In preparation to completely remove zone write locking, replace the
"zone_wlock" mq-debugfs entry that was listing zones that are
write-locked with the zone_wplugs entry which lists the zones that
currently have a write plug allocated.
The write plug information provided is: the zone number, the zone write
plug flags, the zone write plug write pointer offset and the number of
BIOs currently waiting for execution in the zone write plug BIO list.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-26-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Zone append operations are only allowed to target sequential write
required zones. blk_check_zone_append() uses bio_zone_is_seq() to check
this. However, this check is not necessary because:
1) For NVMe ZNS namespace devices, only sequential write required zones
exist, making the zone type check useless.
2) For null_blk, the driver will fail the request anyway, thus notifying
the user that a conventional zone was targeted.
3) For all other zoned devices, zone append is now emulated using zone
write plugging, which checks that a zone append operation does not
target a conventional zone.
In preparation for the removal of zone write locking and its
conventional zone bitmap (used by bio_zone_is_seq()), remove the
bio_zone_is_seq() call from blk_check_zone_append().
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-24-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The only elevator feature ever implemented is ELEVATOR_F_ZBD_SEQ_WRITE
for signaling that a scheduler implements zone write locking to tightly
control the dispatching order of write operations to zoned block
devices. With the removal of zone write locking support in mq-deadline
and the reliance of all block device drivers on the block layer zone
write plugging to control ordering of write operations to zones, the
elevator feature ELEVATOR_F_ZBD_SEQ_WRITE is completely unused.
Remove it, and also remove the now unused code for filtering the
possible schedulers for a block device based on required features.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-23-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
With the block layer generic plugging of write operations for zoned
block devices, mq-deadline, or any other scheduler, can only ever
see at most one write operation per zone at any time. There is thus no
sequentiality requirements for these writes and thus no need to tightly
control the dispatching of write requests using zone write locking.
Remove all the code that implement this control in the mq-deadline
scheduler and remove advertizing support for the
ELEVATOR_F_ZBD_SEQ_WRITE elevator feature.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-22-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for allowing BIO based device drivers to use zone write
plugging and its zone append emulation, allow these drivers to call
blk_revalidate_disk_zones() so that all zone resources necessary to zone
write plugging can be initialized.
To do so, remove the check in blk_revalidate_disk_zones() restricting
the use of this function to mq request-based drivers to allow also
BIO-based drivers to use it. This is safe to do as long as the
BIO-based block device queue is already setup and usable, as it should,
and can be safely frozen.
The helper function disk_need_zone_resources() is added to control the
allocation and initialization of the zone write plug hash table and
of the conventional zone bitmap only for mq devices and for BIO-based
devices that require zone append emulation.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-12-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Given that zone write plugging manages all writes to zones of a zoned
block device and tracks the write pointer position of all zones that are
not full nor empty, emulating zone append operations using regular
writes can be implemented generically, without relying on the underlying
device driver to implement such emulation. This is needed for devices
that do not natively support the zone append command (e.g. SMR
hard-disks).
A device may request zone append emulation by setting its
max_zone_append_sectors queue limit to 0. For such device, the function
blk_zone_wplug_prepare_bio() changes zone append BIOs into
non-mergeable regular write BIOs. Modified zone append BIOs are flagged
with the new BIO flag BIO_EMULATES_ZONE_APPEND. This flag is checked
on completion of the BIO in blk_zone_write_plug_bio_endio() to restore
the original REQ_OP_ZONE_APPEND operation code of the BIO.
The block layer internal inline helper function bio_is_zone_append() is
added to test if a BIO is either a native zone append operation
(REQ_OP_ZONE_APPEND operation code) or if it is flagged with
BIO_EMULATES_ZONE_APPEND. Given that both native and emulated zone
append BIO completion handling should be similar, The functions
blk_update_request() and blk_zone_complete_request_bio() are modified to
use bio_is_zone_append() to execute blk_zone_update_request_bio() for
both native and emulated zone append operations.
