comments
mailmap: update and consolidate Casey Connolly's name and email
nilfs2: remove wbc->for_reclaim handling
fork: define a local GFP_VMAP_STACK
fork: check charging success before zeroing stack
fork: clean-up naming of vm_stack/vm_struct variables in vmap stacks code
fork: clean-up ifdef logic around stack allocation
kernel/rcu/tree_stall: add /sys/kernel/rcu_stall_count
kernel/watchdog: add /sys/kernel/{hard,soft}lockup_count
x86/crash: make the page that stores the dm crypt keys inaccessible
x86/crash: pass dm crypt keys to kdump kernel
Revert "x86/mm: Remove unused __set_memory_prot()"
crash_dump: retrieve dm crypt keys in kdump kernel
...
Pull hardening updates from Kees Cook:
- Update overflow helpers to ease refactoring of on-stack flex array
instances (Gustavo A. R. Silva, Kees Cook)
- lkdtm: Use SLAB_NO_MERGE instead of constructors (Harry Yoo)
- Simplify CONFIG_CC_HAS_COUNTED_BY (Jan Hendrik Farr)
- Disable u64 usercopy KUnit test on 32-bit SPARC (Thomas Weißschuh)
- Add missed designated initializers now exposed by fixed randstruct
(Nathan Chancellor, Kees Cook)
- Document compilers versions for __builtin_dynamic_object_size
- Remove ARM_SSP_PER_TASK GCC plugin
- Fix GCC plugin randstruct, add selftests, and restore COMPILE_TEST
builds
- Kbuild: induce full rebuilds when dependencies change with GCC
plugins, the Clang sanitizer .scl file, or the randstruct seed.
- Kbuild: Switch from -Wvla to -Wvla-larger-than=1
- Correct several __nonstring uses for -Wunterminated-string-initialization
* tag 'hardening-v6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux: (23 commits)
Revert "hardening: Disable GCC randstruct for COMPILE_TEST"
lib/tests: randstruct: Add deep function pointer layout test
lib/tests: Add randstruct KUnit test
randstruct: gcc-plugin: Remove bogus void member
net: qede: Initialize qede_ll_ops with designated initializer
scsi: qedf: Use designated initializer for struct qed_fcoe_cb_ops
md/bcache: Mark __nonstring look-up table
integer-wrap: Force full rebuild when .scl file changes
randstruct: Force full rebuild when seed changes
gcc-plugins: Force full rebuild when plugins change
kbuild: Switch from -Wvla to -Wvla-larger-than=1
hardening: simplify CONFIG_CC_HAS_COUNTED_BY
overflow: Fix direct struct member initialization in _DEFINE_FLEX()
kunit/overflow: Add tests for STACK_FLEX_ARRAY_SIZE() helper
overflow: Add STACK_FLEX_ARRAY_SIZE() helper
input/joystick: magellan: Mark __nonstring look-up table const
watchdog: exar: Shorten identity name to fit correctly
mod_devicetable: Enlarge the maximum platform_device_id name length
overflow: Clarify expectations for getting DEFINE_FLEX variable sizes
compiler_types: Identify compiler versions for __builtin_dynamic_object_size
...
Reported an IO hang and unrecoverable error in our testing environment.
After careful research, we found that bch_allocator_thread is stuck,
the call stack is as follows:
[<0>] __switch_to+0xbc/0x108
[<0>] __closure_sync+0x7c/0xbc [bcache]
[<0>] bch_prio_write+0x430/0x448 [bcache]
[<0>] bch_allocator_thread+0xb44/0xb70 [bcache]
[<0>] kthread+0x124/0x130
[<0>] ret_from_fork+0x10/0x18
Moreover, the RESERVE_BTREE type bucket slot are empty and journal_full
occurs at the same time.
When the cache disk is first used, the sb.nJournal_buckets defaults to 0.
So, only 8 RESERVE_BTREE type buckets are reserved. If RESERVE_BTREE type
buckets used up or btree_check_reserve() failed when request handle btree
split, the request will be repeatedly retried and wait for alloc thread to
fill in.
After the alloc thread fills the buckets, it will call bch_prio_write().
If journal_full occurs simultaneously at this time, journal_reclaim() and
btree_flush_write() will be called sequentially, journal_write cannot be
completed.
This is a low probability event, we believe that reserve more RESERVE_BTREE
buckets can avoid the worst situation.
Fixes: 682811b3ce ("bcache: fix for allocator and register thread race")
Signed-off-by: Mingzhe Zou <mingzhe.zou@easystack.cn>
Signed-off-by: Coly Li <colyli@kernel.org>
Link: https://lore.kernel.org/r/20250527051601.74407-4-colyli@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
