Yu Kuai [Wed, 14 Jan 2026 17:12:29 +0000 (01:12 +0800)]
md/raid5: fix raid5_run() to return error when log_init() fails
Since commit f63f17350e53 ("md/raid5: use the atomic queue limit
update APIs"), the abort path in raid5_run() returns 'ret' instead of
-EIO. However, if log_init() fails, 'ret' is still 0 from the previous
successful call, causing raid5_run() to return success despite the
failure.
Fix this by capturing the return value from log_init().
Link: https://lore.kernel.org/linux-raid/20260114171241.3043364-2-yukuai@fnnas.com Fixes: f63f17350e53 ("md/raid5: use the atomic queue limit update APIs") Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Closes: https://lore.kernel.org/r/202601130531.LGfcZsa4-lkp@intel.com/ Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Li Nan <linan122@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
selftests: ublk: add missing gitignore for metadata_size binary
A new utility metadata_size was added in
commit 261b67f4e347 ("selftests: ublk: add utility to get block device metadata size")
but it was not added to .gitignore. Fix that by adding it there.
While at it sort all entries alphabetically and add a SPDX license header.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Fixes: 261b67f4e347 ("selftests: ublk: add utility to get block device metadata size") Signed-off-by: Alexander Atanasov <alex@zazolabs.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 23 Jan 2026 09:15:43 +0000 (17:15 +0800)]
ublk: fix canceling flag handling in batch I/O recovery
Two issues with ubq->canceling flag handling:
1) In ublk_queue_reset_io_flags(), ubq->canceling is set outside
cancel_lock, violating the locking requirement. Move it inside
the spinlock-protected section.
2) In ublk_batch_unprep_io(), when rolling back after a batch prep
failure, if the queue became ready during prep (which cleared
canceling), the flag is not restored when the queue becomes
not-ready again. This allows new requests to be queued to
uninitialized IO slots.
Fix by restoring ubq->canceling = true under cancel_lock when the
queue transitions from ready to not-ready during rollback.
Ming Lei [Fri, 23 Jan 2026 09:15:42 +0000 (17:15 +0800)]
ublk: move ublk_mark_io_ready() out of __ublk_fetch()
ublk_batch_prep_io() calls __ublk_fetch() while holding io->lock
spinlock. When the last IO makes the device ready, ublk_mark_io_ready()
tries to acquire ub->cancel_mutex which can sleep, causing a
sleeping-while-atomic bug.
Fix by moving ublk_mark_io_ready() out of __ublk_fetch() and into the
callers (ublk_fetch and ublk_batch_prep_io) after the spinlock is
released.
Reported-by: Jens Axboe <axboe@kernel.dk> Fixes: b256795b3606 ("ublk: handle UBLK_U_IO_PREP_IO_CMDS") Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:57 +0000 (22:18 +0800)]
selftests: ublk: support arbitrary threads/queues combination
Enable flexible thread-to-queue mapping in batch I/O mode to support
arbitrary combinations of threads and queues, improving resource
utilization and scalability.
Key improvements:
- Support N:M thread-to-queue mapping (previously limited to 1:1)
- Dynamic buffer allocation based on actual queue assignment per thread
- Thread-safe queue preparation with spinlock protection
- Intelligent buffer index calculation for multi-queue scenarios
- Enhanced validation for thread/queue combination constraints
Implementation details:
- Add q_thread_map matrix to track queue-to-thread assignments
- Dynamic allocation of commit and fetch buffers per thread
- Round-robin queue assignment algorithm for load balancing
- Per-queue spinlock to prevent race conditions during prep
- Updated buffer index calculation using queue position within thread
This enables efficient configurations like:
- Any other N:M combinations for optimal resource matching
Testing:
- Added test_batch_02.sh: 4 threads vs 1 queue
- Added test_batch_03.sh: 1 thread vs 4 queues
- Validates correctness across different mapping scenarios
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:55 +0000 (22:18 +0800)]
selftests: ublk: increase timeout to 150 seconds
More tests need to be covered in existing generic tests, and default
45sec isn't enough, and timeout is often triggered, increase timeout
by adding setting file.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:54 +0000 (22:18 +0800)]
selftests: ublk: handle UBLK_U_IO_FETCH_IO_CMDS
Add support for UBLK_U_IO_FETCH_IO_CMDS to enable efficient batch
fetching of I/O commands using multishot io_uring operations.
Key improvements:
- Implement multishot UBLK_U_IO_FETCH_IO_CMDS for continuous command fetching
- Add fetch buffer management with page-aligned, mlocked buffers
- Process fetched I/O command tags from kernel-provided buffers
- Integrate fetch operations with existing batch I/O infrastructure
- Significantly reduce uring_cmd issuing overhead through batching
The implementation uses two fetch buffers per thread with automatic
requeuing to maintain continuous I/O command flow. Each fetch operation
retrieves multiple command tags in a single syscall, dramatically
improving performance compared to individual command fetching.
Technical details:
- Fetch buffers are page-aligned and mlocked for optimal performance
- Uses IORING_URING_CMD_MULTISHOT for continuous operation
- Automatic buffer management and requeuing on completion
- Enhanced CQE handling for fetch command completions
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:53 +0000 (22:18 +0800)]
selftests: ublk: handle UBLK_U_IO_COMMIT_IO_CMDS
Implement UBLK_U_IO_COMMIT_IO_CMDS to enable efficient batched
completion of I/O operations in the batch I/O framework.
This completes the batch I/O infrastructure by adding the commit
phase that notifies the kernel about completed I/O operations:
Key features:
- Batch multiple I/O completions into single UBLK_U_IO_COMMIT_IO_CMDS
- Dynamic commit buffer allocation and management per thread
- Automatic commit buffer preparation before processing events
- Commit buffer submission after processing completed I/Os
- Integration with existing completion workflows
Implementation details:
- ublk_batch_prep_commit() allocates and initializes commit buffers
- ublk_batch_complete_io() adds completed I/Os to current batch
- ublk_batch_commit_io_cmds() submits batched completions to kernel
- Modified ublk_process_io() to handle batch commit lifecycle
- Enhanced ublk_complete_io() to route to batch or legacy completion
The commit buffer stores completion information (tag, result, buffer
details) for multiple I/Os, then submits them all at once, significantly
reducing syscall overhead compared to individual I/O completions.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:52 +0000 (22:18 +0800)]
selftests: ublk: handle UBLK_U_IO_PREP_IO_CMDS
Implement support for UBLK_U_IO_PREP_IO_CMDS in the batch I/O framework:
- Add batch command initialization and setup functions
- Implement prep command queueing with proper buffer management
- Add command completion handling for prep and commit commands
- Integrate batch I/O setup into thread initialization
- Update CQE handling to support batch commands
The implementation uses the previously established buffer management
infrastructure to queue UBLK_U_IO_PREP_IO_CMDS commands. Commands are
prepared in the first thread context and use commit buffers for
efficient command batching.