This commit contains contributions from Christoph Hellwig <hch@lst.de>.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-11-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for adding a generic zone append emulation using zone
write plugging, allow device drivers supporting zoned block device to
set a the max_zone_append_sectors queue limit of a device to 0 to
indicate the lack of native support for zone append operations and that
the block layer should emulate these operations using regular write
operations.
blk_queue_max_zone_append_sectors() is modified to allow passing 0 as
the max_zone_append_sectors argument. The function
queue_max_zone_append_sectors() is also modified to ensure that the
minimum of the max_hw_sectors and chunk_sectors limit is used whenever
the max_zone_append_sectors limit is 0. This minimum is consistent with
the value set for the max_zone_append_sectors limit by the function
blk_validate_zoned_limits() when limits for a queue are validated.
The helper functions queue_emulates_zone_append() and
bdev_emulates_zone_append() are added to test if a queue (or block
device) emulates zone append operations.
In order for blk_revalidate_disk_zones() to accept zoned block devices
relying on zone append emulation, the direct check to the
max_zone_append_sectors queue limit of the disk is replaced by a check
using the value returned by queue_max_zone_append_sectors(). Similarly,
queue_zone_append_max_show() is modified to use the same accessor so
that the sysfs attribute advertizes the non-zero limit that will be
used, regardless if it is for native or emulated commands.
For stacking drivers, a top device should not need to care if the
underlying devices have native or emulated zone append operations.
blk_stack_limits() is thus modified to set the top device
max_zone_append_sectors limit using the new accessor
queue_limits_max_zone_append_sectors(). queue_max_zone_append_sectors()
is modified to use this function as well. Stacking drivers that require
zone append emulation, e.g. dm-crypt, can still request this feature by
calling blk_queue_max_zone_append_sectors() with a 0 limit.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-10-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For a zoned block device that has no limit on the number of open zones
and no limit on the number of active zones, the zone write plug mempool
is created with a size of 128 zone write plugs. For such case, set the
device max_open_zones queue limit to this value to indicate to the user
the potential performance penalty that may happen when writing
simultaneously to more zones than the mempool size.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20240408014128.205141-9-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Zone write plugging implements a per-zone "plug" for write operations
to control the submission and execution order of write operations to
sequential write required zones of a zoned block device. Per-zone
plugging guarantees that at any time there is at most only one write
request per zone being executed. This mechanism is intended to replace
zone write locking which implements a similar per-zone write throttling
at the scheduler level, but is implemented only by mq-deadline.
Unlike zone write locking which operates on requests, zone write
plugging operates on BIOs. A zone write plug is simply a BIO list that
is atomically manipulated using a spinlock and a kblockd submission
work. A write BIO to a zone is "plugged" to delay its execution if a
write BIO for the same zone was already issued, that is, if a write
request for the same zone is being executed. The next plugged BIO is
unplugged and issued once the write request completes.
This mechanism allows to:
- Untangle zone write ordering from block IO schedulers. This allows
removing the restriction on using mq-deadline for writing to zoned
block devices. Any block IO scheduler, including "none" can be used.
- Zone write plugging operates on BIOs instead of requests. Plugged
BIOs waiting for execution thus do not hold scheduling tags and thus
are not preventing other BIOs from executing (reads or writes to
other zones). Depending on the workload, this can significantly
improve the device use (higher queue depth operation) and
performance.
- Both blk-mq (request based) zoned devices and BIO-based zoned devices
(e.g. device mapper) can use zone write plugging. It is mandatory
for the former but optional for the latter. BIO-based drivers can
use zone write plugging to implement write ordering guarantees, or
the drivers can implement their own if needed.
- The code is less invasive in the block layer and is mostly limited to
blk-zoned.c with some small changes in blk-mq.c, blk-merge.c and
bio.c.
Zone write plugging is implemented using struct blk_zone_wplug. This
structure includes a spinlock, a BIO list and a work structure to
handle the submission of plugged BIOs. Zone write plugs structures are
managed using a per-disk hash table.
Plugging of zone write BIOs is done using the function
blk_zone_write_plug_bio() which returns false if a BIO execution does
not need to be delayed and true otherwise. This function is called
from blk_mq_submit_bio() after a BIO is split to avoid large BIOs
spanning multiple zones which would cause mishandling of zone write
plugs. This ichange enables by default zone write plugging for any mq
request-based block device. BIO-based device drivers can also use zone
write plugging by expliclty calling blk_zone_write_plug_bio() in their
->submit_bio method. For such devices, the driver must ensure that a
BIO passed to blk_zone_write_plug_bio() is already split and not
straddling zone boundaries.
Only write and write zeroes BIOs are plugged. Zone write plugging does
not introduce any significant overhead for other operations. A BIO that
is being handled through zone write plugging is flagged using the new
BIO flag BIO_ZONE_WRITE_PLUGGING. A request handling a BIO flagged with
this new flag is flagged with the new RQF_ZONE_WRITE_PLUGGING flag.
The completion of BIOs and requests flagged trigger respectively calls
to the functions blk_zone_write_bio_endio() and
blk_zone_write_complete_request(). The latter function is used to
trigger submission of the next plugged BIO using the zone plug work.
blk_zone_write_bio_endio() does the same for BIO-based devices.
This ensures that at any time, at most one request (blk-mq devices) or
one BIO (BIO-based devices) is being executed for any zone. The
handling of zone write plugs using a per-zone plug spinlock maximizes
parallelism and device usage by allowing multiple zones to be writen
simultaneously without lock contention.
Zone write plugging ignores flush BIOs without data. Hovever, any flush
BIO that has data is always plugged so that the write part of the flush
sequence is serialized with other regular writes.
Given that any BIO handled through zone write plugging will be the only
BIO in flight for the target zone when it is executed, the unplugging
and submission of a BIO will have no chance of successfully merging with
plugged requests or requests in the scheduler. To overcome this
potential performance degradation, blk_mq_submit_bio() calls the
function blk_zone_write_plug_attempt_merge() to try to merge other
plugged BIOs with the one just unplugged and submitted. Successful
merging is signaled using blk_zone_write_plug_bio_merged(), called from
bio_attempt_back_merge(). Furthermore, to avoid recalculating the number
of segments of plugged BIOs to attempt merging, the number of segments
of a plugged BIO is saved using the new struct bio field
__bi_nr_segments. To avoid growing the size of struct bio, this field is
added as a union with the bio_cookie field. This is safe to do as
polling is always disabled for plugged BIOs.
When BIOs are plugged in a zone write plug, the device request queue
usage counter is always incremented. This reference is kept and reused
for blk-mq devices when the plugged BIO is unplugged and submitted
again using submit_bio_noacct_nocheck(). For this case, the unplugged
BIO is already flagged with BIO_ZONE_WRITE_PLUGGING and
blk_mq_submit_bio() proceeds directly to allocating a new request for
the BIO, re-using the usage reference count taken when the BIO was
plugged. This extra reference count is dropped in
blk_zone_write_plug_attempt_merge() for any plugged BIO that is
successfully merged. Given that BIO-based devices will not take this
path, the extra reference is dropped after a plugged BIO is unplugged
and submitted.
Zone write plugs are dynamically allocated and managed using a hash
table (an array of struct hlist_head) with RCU protection.
A zone write plug is allocated when a write BIO is received for the
zone and not freed until the zone is fully written, reset or finished.
To detect when a zone write plug can be freed, the write state of each
zone is tracked using a write pointer offset which corresponds to the
offset of a zone write pointer relative to the zone start. Write
operations always increment this write pointer offset. Zone reset
operations set it to 0 and zone finish operations set it to the zone
size.
If a write error happens, the wp_offset value of a zone write plug may
become incorrect and out of sync with the device managed write pointer.
This is handled using the zone write plug flag BLK_ZONE_WPLUG_ERROR.
The function blk_zone_wplug_handle_error() is called from the new disk
zone write plug work when this flag is set. This function executes a
report zone to update the zone write pointer offset to the current
value as indicated by the device. The disk zone write plug work is
scheduled whenever a BIO flagged with BIO_ZONE_WRITE_PLUGGING completes
with an error or when bio_zone_wplug_prepare_bio() detects an unaligned
write. Once scheduled, the disk zone write plugs work keeps running
until all zone errors are handled.
To match the new data structures used for zoned disks, the function
disk_free_zone_bitmaps() is renamed to the more generic
disk_free_zone_resources(). The function disk_init_zone_resources() is
also introduced to initialize zone write plugs resources when a gendisk
is allocated.