Key changes:
- ublk_batch_queue_prep_io_cmds() prepares I/O command batches
- ublk_batch_compl_cmd() handles batch command completions
- Modified thread setup to use batch operations when enabled
- Enhanced buffer index calculation for batch mode
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add the foundational infrastructure for UBLK_F_BATCH_IO buffer
management including:
- Allocator utility functions for small sized per-thread allocation
- Batch buffer allocation and deallocation functions
- Buffer index management for commit buffers
- Thread state management for batch I/O mode
- Buffer size calculation based on device features
This prepares the groundwork for handling batch I/O commands by
establishing the buffer management layer needed for UBLK_U_IO_PREP_IO_CMDS
and UBLK_U_IO_COMMIT_IO_CMDS operations.
The allocator uses CPU sets for efficient per-thread buffer tracking,
and commit buffers are pre-allocated with 2 buffers per thread to handle
overlapping command operations.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:50 +0000 (22:18 +0800)]
selftests: ublk: add ublk_io_buf_idx() for returning io buffer index
Since UBLK_F_PER_IO_DAEMON is added, io buffer index may depend on current
thread because the common way is to use per-pthread io_ring_ctx for issuing
ublk uring_cmd.
Add one helper for returning io buffer index, so we can hide the buffer
index implementation details for target code.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:48 +0000 (22:18 +0800)]
selftests: ublk: fix user_data truncation for tgt_data >= 256
The build_user_data() function packs multiple fields into a __u64
value using bit shifts. Without explicit __u64 casts before shifting,
the shift operations are performed on 32-bit unsigned integers before
being promoted to 64-bit, causing data loss.
Specifically, when tgt_data >= 256, the expression (tgt_data << 24)
shifts on a 32-bit value, truncating the upper 8 bits before promotion
to __u64. Since tgt_data can be up to 16 bits (assertion allows up to
65535), values >= 256 would have their high byte lost.
Add explicit __u64 casts to both op and tgt_data before shifting to
ensure the shift operations happen in 64-bit space, preserving all
bits of the input values.
user_data_to_tgt_data() is only used by stripe.c, in which the max
supported member disks are 4, so won't trigger this issue.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:47 +0000 (22:18 +0800)]
ublk: fix batch I/O recovery -ENODEV error
During recovery with batch I/O, UBLK_U_IO_FETCH_IO_CMDS command fails with
-ENODEV because ublk_batch_attach() rejects them when ubq->canceling is set.
The canceling flag remains set until all queues are ready.
Fix this by tracking per-queue readiness and clearing ubq->canceling as
soon as each individual queue becomes ready, rather than waiting for all
queues. This allows subsequent UBLK_U_IO_FETCH_IO_CMDS commands to succeed
during recovery.
Changes:
- Add ubq->nr_io_ready to track I/Os ready per queue
- Add ub->nr_queue_ready to track number of ready queues
- Add ublk_queue_ready() helper to check queue readiness
- Redefine ublk_dev_ready() based on queue count instead of I/O count
- Clear ubq->canceling immediately when queue becomes ready
- Add ublk_queue_reset_io_flags() to reset per-queue flags
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:46 +0000 (22:18 +0800)]
ublk: implement batch request completion via blk_mq_end_request_batch()
Reduce overhead when completing multiple requests in batch I/O mode by
accumulating them in an io_comp_batch structure and completing them
together via blk_mq_end_request_batch(). This minimizes per-request
completion overhead and improves performance for high IOPS workloads.
The implementation adds an io_comp_batch pointer to struct ublk_io and
initializes it in __ublk_fetch(). For batch I/O, the pointer is set to
the batch structure in ublk_batch_commit_io(). The __ublk_complete_rq()
function uses io->iob to call blk_mq_add_to_batch() for batch mode.
After processing all batch I/Os, the completion callback is invoked in
ublk_handle_batch_commit_cmd() to complete all accumulated requests
efficiently.
So far just covers direct completion. For deferred completion(zero copy,
auto buffer reg), ublk_io_release() is often delayed in freeing buffer
consumer io_uring request's code path, so this patch often doesn't work,
also it is hard to pass the per-task 'struct io_comp_batch' for deferred
completion.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:44 +0000 (22:18 +0800)]
ublk: add new feature UBLK_F_BATCH_IO
Add new feature UBLK_F_BATCH_IO which replaces the following two
per-io commands:
- UBLK_U_IO_FETCH_REQ
- UBLK_U_IO_COMMIT_AND_FETCH_REQ
with three per-queue batch io uring_cmd:
- UBLK_U_IO_PREP_IO_CMDS
- UBLK_U_IO_COMMIT_IO_CMDS
- UBLK_U_IO_FETCH_IO_CMDS
Then ublk can deliver batch io commands to ublk server in single
multishort uring_cmd, also allows to prepare & commit multiple
commands in batch style via single uring_cmd, communication cost is
reduced a lot.
This feature also doesn't limit task context any more for all supported
commands, so any allowed uring_cmd can be issued in any task context.
ublk server implementation becomes much easier.
Meantime load balance becomes much easier to support with this feature.
The command `UBLK_U_IO_FETCH_IO_CMDS` can be issued from multiple task
contexts, so each task can adjust this command's buffer length or number
of inflight commands for controlling how much load is handled by current
task.
Later, priority parameter will be added to command `UBLK_U_IO_FETCH_IO_CMDS`
for improving load balance support.
UBLK_U_IO_NEED_GET_DATA isn't supported in batch io yet, but it may be
enabled in future via its batch pair.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:43 +0000 (22:18 +0800)]
ublk: abort requests filled in event kfifo
In case of BATCH_IO, any request filled in event kfifo, they don't get
chance to be dispatched any more when releasing ublk char device, so
we have to abort them too.
Add ublk_abort_batch_queue() for aborting this kind of requests.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:42 +0000 (22:18 +0800)]
ublk: refactor ublk_queue_rq() and add ublk_batch_queue_rq()
Extract common request preparation and cancellation logic into
__ublk_queue_rq_common() helper function. Add dedicated
ublk_batch_queue_rq() for batch mode operations to eliminate runtime check
in ublk_queue_rq().
Signed-off-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:41 +0000 (22:18 +0800)]
ublk: add UBLK_U_IO_FETCH_IO_CMDS for batch I/O processing
Add UBLK_U_IO_FETCH_IO_CMDS command to enable efficient batch processing
of I/O requests. This multishot uring_cmd allows the ublk server to fetch
multiple I/O commands in a single operation, significantly reducing
submission overhead compared to individual FETCH_REQ* commands.
Key Design Features:
1. Multishot Operation: One UBLK_U_IO_FETCH_IO_CMDS can fetch many I/O
commands, with the batch size limited by the provided buffer length.
2. Dynamic Load Balancing: Multiple fetch commands can be submitted
simultaneously, but only one is active at any time. This enables
efficient load distribution across multiple server task contexts.
3. Implicit State Management: The implementation uses three key variables
to track state:
- evts_fifo: Queue of request tags awaiting processing
- fcmd_head: List of available fetch commands
- active_fcmd: Currently active fetch command (NULL = none active)
States are derived implicitly:
- IDLE: No fetch commands available
- READY: Fetch commands available, none active
- ACTIVE: One fetch command processing events
4. Lockless Reader Optimization: The active fetch command can read from
evts_fifo without locking (single reader guarantee), while writers
(ublk_queue_rq/ublk_queue_rqs) use evts_lock protection. The memory
barrier pairing plays key role for the single lockless reader
optimization.
Implementation Details:
- ublk_queue_rq() and ublk_queue_rqs() save request tags to evts_fifo
- __ublk_acquire_fcmd() selects an available fetch command when
events arrive and no command is currently active
- ublk_batch_dispatch() moves tags from evts_fifo to the fetch command's
buffer and posts completion via io_uring_mshot_cmd_post_cqe()
- State transitions are coordinated via evts_lock to maintain consistency
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:40 +0000 (22:18 +0800)]
ublk: add batch I/O dispatch infrastructure
Add infrastructure for delivering I/O commands to ublk server in batches,
preparing for the upcoming UBLK_U_IO_FETCH_IO_CMDS feature.
Key components:
- struct ublk_batch_fetch_cmd: Represents a batch fetch uring_cmd that will
receive multiple I/O tags in a single operation, using io_uring's
multishot command for efficient ublk IO delivery.
- ublk_batch_dispatch(): Batch version of ublk_dispatch_req() that:
* Pulls multiple request tags from the events FIFO (lock-free reader)
* Prepares each I/O for delivery (including auto buffer registration)
* Delivers tags to userspace via single uring_cmd notification
* Handles partial failures by restoring undelivered tags to FIFO
The batch approach significantly reduces notification overhead by aggregating
multiple I/O completions into single uring_cmd, while maintaining the same
I/O processing semantics as individual operations.
Error handling ensures system consistency: if buffer selection or CQE
posting fails, undelivered tags are restored to the FIFO for retry,
meantime IO state has to be restored.
This runs in task work context, scheduled via io_uring_cmd_complete_in_task()
or called directly from ->uring_cmd(), enabling efficient batch processing
without blocking the I/O submission path.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:39 +0000 (22:18 +0800)]
ublk: add io events fifo structure
Add ublk io events fifo structure and prepare for supporting command
batch, which will use io_uring multishot uring_cmd for fetching one
batch of io commands each time.
One nice feature of kfifo is to allow multiple producer vs single
consumer. We just need lock the producer side, meantime the single
consumer can be lockless.
The producer is actually from ublk_queue_rq() or ublk_queue_rqs(), so
lock contention can be eased by setting proper blk-mq nr_queues.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:37 +0000 (22:18 +0800)]
ublk: handle UBLK_U_IO_PREP_IO_CMDS
This commit implements the handling of the UBLK_U_IO_PREP_IO_CMDS command,
which allows userspace to prepare a batch of I/O requests.
The core of this change is the `ublk_walk_cmd_buf` function, which iterates
over the elements in the uring_cmd fixed buffer. For each element, it parses
the I/O details, finds the corresponding `ublk_io` structure, and prepares it
for future dispatch.
Add per-io lock for protecting concurrent delivery and committing.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 14:18:34 +0000 (22:18 +0800)]
ublk: define ublk_ch_batch_io_fops for the coming feature F_BATCH_IO
Introduces the basic structure for a batched I/O feature in the ublk driver.
It adds placeholder functions and a new file operations structure,
ublk_ch_batch_io_fops, which will be used for fetching and committing I/O
commands in batches. Currently, the feature is disabled.
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 16 Jan 2026 07:46:38 +0000 (15:46 +0800)]
nvme/io_uring: optimize IOPOLL completions for local ring context
When multiple io_uring rings poll on the same NVMe queue, one ring can
find completions belonging to another ring. The current code always
uses task_work to handle this, but this adds overhead for the common
single-ring case.
This patch passes the polling io_ring_ctx through io_comp_batch's new
poll_ctx field. In io_do_iopoll(), the polling ring's context is stored
in iob.poll_ctx before calling the iopoll callbacks.
In nvme_uring_cmd_end_io(), we now compare iob->poll_ctx with the
request's owning io_ring_ctx (via io_uring_cmd_ctx_handle()). If they
match (local context), we complete inline with io_uring_cmd_done32().
If they differ (remote context) or iob is NULL (non-iopoll path), we
use task_work as before.
This optimization eliminates task_work scheduling overhead for the
common case where a ring polls and finds its own completions.
~10% IOPS improvement is observed in the following benchmark:
Ming Lei [Fri, 16 Jan 2026 07:46:37 +0000 (15:46 +0800)]
block: pass io_comp_batch to rq_end_io_fn callback
Add a third parameter 'const struct io_comp_batch *' to the rq_end_io_fn
callback signature. This allows end_io handlers to access the completion
batch context when requests are completed via blk_mq_end_request_batch().
The io_comp_batch is passed from blk_mq_end_request_batch(), while NULL
is passed from __blk_mq_end_request() and blk_mq_put_rq_ref() which don't
have batch context.
This infrastructure change enables drivers to detect whether they're
being called from a batched completion path (like iopoll) and access
additional context stored in the io_comp_batch.
Damien Le Moal [Tue, 6 Jan 2026 07:00:56 +0000 (16:00 +0900)]
block: fix blk_zone_cond_str() comment
Fix the comment for blk_zone_cond_str() by replacing the meaningless
BLK_ZONE_ZONE_XXX comment with the correct BLK_ZONE_COND_name, thus also
replacing the XXX with what that actually means.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Mon, 12 Jan 2026 22:05:02 +0000 (00:05 +0200)]
selftests: ublk: add stop command with --safe option
Add 'stop' subcommand to kublk utility that uses the new
UBLK_CMD_TRY_STOP_DEV command when --safe option is specified.
This allows stopping a device only if it has no active openers,
returning -EBUSY otherwise.
Also add test_generic_16.sh to test the new functionality.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Yoav Cohen [Mon, 12 Jan 2026 22:05:01 +0000 (00:05 +0200)]
ublk: add UBLK_CMD_TRY_STOP_DEV command
Add a best-effort stop command, UBLK_CMD_TRY_STOP_DEV, which only stops a
ublk device when it has no active openers.
Unlike UBLK_CMD_STOP_DEV, this command does not disrupt existing users.