In order to guarantee that the user can simultaneously write up to a
number of zones equal to a device max active zone limit or max open zone
limit, zone write plugs are allocated using a mempool sized to the
maximum of these 2 device limits. For a device that does not have
active and open zone limits, 128 is used as the default mempool size.
If a change to the device active and open zone limits is detected, the
disk mempool is resized when blk_revalidate_disk_zones() is executed.
This commit contains contributions from Christoph Hellwig <hch@lst.de>.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20240408014128.205141-8-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for adding zone write plugging, modify
blk_revalidate_disk_zones() to get the capacity of zones of a zoned
block device. This capacity value as a number of 512B sectors is stored
in the gendisk zone_capacity field.
Given that host-managed SMR disks (including zoned UFS drives) and all
known NVMe ZNS devices have the same zone capacity for all zones
blk_revalidate_disk_zones() returns an error if different capacities are
detected for different zones.
This also adds check to verify that the values reported by the device
for zone capacities are correct, that is, that the zone capacity is
never 0, does not exceed the zone size and is equal to the zone size for
conventional zones.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20240408014128.205141-7-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
On completion of a zone append request, the request sector indicates the
location of the written data. This value must be returned to the user
through the BIO iter sector. This is done in 2 places: in
blk_complete_request() and in blk_update_request(). Introduce the inline
helper function blk_zone_update_request_bio() to avoid duplicating
this BIO update for zone append requests, and to compile out this
helper call when CONFIG_BLK_DEV_ZONED is not enabled.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20240408014128.205141-4-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
On completion of a flush sequence, blk_flush_restore_request() restores
the bio of a request to the original submitted BIO. However, the last
use of the request in the flush sequence may have been for a POSTFLUSH
which does not have a sector. So make sure to restore the request sector
using the iter sector of the original BIO. This BIO has not changed yet
since the completions of the flush sequence intermediate steps use
requeueing of the request until all steps are completed.
Restoring the request sector ensures that blk_mq_end_request() will see
a valid sector as originally set when the flush BIO was submitted.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20240408014128.205141-2-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blkdev_dio_unaligned() is called from __blkdev_direct_IO(),
__blkdev_direct_IO_simple(), and __blkdev_direct_IO_async(), and all these
are only called from blkdev_direct_IO().
Move the blkdev_dio_unaligned() call to the common callsite,
blkdev_direct_IO().
Pass those functions the bdev pointer from blkdev_direct_IO(), as it is
non-trivial to look up.
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: John Garry <john.g.garry@oracle.com>
Link: https://lore.kernel.org/r/20240415122020.1541594-1-john.g.garry@oracle.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull block fixes from Jens Axboe:
- MD pull request via Song:
- UAF fix (Yu)
- Avoid out-of-bounds shift in blk-iocost (Rik)
- Fix for q->blkg_list corruption (Ming)
- Relax virt boundary mask/size segment checking (Ming)
* tag 'block-6.9-20240412' of git://git.kernel.dk/linux:
block: fix that blk_time_get_ns() doesn't update time after schedule
block: allow device to have both virt_boundary_mask and max segment size
block: fix q->blkg_list corruption during disk rebind
blk-iocost: avoid out of bounds shift
raid1: fix use-after-free for original bio in raid1_write_request()
While monitoring the throttle time of IO from iocost, it's found that
such time is always zero after the io_schedule() from ioc_rqos_throttle,
for example, with the following debug patch:
+ printk("%s-%d: %s enter %llu\n", current->comm, current->pid, __func__, blk_time_get_ns());
while (true) {
set_current_state(TASK_UNINTERRUPTIBLE);
if (wait.committed)
break;
io_schedule();
}
+ printk("%s-%d: %s exit %llu\n", current->comm, current->pid, __func__, blk_time_get_ns());
It can be observerd that blk_time_get_ns() always return the same time:
[ 1068.096579] fio-1268: ioc_rqos_throttle enter 1067901962288
[ 1068.272587] fio-1268: ioc_rqos_throttle exit 1067901962288
[ 1068.274389] fio-1268: ioc_rqos_throttle enter 1067901962288
[ 1068.472690] fio-1268: ioc_rqos_throttle exit 1067901962288
[ 1068.474485] fio-1268: ioc_rqos_throttle enter 1067901962288
[ 1068.672656] fio-1268: ioc_rqos_throttle exit 1067901962288
[ 1068.674451] fio-1268: ioc_rqos_throttle enter 1067901962288
[ 1068.872655] fio-1268: ioc_rqos_throttle exit 1067901962288
And I think the root cause is that 'PF_BLOCK_TS' is always cleared
by blk_flush_plug() before scheduel(), hence blk_plug_invalidate_ts()
will never be called:
blk_time_get_ns
plug->cur_ktime = ktime_get_ns();
current->flags |= PF_BLOCK_TS;
io_schedule:
io_schedule_prepare
blk_flush_plug
__blk_flush_plug
/* the flag is cleared, while time is not */
current->flags &= ~PF_BLOCK_TS;
schedule
sched_update_worker
/* the flag is not set, hence plug->cur_ktime is not cleared */
if (tsk->flags & PF_BLOCK_TS)
blk_plug_invalidate_ts()
blk_time_get_ns
/* got the time stashed before schedule */
return plug->cur_ktime;
Fix the problem by clearing cached time in __blk_flush_plug().