New opens are blocked only after disk_openers has reached zero; if the
device is busy, the command returns -EBUSY and leaves it running.
The ub->block_open flag is used only to close a race with an in-progress
open and does not otherwise change open behavior.
Advertise support via the UBLK_F_SAFE_STOP_DEV feature flag.
Signed-off-by: Yoav Cohen <yoav@nvidia.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add test case loop_08 to verify the ublk integrity data flow. It uses
the kublk loop target to create a ublk device with integrity on top of
backing data and integrity files. It then writes to the whole device
with fio configured to generate integrity data. Then it reads back the
whole device with fio configured to verify the integrity data.
It also verifies that injected guard, reftag, and apptag corruptions are
correctly detected.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add test case null_04 to exercise all the different integrity params. It
creates 4 different ublk devices with different combinations of
integrity arguments and verifies their integrity limits via sysfs and
the metadata_size utility.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
selftests: ublk: add integrity data support to loop target
To perform and end-to-end test of integrity information through a ublk
device, we need to actually store it somewhere and retrieve it. Add this
support to kublk's loop target. It uses a second backing file for the
integrity data corresponding to the data stored in the first file.
The integrity file is initialized with byte 0xFF, which ensures the app
and reference tags are set to the "escape" pattern to disable the
bio-integrity-auto guard and reftag checks until the blocks are written.
The integrity file is opened without O_DIRECT since it will be accessed
at sub-block granularity. Each incoming read/write results in a pair of
reads/writes, one to the data file, and one to the integrity file. If
either backing I/O fails, the error is propagated to the ublk request.
If both backing I/Os read/write some bytes, the ublk request is
completed with the smaller of the number of blocks accessed by each I/O.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
selftests: ublk: support non-O_DIRECT backing files
A subsequent commit will add support for using a backing file to store
integrity data. Since integrity data is accessed in intervals of
metadata_size, which may be much smaller than a logical block on the
backing device, direct I/O cannot be used. Add an argument to
backing_file_tgt_init() to specify the number of files to open for
direct I/O. The remaining files will use buffered I/O. For now, continue
to request direct I/O for all the files.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
selftests: ublk: implement integrity user copy in kublk
If integrity data is enabled for kublk, allocate an integrity buffer for
each I/O. Extend ublk_user_copy() to copy the integrity data between the
ublk request and the integrity buffer if the ublksrv_io_desc indicates
that the request has integrity data.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
selftests: ublk: add kublk support for integrity params
Add integrity param command line arguments to kublk. Plumb these to
struct ublk_params for the null and fault_inject targets, as they don't
need to actually read or write the integrity data. Forbid the integrity
params for loop or stripe until the integrity data copy is implemented.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
selftests: ublk: add utility to get block device metadata size
Some block device integrity parameters are available in sysfs, but
others are only accessible using the FS_IOC_GETLBMD_CAP ioctl. Add a
metadata_size utility program to print out the logical block metadata
size, PI offset, and PI size within the metadata. Example output:
$ metadata_size /dev/ublkb0
metadata_size: 64
pi_offset: 56
pi_tuple_size: 8
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
ublk user copy syscalls may be issued from any task, so they take a
reference count on the struct ublk_io to check whether it is owned by
the ublk server and prevent a concurrent UBLK_IO_COMMIT_AND_FETCH_REQ
from completing the request. However, if the user copy syscall is issued
on the io's daemon task, a concurrent UBLK_IO_COMMIT_AND_FETCH_REQ isn't
possible, so the atomic reference count dance is unnecessary. Check for
UBLK_IO_FLAG_OWNED_BY_SRV to ensure the request is dispatched to the
sever and obtain the request from ublk_io's req field instead of looking
it up on the tagset. Skip the reference count increment and decrement.
Commit 8a8fe42d765b ("ublk: optimize UBLK_IO_REGISTER_IO_BUF on daemon
task") made an analogous optimization for ublk zero copy buffer
registration.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stanley Zhang [Thu, 8 Jan 2026 09:19:38 +0000 (02:19 -0700)]
ublk: support UBLK_F_INTEGRITY
Now that all the components of the ublk integrity feature have been
implemented, add UBLK_F_INTEGRITY to UBLK_F_ALL, conditional on block
layer integrity support (CONFIG_BLK_DEV_INTEGRITY). This allows ublk
servers to create ublk devices with UBLK_F_INTEGRITY set and
UBLK_U_CMD_GET_FEATURES to report the feature as supported.
Signed-off-by: Stanley Zhang <stazhang@purestorage.com>
[csander: make feature conditional on CONFIG_BLK_DEV_INTEGRITY] Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stanley Zhang [Thu, 8 Jan 2026 09:19:37 +0000 (02:19 -0700)]
ublk: implement integrity user copy
Add a function ublk_copy_user_integrity() to copy integrity information
between a request and a user iov_iter. This mirrors the existing
ublk_copy_user_pages() but operates on request integrity data instead of
regular data. Check UBLKSRV_IO_INTEGRITY_FLAG in iocb->ki_pos in
ublk_user_copy() to choose between copying data or integrity data.
[csander: change offset units from data bytes to integrity data bytes,
fix CONFIG_BLK_DEV_INTEGRITY=n build, rebase on user copy refactor]
Signed-off-by: Stanley Zhang <stazhang@purestorage.com> Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
ublk: move offset check out of __ublk_check_and_get_req()
__ublk_check_and_get_req() checks that the passed in offset is within
the data length of the specified ublk request. However, only user copy
(ublk_check_and_get_req()) supports accessing ublk request data at a
nonzero offset. Zero-copy buffer registration (ublk_register_io_buf())
always passes 0 for the offset, so the check is unnecessary. Move the
check from __ublk_check_and_get_req() to ublk_check_and_get_req().
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
ublk: inline ublk_check_and_get_req() into ublk_user_copy()
ublk_check_and_get_req() has a single callsite in ublk_user_copy(). It
takes a ton of arguments in order to pass local variables from
ublk_user_copy() to ublk_check_and_get_req() and vice versa. And more
are about to be added. Combine the functions to reduce the argument
passing noise.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
ublk_ch_read_iter() and ublk_ch_write_iter() are nearly identical except
for the iter direction. Split out a helper function ublk_user_copy() to
reduce the code duplication as these functions are about to get larger.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stanley Zhang [Thu, 8 Jan 2026 09:19:31 +0000 (02:19 -0700)]
ublk: support UBLK_PARAM_TYPE_INTEGRITY in device creation
Add a feature flag UBLK_F_INTEGRITY for a ublk server to request
integrity/metadata support when creating a ublk device. The ublk server
can also check for the feature flag on the created device or the result
of UBLK_U_CMD_GET_FEATURES to tell if the ublk driver supports it.