Fixes: 06b23f92af ("block: update cached timestamp post schedule/preemption")
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20240411032349.3051233-2-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Multiple gendisk instances can allocated/added for single request queue
in case of disk rebind. blkg may still stay in q->blkg_list when calling
blkcg_init_disk() for rebind, then q->blkg_list becomes corrupted.
Fix the list corruption issue by:
- add blkg_init_queue() to initialize q->blkg_list & q->blkcg_mutex only
- move calling blkg_init_queue() into blk_alloc_queue()
The list corruption should be started since commit f1c006f1c6 ("blk-cgroup:
synchronize pd_free_fn() from blkg_free_workfn() and blkcg_deactivate_policy()")
which delays removing blkg from q->blkg_list into blkg_free_workfn().
Fixes: f1c006f1c6 ("blk-cgroup: synchronize pd_free_fn() from blkg_free_workfn() and blkcg_deactivate_policy()")
Fixes: 1059699f87 ("block: move blkcg initialization/destroy into disk allocation/release handler")
Cc: Yu Kuai <yukuai3@huawei.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20240407125910.4053377-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There's a bunch of flags that are purely based on what the file
operations support while also never being conditionally set or unset.
IOW, they're not subject to change for individual files. Imho, such
flags don't need to live in f_mode they might as well live in the fops
structs itself. And the fops struct already has that lonely
mmap_supported_flags member. We might as well turn that into a generic
fop_flags member and move a few flags from FMODE_* space into FOP_*
space. That gets us four FMODE_* bits back and the ability for new
static flags that are about file ops to not have to live in FMODE_*
space but in their own FOP_* space. It's not the most beautiful thing
ever but it gets the job done. Yes, there'll be an additional pointer
chase but hopefully that won't matter for these flags.
I suspect there's a few more we can move into there and that we can also
redirect a bunch of new flag suggestions that follow this pattern into
the fop_flags field instead of f_mode.
Link: https://lore.kernel.org/r/20240328-gewendet-spargel-aa60a030ef74@brauner
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Christian Brauner <brauner@kernel.org>
UBSAN catches undefined behavior in blk-iocost, where sometimes
iocg->delay is shifted right by a number that is too large,
resulting in undefined behavior on some architectures.
[ 186.556576] ------------[ cut here ]------------
UBSAN: shift-out-of-bounds in block/blk-iocost.c:1366:23
shift exponent 64 is too large for 64-bit type 'u64' (aka 'unsigned long long')
CPU: 16 PID: 0 Comm: swapper/16 Tainted: G S E N 6.9.0-0_fbk700_debug_rc2_kbuilder_0_gc85af715cac0 #1
Hardware name: Quanta Twin Lakes MP/Twin Lakes Passive MP, BIOS F09_3A23 12/08/2020
Call Trace:
<IRQ>
dump_stack_lvl+0x8f/0xe0
__ubsan_handle_shift_out_of_bounds+0x22c/0x280
iocg_kick_delay+0x30b/0x310
ioc_timer_fn+0x2fb/0x1f80
__run_timer_base+0x1b6/0x250
...
Avoid that undefined behavior by simply taking the
"delay = 0" branch if the shift is too large.
I am not sure what the symptoms of an undefined value
delay will be, but I suspect it could be more than a
little annoying to debug.
Signed-off-by: Rik van Riel <riel@surriel.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Jens Axboe <axboe@kernel.dk>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20240404123253.0f58010f@imladris.surriel.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull vfs fixes from Christian Brauner:
"This contains a few small fixes. This comes with some delay because I
wanted to wait on people running their reproducers and the Easter
Holidays meant that those replies came in a little later than usual:
- Fix handling of preventing writes to mounted block devices.
Since last kernel we allow to prevent writing to mounted block
devices provided CONFIG_BLK_DEV_WRITE_MOUNTED isn't set and the
block device is opened with restricted writes. When we switched to
opening block devices as files we altered the mechanism by which we
recognize when a block device has been opened with write
restrictions.
The detection logic assumed that only read-write mounted
filesystems would apply write restrictions to their block devices
from other openers. That of course is not true since it also makes
sense to apply write restrictions for filesystems that are
read-only.
Fix the detection logic using an FMODE_* bit. We still have a few
left since we freed up a couple a while ago. I also picked up a
patch to free up four additional FMODE_* bits scheduled for the
next merge window.
- Fix counting the number of writers to a block device. This just
changes the logic to be consistent.
- Fix a bug in aio causing a NULL pointer derefernce after we
implemented batched processing in aio.
- Finally, add the changes we discussed that allows to yield block
devices early even though file closing itself is deferred.
This also allows us to remove two holder operations to get and
release the holder to align lifetime of file and holder of the
block device"
* tag 'vfs-6.9-rc3.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
aio: Fix null ptr deref in aio_complete() wakeup
fs,block: yield devices early
block: count BLK_OPEN_RESTRICT_WRITES openers
block: handle BLK_OPEN_RESTRICT_WRITES correctly
Kernel parameter of `isolcpus=` or 'nohz_full=' are used to isolate CPUs
for specific task, and it isn't expected to let block IO disturb these CPUs.
blk-mq kworker shouldn't be scheduled on isolated CPUs. Also if isolated
CPUs is run for blk-mq kworker, long block IO latency can be caused.
Kernel workqueue only respects CPU isolation for WQ_UNBOUND, for bound
WQ, the responsibility is on user because CPU is specified as WQ API
parameter, such as mod_delayed_work_on(cpu), queue_delayed_work_on(cpu)
and queue_work_on(cpu).
So not run blk-mq kworker on isolated CPUs by removing isolated CPUs
from hctx->cpumask. Meantime use queue map to check if all CPUs in this
hw queue are offline instead of hctx->cpumask, this way can avoid any
cost in fast IO code path, and is safe since hctx->cpumask are only
used in the two cases.
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Andrew Theurer <atheurer@redhat.com>
Cc: Joe Mario <jmario@redhat.com>
Cc: Sebastian Jug <sejug@redhat.com>
Cc: Frederic Weisbecker <frederic@kernel.org>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Tejun Heo <tj@kernel.org>
Tesed-by: Joe Mario <jmario@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Ewan D. Milne <emilne@redhat.com>
Link: https://lore.kernel.org/r/20240322021244.1056223-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull block fixes from Jens Axboe:
"Small round of minor fixes or cleanups for the 6.9-rc2 kernel, one
fixing an issue introduced in 6.8"
* tag 'block-6.9-20240329' of git://git.kernel.dk/linux:
block: Do not force full zone append completion in req_bio_endio()
block: don't reject too large max_user_sectors in blk_validate_limits
block: Make blk_rq_set_mixed_merge() static
This reverts commit 748dc0b65e.
Partial zone append completions cannot be supported as there is no
guarantees that the fragmented data will be written sequentially in the
same manner as with a full command. Commit 748dc0b65e ("block: fix
partial zone append completion handling in req_bio_endio()") changed
req_bio_endio() to always advance a partially failed BIO by its full
length, but this can lead to incorrect accounting. So revert this
change and let low level device drivers handle this case by always
failing completely zone append operations. With this revert, users will
still see an IO error for a partially completed zone append BIO.