UBLK_F_INTEGRITY requires UBLK_F_USER_COPY, as user copy is the only
data copy mode initially supported for integrity data.
Add UBLK_PARAM_TYPE_INTEGRITY and struct ublk_param_integrity to struct
ublk_params to specify the integrity params of a ublk device.
UBLK_PARAM_TYPE_INTEGRITY requires UBLK_F_INTEGRITY and a nonzero
metadata_size. The LBMD_PI_CAP_* and LBMD_PI_CSUM_* values from the
linux/fs.h UAPI header are used for the flags and csum_type fields.
If the UBLK_PARAM_TYPE_INTEGRITY flag is set, validate the integrity
parameters and apply them to the blk_integrity limits.
The struct ublk_param_integrity validations are based on the checks in
blk_validate_integrity_limits(). Any invalid parameters should be
rejected before being applied to struct blk_integrity.
[csander: drop redundant pi_tuple_size field, use block metadata UAPI
constants, add param validation]
Signed-off-by: Stanley Zhang <stazhang@purestorage.com> Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
ublk_dev_support_user_copy() will be used in ublk_validate_params().
Move these functions next to ublk_{dev,queue}_is_zoned() to avoid
needing to forward-declare them.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk-integrity: take const pointer in blk_integrity_rq()
blk_integrity_rq() doesn't modify the struct request passed in, so allow
a const pointer to be passed. Use a matching signature for the
!CONFIG_BLK_DEV_INTEGRITY version.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Sun, 11 Jan 2026 20:16:36 +0000 (13:16 -0700)]
Merge branch 'block-6.19' into for-7.0/block
Merge in fixes that went to 6.19 after for-7.0/block was branched.
Pending ublk changes depend on particularly the async scan work.
* block-6.19:
block: zero non-PI portion of auto integrity buffer
ublk: fix use-after-free in ublk_partition_scan_work
blk-mq: avoid stall during boot due to synchronize_rcu_expedited
loop: add missing bd_abort_claiming in loop_set_status
block: don't merge bios with different app_tags
blk-rq-qos: Remove unlikely() hints from QoS checks
loop: don't change loop device under exclusive opener in loop_set_status
block, bfq: update outdated comment
blk-mq: skip CPU offline notify on unmapped hctx
selftests/ublk: fix Makefile to rebuild on header changes
selftests/ublk: add test for async partition scan
ublk: scan partition in async way
block,bfq: fix aux stat accumulation destination
md: Fix forward incompatibility from configurable logical block size
md: Fix logical_block_size configuration being overwritten
md: suspend array while updating raid_disks via sysfs
md/raid5: fix possible null-pointer dereferences in raid5_store_group_thread_cnt()
md: Fix static checker warning in analyze_sbs
blk-crypto: handle the fallback above the block layer
Add a blk_crypto_submit_bio helper that either submits the bio when
it is not encrypted or inline encryption is provided, but otherwise
handles the encryption before going down into the low-level driver.
This reduces the risk from bio reordering and keeps memory allocation
as high up in the stack as possible.
Note that if the submitter knows that inline enctryption is known to
be supported by the underyling driver, it can still use plain
submit_bio.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Avoid the relatively high overhead of constructing and walking per-page
segment bio_vecs for data unit alignment checking by merging the checks
into existing loops.
For hardware support crypto, perform the check in bio_split_io_at, which
already contains a similar alignment check applied for all I/O. This
means bio-based drivers that do not call bio_split_to_limits, should they
ever grow blk-crypto support, need to implement the check themselves,
just like all other queue limits checks.
For blk-crypto-fallback do it in the encryption/decryption loops. This
means alignment errors for decryption will only be detected after I/O
has completed, but that seems like a worthwhile trade off.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk-crypto: use mempool_alloc_bulk for encrypted bio page allocation
Calling mempool_alloc in a loop is not safe unless the maximum allocation
size times the maximum number of threads using it is less than the
minimum pool size. Use the new mempool_alloc_bulk helper to allocate
all missing elements in one pass to remove this deadlock risk. This
also means that non-pool allocations now use alloc_pages_bulk which can
be significantly faster than a loop over individual page allocations.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk-crypto: use on-stack skcipher requests for fallback en/decryption
Allocating a skcipher request dynamically can deadlock or cause
unexpected I/O failures when called from writeback context. Avoid the
allocation entirely by using on-stack skciphers, similar to what the
non-blk-crypto fscrypt path already does.
This drops the incomplete support for asynchronous algorithms, which
previously could be used, but only synchronously.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk-crypto: optimize bio splitting in blk_crypto_fallback_encrypt_bio
The current code in blk_crypto_fallback_encrypt_bio is inefficient and
prone to deadlocks under memory pressure: It first walks the passed in
plaintext bio to see how much of it can fit into a single encrypted
bio using up to BIO_MAX_VEC PAGE_SIZE segments, and then allocates a
plaintext clone that fits the size, only to allocate another bio for
the ciphertext later. While the plaintext clone uses a bioset to avoid
deadlocks when allocations could fail, the ciphertex one uses bio_kmalloc
which is a no-go in the file system I/O path.
Switch blk_crypto_fallback_encrypt_bio to walk the source plaintext bio
while consuming bi_iter without cloning it, and instead allocate a
ciphertext bio at the beginning and whenever we fille up the previous
one. The existing bio_set for the plaintext clones is reused for the
ciphertext bios to remove the deadlock risk.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk-crypto: submit the encrypted bio in blk_crypto_fallback_bio_prep
Restructure blk_crypto_fallback_bio_prep so that it always submits the
encrypted bio instead of passing it back to the caller, which allows
to simplify the calling conventions for blk_crypto_fallback_bio_prep and
blk_crypto_bio_prep so that they never have to return a bio, and can
use a true return value to indicate that the caller should submit the
bio, and false that the blk-crypto code consumed it.
The submission is handled by the on-stack bio list in the current
task_struct by the block layer and does not cause additional stack
usage or major overhead. It also prepares for the following optimization
and fixes for the blk-crypto fallback write path.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
fscrypt: keep multiple bios in flight in fscrypt_zeroout_range_inline_crypt
This should slightly improve performance for large zeroing operations,
but more importantly prepares for blk-crypto refactoring that requires
all fscrypt users to call submit_bio directly.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
fscrypt: pass a real sector_t to fscrypt_zeroout_range_inline_crypt
While the pblk argument to fscrypt_zeroout_range_inline_crypt is
declared as a sector_t it actually is interpreted as a logical block
size unit, which is highly unusual. Switch to passing the 512 byte
units that sector_t is defined for.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Sat, 10 Jan 2026 13:42:36 +0000 (21:42 +0800)]
block: account for bi_bvec_done in bio_may_need_split()
When checking if a bio fits in a single segment, bio_may_need_split()
compares bi_size against the current bvec's bv_len. However, for
partially consumed bvecs (bi_bvec_done > 0), such as in cloned or
split bios, the remaining bytes in the current bvec is actually
(bv_len - bi_bvec_done), not bv_len.