Fixes: 748dc0b65e ("block: fix partial zone append completion handling in req_bio_endio()")
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/20240328004409.594888-2-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently a device is only really released once the umount returns to
userspace due to how file closing works. That ultimately could cause
an old umount assumption to be violated that concurrent umount and mount
don't fail. So an exclusively held device with a temporary holder should
be yielded before the filesystem is gone. Add a helper that allows
callers to do that. This also allows us to remove the two holder ops
that Linus wasn't excited about.
Link: https://lore.kernel.org/r/20240326-vfs-bdev-end_holder-v1-1-20af85202918@kernel.org
Fixes: f3a608827d ("bdev: open block device as files") # mainline only
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
The original changes in v6.8 do allow for a block device to be reopened
with BLK_OPEN_RESTRICT_WRITES provided the same holder is used as per
bdev_may_open(). I think this has a bug.
The first opener @f1 of that block device will set bdev->bd_writers to
-1. The second opener @f2 using the same holder will pass the check in
bdev_may_open() that bdev->bd_writers must not be greater than zero.
The first opener @f1 now closes the block device and in bdev_release()
will end up calling bdev_yield_write_access() which calls
bdev_writes_blocked() and sets bdev->bd_writers to 0 again.
Now @f2 holds a file to that block device which was opened with
exclusive write access but bdev->bd_writers has been reset to 0.
So now @f3 comes along and succeeds in opening the block device with
BLK_OPEN_WRITE betraying @f2's request to have exclusive write access.
This isn't a practical issue yet because afaict there's no codepath
inside the kernel that reopenes the same block device with
BLK_OPEN_RESTRICT_WRITES but it will be if there is.
Fix this by counting the number of BLK_OPEN_RESTRICT_WRITES openers. So
we only allow writes again once all BLK_OPEN_RESTRICT_WRITES openers are
done.
Link: https://lore.kernel.org/r/20240323-abtauchen-klauen-c2953810082d@brauner
Fixes: ed5cc702d3 ("block: Add config option to not allow writing to mounted devices")
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Last kernel release we introduce CONFIG_BLK_DEV_WRITE_MOUNTED. By
default this option is set. When it is set the long-standing behavior
of being able to write to mounted block devices is enabled.
But in order to guard against unintended corruption by writing to the
block device buffer cache CONFIG_BLK_DEV_WRITE_MOUNTED can be turned
off. In that case it isn't possible to write to mounted block devices
anymore.
A filesystem may open its block devices with BLK_OPEN_RESTRICT_WRITES
which disallows concurrent BLK_OPEN_WRITE access. When we still had the
bdev handle around we could recognize BLK_OPEN_RESTRICT_WRITES because
the mode was passed around. Since we managed to get rid of the bdev
handle we changed that logic to recognize BLK_OPEN_RESTRICT_WRITES based
on whether the file was opened writable and writes to that block device
are blocked. That logic doesn't work because we do allow
BLK_OPEN_RESTRICT_WRITES to be specified without BLK_OPEN_WRITE.
Fix the detection logic and use an FMODE_* bit. We could've also abused
O_EXCL as an indicator that BLK_OPEN_RESTRICT_WRITES has been requested.
For userspace open paths O_EXCL will never be retained but for internal
opens where we open files that are never installed into a file
descriptor table this is fine. But it would be a gamble that this
doesn't cause bugs. Note that BLK_OPEN_RESTRICT_WRITES is an internal
only flag that cannot directly be raised by userspace. It is implicitly
raised during mounting.
Passes xftests and blktests with CONFIG_BLK_DEV_WRITE_MOUNTED set and
unset.
Link: https://lore.kernel.org/r/ZfyyEwu9Uq5Pgb94@casper.infradead.org
Link: https://lore.kernel.org/r/20240323-zielbereich-mittragen-6fdf14876c3e@brauner
Fixes: 321de651fa ("block: don't rely on BLK_OPEN_RESTRICT_WRITES when yielding write access")
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reported-by: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull vfs fixes from Christian Brauner:
"This contains a few small fixes for this merge window:
- Undo the hiding of silly-rename files in afs. If they're hidden
they can't be deleted by rm manually anymore causing regressions
- Avoid caching the preferred address for an afs server to avoid
accidently overriding an explicitly specified preferred server
address
- Fix bad stat() and rmdir() interaction in afs
- Take a passive reference on the superblock when opening a block
device so the holder is available to concurrent callers from the
block layer
- Clear private data pointer in fscache_begin_operation() to avoid it
being falsely treated as valid"
* tag 'vfs-6.9-rc1.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
fscache: Fix error handling in fscache_begin_operation()
fs,block: get holder during claim
afs: Fix occasional rmdir-then-VNOVNODE with generic/011
afs: Don't cache preferred address
afs: Revert "afs: Hide silly-rename files from userspace"
This reverts commit 8a08c5fd89.
It turns out while this is a perfectly valid and long overdue thing to do
for user initiated discards / zeroing from the ioctl handler, it actually
breaks file system use of the discard helper by interrupting in places
the file system doesn't expect, and by leaving the bio chain in a state
that the file system callers of (at least) __blkdev_issue_discard do
not expect.
Revert the change for now, we'll redo it for the next merge window
after refactoring the code to better split the file system vs ioctl
callers and cleaning up a few other loose ends.
Reported-by: Chandan Babu R <chandanbabu@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Keith Busch <kbusch@kernel.org>
Link: https://lore.kernel.org/r/20240314021623.1908895-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull block updates from Jens Axboe:
- MD pull requests via Song:
- Cleanup redundant checks (Yu Kuai)
- Remove deprecated headers (Marc Zyngier, Song Liu)
- Concurrency fixes (Li Lingfeng)
- Memory leak fix (Li Nan)
- Refactor raid1 read_balance (Yu Kuai, Paul Luse)
- Clean up and fix for md_ioctl (Li Nan)
- Other small fixes (Gui-Dong Han, Heming Zhao)
- MD atomic limits (Christoph)
- NVMe pull request via Keith:
- RDMA target enhancements (Max)
- Fabrics fixes (Max, Guixin, Hannes)
- Atomic queue_limits usage (Christoph)
- Const use for class_register (Ricardo)
- Identification error handling fixes (Shin'ichiro, Keith)
- Improvement and cleanup for cached request handling (Christoph)
- Moving towards atomic queue limits. Core changes and driver bits so
far (Christoph)
- Fix UAF issues in aoeblk (Chun-Yi)
- Zoned fix and cleanups (Damien)
- s390 dasd cleanups and fixes (Jan, Miroslav)
- Block issue timestamp caching (me)
- noio scope guarding for zoned IO (Johannes)
- block/nvme PI improvements (Kanchan)
- Ability to terminate long running discard loop (Keith)
- bdev revalidation fix (Li)
- Get rid of old nr_queues hack for kdump kernels (Ming)
- Support for async deletion of ublk (Ming)
- Improve IRQ bio recycling (Pavel)
- Factor in CPU capacity for remote vs local completion (Qais)
- Add shared_tags configfs entry for null_blk (Shin'ichiro
- Fix for a regression in page refcounts introduced by the folio
unification (Tony)
- Misc fixes and cleanups (Arnd, Colin, John, Kunwu, Li, Navid,
Ricardo, Roman, Tang, Uwe)
* tag 'for-6.9/block-20240310' of git://git.kernel.dk/linux: (221 commits)
block: partitions: only define function mac_fix_string for CONFIG_PPC_PMAC
block/swim: Convert to platform remove callback returning void
cdrom: gdrom: Convert to platform remove callback returning void
block: remove disk_stack_limits
md: remove mddev->queue
md: don't initialize queue limits
md/raid10: use the atomic queue limit update APIs
md/raid5: use the atomic queue limit update APIs
md/raid1: use the atomic queue limit update APIs
md/raid0: use the atomic queue limit update APIs
md: add queue limit helpers
md: add a mddev_is_dm helper
md: add a mddev_add_trace_msg helper
md: add a mddev_trace_remap helper
bcache: move calculation of stripe_size and io_opt into bcache_device_init
virtio_blk: Do not use disk_set_max_open/active_zones()
aoe: fix the potential use-after-free problem in aoecmd_cfg_pkts
block: move capacity validation to blkpg_do_ioctl()
block: prevent division by zero in blk_rq_stat_sum()
drbd: atomically update queue limits in drbd_reconsider_queue_parameters
...
Damien Le Moal