This could cause bio_may_need_split() to incorrectly return false,
leading to nr_phys_segments being set to 1 when the bio actually
spans multiple segments. This triggers the WARN_ON in __blk_rq_map_sg()
when the actual mapped segments exceed the expected count.
Fix by subtracting bi_bvec_done from bv_len in the comparison.
Reported-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Close: https://lore.kernel.org/linux-block/9687cf2b-1f32-44e1-b58d-2492dc6e7185@linux.ibm.com/ Repored-and-bisected-by: Christoph Hellwig <hch@infradead.org> Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com> Tested-by: Christoph Hellwig <hch@infradead.org> Fixes: ee623c892aa5 ("block: use bvec iterator helper for bio_may_need_split()") Cc: Nitesh Shetty <nj.shetty@samsung.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
bi_offload_capable() returns whether a block device's metadata size
matches its PI tuple size. Use pi_tuple_size instead of switching on
csum_type. This makes the code considerably simpler and less branchy.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Anuj Gupta <anuj20.g@samsung.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
block: zero non-PI portion of auto integrity buffer
The auto-generated integrity buffer for writes needs to be fully
initialized before being passed to the underlying block device,
otherwise the uninitialized memory can be read back by userspace or
anyone with physical access to the storage device. If protection
information is generated, that portion of the integrity buffer is
already initialized. The integrity data is also zeroed if PI generation
is disabled via sysfs or the PI tuple size is 0. However, this misses
the case where PI is generated and the PI tuple size is nonzero, but the
metadata size is larger than the PI tuple. In this case, the remainder
("opaque") of the metadata is left uninitialized.
Generalize the BLK_INTEGRITY_CSUM_NONE check to cover any case when the
metadata is larger than just the PI tuple.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Fixes: c546d6f43833 ("block: only zero non-PI metadata tuples in bio_integrity_prep") Reviewed-by: Anuj Gupta <anuj20.g@samsung.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Fri, 9 Jan 2026 12:14:54 +0000 (20:14 +0800)]
ublk: fix use-after-free in ublk_partition_scan_work
A race condition exists between the async partition scan work and device
teardown that can lead to a use-after-free of ub->ub_disk:
1. ublk_ctrl_start_dev() schedules partition_scan_work after add_disk()
2. ublk_stop_dev() calls ublk_stop_dev_unlocked() which does:
- del_gendisk(ub->ub_disk)
- ublk_detach_disk() sets ub->ub_disk = NULL
- put_disk() which may free the disk
3. The worker ublk_partition_scan_work() then dereferences ub->ub_disk
leading to UAF
Fix this by using ublk_get_disk()/ublk_put_disk() in the worker to hold
a reference to the disk during the partition scan. The spinlock in
ublk_get_disk() synchronizes with ublk_detach_disk() ensuring the worker
either gets a valid reference or sees NULL and exits early.
Also change flush_work() to cancel_work_sync() to avoid running the
partition scan work unnecessarily when the disk is already detached.
Fixes: 7fc4da6a304b ("ublk: scan partition in async way") Reported-by: Ruikai Peng <ruikai@pwno.io> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Wed, 31 Dec 2025 03:00:57 +0000 (11:00 +0800)]
io_uring: remove nr_segs recalculation in io_import_kbuf()
io_import_kbuf() recalculates iter->nr_segs to reflect only the bvecs
needed for the requested byte range. This was added to provide an
accurate segment count to bio_iov_bvec_set(), which copied nr_segs to
bio->bi_vcnt for use as a bio split hint.
The previous two patches eliminated this dependency:
- bio_may_need_split() now uses bi_iter instead of bi_vcnt for split
decisions
- bio_iov_bvec_set() no longer copies nr_segs to bi_vcnt
Since nr_segs is no longer used for bio split decisions, the
recalculation loop is unnecessary. The iov_iter already has the correct
bi_size to cap iteration, so an oversized nr_segs is harmless.
Link: https://lkml.org/lkml/2025/4/16/351 Signed-off-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Nitesh Shetty <nj.shetty@samsung.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Wed, 31 Dec 2025 03:00:56 +0000 (11:00 +0800)]
block: don't initialize bi_vcnt for cloned bio in bio_iov_bvec_set()
bio_iov_bvec_set() creates a cloned bio that borrows a bvec array from
an iov_iter. For cloned bios, bi_vcnt is meaningless because iteration
is controlled entirely by bi_iter (bi_idx, bi_size, bi_bvec_done), not
by bi_vcnt. Remove the incorrect bi_vcnt assignment.
Explicitly initialize bi_iter.bi_idx to 0 to ensure iteration starts
at the first bvec. While bi_idx is typically already zero from bio
initialization, making this explicit improves clarity and correctness.
This change also avoids accessing iter->nr_segs, which is an iov_iter
implementation detail that block code should not depend on.
Signed-off-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Nitesh Shetty <nj.shetty@samsung.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ming Lei [Wed, 31 Dec 2025 03:00:55 +0000 (11:00 +0800)]
block: use bvec iterator helper for bio_may_need_split()
bio_may_need_split() uses bi_vcnt to determine if a bio has a single
segment, but bi_vcnt is unreliable for cloned bios. Cloned bios share
the parent's bi_io_vec array but iterate over a subset via bi_iter,
so bi_vcnt may not reflect the actual segment count being iterated.
Replace the bi_vcnt check with bvec iterator access via
__bvec_iter_bvec(), comparing bi_iter.bi_size against the current
bvec's length. This correctly handles both cloned and non-cloned bios.
Move bi_io_vec into the first cache line adjacent to bi_iter. This is
a sensible layout since bi_io_vec and bi_iter are commonly accessed
together throughout the block layer - every bvec iteration requires
both fields. This displaces bi_end_io to the second cache line, which
is acceptable since bi_end_io and bi_private are always fetched
together in bio_endio() anyway.
The struct layout change requires bio_reset() to preserve and restore
bi_io_vec across the memset, since it now falls within BIO_RESET_BYTES.
Nitesh verified that this patch doesn't regress NVMe 512-byte IO perf [1].
Tetsuo Handa [Wed, 7 Jan 2026 10:41:43 +0000 (19:41 +0900)]
loop: add missing bd_abort_claiming in loop_set_status
Commit 08e136ebd193 ("loop: don't change loop device under exclusive
opener in loop_set_status") forgot to call bd_abort_claiming() when
mutex_lock_killable() failed.
Fixes: 08e136ebd193 ("loop: don't change loop device under exclusive opener in loop_set_status") Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Jens Axboe <axboe@kernel.dk>
nvme_set_app_tag() uses the app_tag value from the bio_integrity_payload
of the struct request's first bio. This assumes all the request's bios
have the same app_tag. However, it is possible for bios with different
app_tag values to be merged into a single request.
Add a check in blk_integrity_merge_{bio,rq}() to prevent the merging of
bios/requests with different app_tag values if BIP_CHECK_APPTAG is set.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Fixes: 3d8b5a22d404 ("block: add support to pass user meta buffer") Signed-off-by: Jens Axboe <axboe@kernel.dk>
Breno Leitao [Tue, 6 Jan 2026 14:26:57 +0000 (06:26 -0800)]
blk-rq-qos: Remove unlikely() hints from QoS checks
The unlikely() annotations on QUEUE_FLAG_QOS_ENABLED checks are
counterproductive. Writeback throttling (WBT) might be enabled by
default, mainly because CONFIG_BLK_WBT_MQ defaults to 'y'.
Branch profiling on Meta servers, which have WBT enabled, confirms 100%
misprediction rates on these checks.
Remove the unlikely() annotations to let the CPU's branch predictor
learn the actual behavior, potentially improving I/O path performance.
Leon Romanovsky [Wed, 17 Dec 2025 09:41:24 +0000 (11:41 +0200)]
types: move phys_vec definition to common header
Move the struct phys_vec definition from block/blk-mq-dma.c to
include/linux/types.h to make it available for use across the kernel.
The phys_vec structure represents a physical address range with a
length, which is used by the new physical address-based DMA mapping
API. This structure is already used by the block layer and will be
needed for DMA phys API users.
Moving this definition to types.h provides a centralized location
for this common data structure and eliminates code duplication
across subsystems that need to work with physical address ranges.
Leon Romanovsky [Wed, 17 Dec 2025 09:41:23 +0000 (11:41 +0200)]
nvme-pci: Use size_t for length fields to handle larger sizes
This patch changes the length variables from unsigned int to size_t.
Using size_t ensures that we can handle larger sizes, as size_t is
always equal to or larger than the previously used u32 type.
Originally, u32 was used because blk-mq-dma code evolved from
scatter-gather implementation, which uses unsigned int to describe length.
This change will also allow us to reuse the existing struct phys_vec in places
that don't need scatter-gather.
loop: don't change loop device under exclusive opener in loop_set_status
loop_set_status() is allowed to change the loop device while there
are other openers of the device, even exclusive ones.
In this case, it causes a KASAN: slab-out-of-bounds Read in
ext4_search_dir(), since when looking for an entry in an inlined
directory, e_value_offs is changed underneath the filesystem by
loop_set_status().
Fix the problem by forbidding loop_set_status() from modifying the loop
device while there are exclusive openers of the device. This is similar
to the fix in loop_configure() by commit 33ec3e53e7b1 ("loop: Don't
change loop device under exclusive opener") alongside commit ecbe6bc0003b
("block: use bd_prepare_to_claim directly in the loop driver").
Reported-by: syzbot+3ee481e21fd75e14c397@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=3ee481e21fd75e14c397 Tested-by: syzbot+3ee481e21fd75e14c397@syzkaller.appspotmail.com Tested-by: Yongpeng Yang <yangyongpeng@xiaomi.com> Signed-off-by: Raphael Pinsonneault-Thibeault <rpthibeault@gmail.com> Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Md Haris Iqbal [Fri, 5 Dec 2025 12:47:33 +0000 (13:47 +0100)]
rnbd-srv: Zero the rsp buffer before using it
Before using the data buffer to send back the response message, zero it
completely. This prevents any stray bytes to be picked up by the client
side when there the message is exchanged between different protocol
versions.
Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Grzegorz Prajsner <grzegorz.prajsner@ionos.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
rnbd-srv: Fix server side setting of bi_size for special IOs
On rnbd-srv, the bi_size of the bio is set during the bio_add_page
function, to which datalen is passed. But for special IOs like DISCARD
and WRITE_ZEROES, datalen is 0, since there is no data to write. For
these special IOs, use the bi_size of the rnbd_msg_io.
Fixes: f6f84be089c9 ("block/rnbd-srv: Add sanity check and remove redundant assignment") Signed-off-by: Florian-Ewald Mueller <florian-ewald.mueller@ionos.com> Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Grzegorz Prajsner <grzegorz.prajsner@ionos.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jack Wang [Fri, 5 Dec 2025 12:47:31 +0000 (13:47 +0100)]
rnbd-srv: fix the trace format for flags
The __print_flags helper meant for bitmask, while the rnbd_rw_flags is
mixed with bitmask and enum, to avoid confusion, just print the data
as it is.
Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Grzegorz Prajsner <grzegorz.prajsner@ionos.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Md Haris Iqbal [Fri, 5 Dec 2025 12:47:30 +0000 (13:47 +0100)]
block/rnbd-proto: Check and retain the NOUNMAP flag for requests
The NOUNMAP flag is in combination with WRITE_ZEROES flag to indicate
that the upper layers wants the sectors zeroed, but does not want it to
get freed. This instruction is especially important for storage stacks
which involves a layer capable of thin provisioning.
This commit makes RNBD block device transfer and retain this NOUNMAP flag
for requests, so it can be passed onto the backend device on the server
side.
Since it is a change in the wire protocol, bump the minor version of
protocol.
Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Grzegorz Prajsner <grzegorz.prajsner@ionos.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Md Haris Iqbal [Fri, 5 Dec 2025 12:47:28 +0000 (13:47 +0100)]
block/rnbd-proto: Handle PREFLUSH flag properly for IOs
In RNBD client, for a WRITE request of size 0, with only the REQ_PREFLUSH
bit set, while converting from bio_opf to rnbd_opf, we do REQ_OP_WRITE to
RNBD_OP_WRITE, and then check if the rq is flush through function
op_is_flush. That function checks both REQ_PREFLUSH and REQ_FUA flag, and
if any of them is set, the RNBD_F_FUA is set.
On the RNBD server side, while converting the RNBD flags to req flags, if
the RNBD_F_FUA flag is set, we just set the REQ_FUA flag. This means we
have lost the PREFLUSH flag, and added the REQ_FUA flag in its place.
This commits adds a new RNBD_F_PREFLUSH flag, and also adds separate
handling for REQ_PREFLUSH flag. On the server side, if the RNBD_F_PREFLUSH
is present, the REQ_PREFLUSH is added to the bio.
Since it is a change in the wire protocol, bump the minor version of
protocol.
The change is backwards compatible, and does not change the functionality
if either the client or the server is running older/newer versions.
If the client side is running the older version, both REQ_PREFLUSH and
REQ_FUA is converted to RNBD_F_FUA. The server running newer one would
still add only the REQ_FUA flag which is what happens when both client and
server is running the older version.
If the client side is running the newer version, just like before a
RNBD_F_FUA is added, but now a RNBD_F_PREFLUSH is also added to the
rnbd_opf. In case the server is running the older version the
RNBD_F_PREFLUSH is ignored, and only the RNBD_F_FUA is processed.
Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Reviewed-by: Jack Wang <jinpu.wang@ionos.com> Reviewed-by: Florian-Ewald Mueller <florian-ewald.mueller@ionos.com> Signed-off-by: Grzegorz Prajsner <grzegorz.prajsner@ionos.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Julia Lawall [Wed, 31 Dec 2025 17:22:07 +0000 (18:22 +0100)]
block, bfq: update outdated comment
The function bfq_bfqq_may_idle() was renamed as bfq_better_to_idle()
in commit 277a4a9b56cd ("block, bfq: give a better name to
bfq_bfqq_may_idle"). Update the comment accordingly.
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 31 Dec 2025 13:55:07 +0000 (06:55 -0700)]
Merge tag 'md-6.19-20251231' of gitolite.kernel.org:pub/scm/linux/kernel/git/mdraid/linux into block-6.19
Pull MD fixes from Yu Kuai:
"- Fix null-pointer dereference in raid5 sysfs group_thread_cnt store
(Tuo Li)
- Fix possible mempool corruption during raid1 raid_disks update via
sysfs (FengWei Shih)
- Fix logical_block_size configuration being overwritten during
super_1_validate() (Li Nan)
- Fix forward incompatibility with configurable logical block size:
arrays assembled on new kernels could not be assembled on kernels
<=6.18 due to non-zero reserved pad rejection (Li Nan)
- Fix static checker warning about iterator not incremented (Li Nan)"
* tag 'md-6.19-20251231' of gitolite.kernel.org:pub/scm/linux/kernel/git/mdraid/linux:
md: Fix forward incompatibility from configurable logical block size
md: Fix logical_block_size configuration being overwritten
md: suspend array while updating raid_disks via sysfs
md/raid5: fix possible null-pointer dereferences in raid5_store_group_thread_cnt()
md: Fix static checker warning in analyze_sbs
Cong Zhang [Tue, 30 Dec 2025 09:17:05 +0000 (17:17 +0800)]
blk-mq: skip CPU offline notify on unmapped hctx
If an hctx has no software ctx mapped, blk_mq_map_swqueue() never
allocates tags and leaves hctx->tags NULL. The CPU hotplug offline
notifier can still run for that hctx, return early since hctx cannot
hold any requests.
Signed-off-by: Cong Zhang <cong.zhang@oss.qualcomm.com> Fixes: bf0beec0607d ("blk-mq: drain I/O when all CPUs in a hctx are offline") Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
null_blk: Constify struct configfs_item_operations and configfs_group_operations
'struct configfs_item_operations' and 'configfs_group_operations' are not
modified in this driver.
Constifying these structures moves some data to a read-only section, so
increases overall security, especially when the structure holds some
function pointers.
On a x86_64, with allmodconfig:
Before:
======
text data bss dec hex filename
100263 37808 2752 140823 22617 drivers/block/null_blk/main.o
After:
=====
text data bss dec hex filename
100423 37648 2752 140823 22617 drivers/block/null_blk/main.o
Thorsten Blum [Sat, 20 Dec 2025 23:59:23 +0000 (00:59 +0100)]
brd: replace simple_strtol with kstrtoul in ramdisk_size
Replace simple_strtol() with the recommended kstrtoul() for parsing the
'ramdisk_size=' boot parameter. Unlike simple_strtol(), which returns a
long, kstrtoul() converts the string directly to an unsigned long and
avoids implicit casting.
Check the return value of kstrtoul() and reject invalid values. This
adds error handling while preserving behavior for existing values, and
removes use of the deprecated simple_strtol() helper. The current code
silently sets 'rd_size = 0' if parsing fails, instead of leaving the
default value (CONFIG_BLK_DEV_RAM_SIZE) unchanged.
Linus Torvalds [Sun, 28 Dec 2025 18:21:47 +0000 (10:21 -0800)]
Merge tag 'usb-6.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB fixes from Greg KH:
"Here are some small USB fixes, and bunch of reverts for 6.19-rc3.
Included in here are:
- reverts of some typec ucsi driver changes that had a lot of
regression reports after -rc1. Let's just revert it all for now and
it will come back in a way that is better tested.
- other typec bugfixes
- usb-storage quirk fixups
- dwc3 driver fix
- other minor USB fixes for reported problems.
All of these have passed 0-day testing and individual testing"
* tag 'usb-6.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (22 commits)
Revert "usb: typec: ucsi: Update UCSI structure to have message in and message out fields"
Revert "usb: typec: ucsi: Add support for message out data structure"
Revert "usb: typec: ucsi: Enable debugfs for message_out data structure"
Revert "usb: typec: ucsi: Add support for SET_PDOS command"
Revert "usb: typec: ucsi: Fix null pointer dereference in ucsi_sync_control_common"
Revert "usb: typec: ucsi: Get connector status after enable notifications"
usb: ohci-nxp: clean up probe error labels
usb: gadget: lpc32xx_udc: clean up probe error labels
usb: ohci-nxp: fix device leak on probe failure
usb: phy: isp1301: fix non-OF device reference imbalance
usb: gadget: lpc32xx_udc: fix clock imbalance in error path
usb: typec: ucsi: Get connector status after enable notifications
usb: usb-storage: Maintain minimal modifications to the bcdDevice range.
usb: dwc3: of-simple: fix clock resource leak in dwc3_of_simple_probe
usb: typec: ucsi: Fix null pointer dereference in ucsi_sync_control_common
USB: lpc32xx_udc: Fix error handling in probe
usb: typec: altmodes/displayport: Drop the device reference in dp_altmode_probe()
usb: phy: fsl-usb: Fix use-after-free in delayed work during device removal
usb: renesas_usbhs: Fix a resource leak in usbhs_pipe_malloc()
usb: typec: ucsi: huawei-gaokin: add DRM dependency
...
Linus Torvalds [Sun, 28 Dec 2025 18:14:49 +0000 (10:14 -0800)]
Merge tag 'tty-6.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull serial driver fixes from Greg KH:
"Here are some small serial driver fixes for some reported issues.
Included in here are:
- serial sysfs fwnode fix that was much reported
- sh-sci driver fix
- serial device init bugfix
- 8250 bugfix
- xilinx_uartps bugfix
All of these have passed 0-day testing and individual testing"
* tag 'tty-6.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
serial: xilinx_uartps: fix rs485 delay_rts_after_send
serial: sh-sci: Check that the DMA cookie is valid
serial: core: Fix serial device initialization
serial: 8250: longson: Fix NULL vs IS_ERR() bug in probe
serial: core: Restore sysfs fwnode information
projects / thirdparty / kernel / linux.git / log
