| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Currently, io_ticks is accounted based on sampling, specifically
update_io_ticks() will always account io_ticks by 1 jiffies from
bdev_start_io_acct()/blk_account_io_start(), and the result can be
inaccurate, for example(HZ is 250):
Test script:
fio -filename=/dev/sda -bs=4k -rw=write -direct=1 -name=test -thinktime=4ms
Test result: util is about 90%, while the disk is really idle.
This behaviour is introduced by commit 5b18b5a73760 ("block: delete
part_round_stats and switch to less precise counting"), however, there
was a key point that is missed that this patch also improve performance
a lot:
Before the commit:
part_round_stats:
if (part->stamp != now)
stats |= 1;
part_in_flight()
-> there can be lots of task here in 1 jiffies.
part_round_stats_single()
__part_stat_add()
part->stamp = now;
After the commit:
update_io_ticks:
stamp = part->bd_stamp;
if (time_after(now, stamp))
if (try_cmpxchg())
__part_stat_add()
-> only one task can reach here in 1 jiffies.
Hence in order to account io_ticks precisely, we only need to know if
there are IO inflight at most once in one jiffies. Noted that for
rq-based device, iterating tags should not be used here because
'tags->lock' is grabbed in blk_mq_find_and_get_req(), hence
part_stat_lock_inc/dec() and part_in_flight() is used to trace inflight.
The additional overhead is quite little:
- per cpu add/dec for each IO for rq-based device;
- per cpu sum for each jiffies;
And it's verified by null-blk that there are no performance degration
under heavy IO pressure.
Fixes: 5b18b5a73760 ("block: delete part_round_stats and switch to less precise counting")
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20240509123717.3223892-2-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
With the block layer zone write plugging being automatically done for
any write operation to a zone of a zoned block device, a regular request
plugging handled through current->plug can only ever see at most a
single write request per zone. In such case, any potential reordering
of the plugged requests will be harmless. We can thus remove the special
casing for write operations to zones and have these requests plugged as
well. This allows removing the function blk_mq_plug and instead directly
using current->plug where needed.
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/20240408014128.205141-29-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>
|
|
Remove "static" from the definition of bio_attempt_back_merge() and
declare this function in block/blk.h to allow using it internally from
other block layer files. The definition of enum bio_merge_status is
also moved to block/blk.h.
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/20240408014128.205141-6-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Since commit 8e756373d7c8 ("block: Move bio merge related functions into
blk-merge.c"), blk_rq_set_mixed_merge() has only been referenced in
blk-merge.c, so make it static.
Signed-off-by: John Garry <john.g.garry@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20240325083501.2816408-1-john.g.garry@oracle.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Restore support for passing data lifetime information from filesystems to
block drivers. This patch reverts commit b179c98f7697 ("block: Remove
request.write_hint") and commit c75e707fe1aa ("block: remove the
per-bio/request write hint").
This patch does not modify the size of struct bio because the new
bi_write_hint member fills a hole in struct bio. pahole reports the
following for struct bio on an x86_64 system with this patch applied:
/* size: 112, cachelines: 2, members: 20 */
/* sum members: 110, holes: 1, sum holes: 2 */
/* last cacheline: 48 bytes */
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20240202203926.2478590-7-bvanassche@acm.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
A zero discard_granularity is not treated the same as a single-block one,
and not having any segments after taking alignment is perfectly fine
and does not need a warning.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231228075545.362768-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Once all I/O using a blk_crypto_key has completed, filesystems can call
blk_crypto_evict_key(). However, the block layer currently doesn't call
blk_crypto_put_keyslot() until the request is being freed, which happens
after upper layers have been told (via bio_endio()) the I/O has
completed. This causes a race condition where blk_crypto_evict_key()
can see 'slot_refs != 0' without there being an actual bug.
This makes __blk_crypto_evict_key() hit the
'WARN_ON_ONCE(atomic_read(&slot->slot_refs) != 0)' and return without
doing anything, eventually causing a use-after-free in
blk_crypto_reprogram_all_keys(). (This is a very rare bug and has only
been seen when per-file keys are being used with fscrypt.)
There are two options to fix this: either release the keyslot before
bio_endio() is called on the request's last bio, or make
__blk_crypto_evict_key() ignore slot_refs. Let's go with the first
solution, since it preserves the ability to report bugs (via
WARN_ON_ONCE) where a key is evicted while still in-use.
Fixes: a892c8d52c02 ("block: Inline encryption support for blk-mq")
Cc: stable@vger.kernel.org
Reviewed-by: Nathan Huckleberry <nhuck@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20230315183907.53675-2-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Pull block fixes from Jens Axboe:
- NVMe pull request via Christoph:
- Don't access released socket during error recovery (Akinobu
Mita)
- Bring back auto-removal of deleted namespaces during sequential
scan (Christoph Hellwig)
- Fix an error code in nvme_auth_process_dhchap_challenge (Dan
Carpenter)
- Show well known discovery name (Daniel Wagner)
- Add a missing endianess conversion in effects masking (Keith
Busch)
- Fix for a regression introduced in blk-rq-qos during init in this
merge window (Breno)
- Reorder a few fields in struct blk_mq_tag_set, eliminating a few
holes and shrinking it (Christophe)
- Remove redundant bdev_get_queue() NULL checks (Juhyung)
- Add sed-opal single user mode support flag (Luca)
- Remove SQE128 check in ublk as it isn't needed, saving some memory
(Ming)
- Op specific segment checking for cloned requests (Uday)
- Exclusive open partition scan fixes (Yu)
- Loop offset/size checking before assigning them in the device (Zhong)
- Bio polling fixes (me)
* tag 'block-6.3-2023-03-03' of git://git.kernel.dk/linux:
blk-mq: enforce op-specific segment limits in blk_insert_cloned_request
nvme-fabrics: show well known discovery name
nvme-tcp: don't access released socket during error recovery
nvme-auth: fix an error code in nvme_auth_process_dhchap_challenge()
nvme: bring back auto-removal of deleted namespaces during sequential scan
blk-iocost: Pass gendisk to ioc_refresh_params
nvme: fix sparse warning on effects masking
block: be a bit more careful in checking for NULL bdev while polling
block: clear bio->bi_bdev when putting a bio back in the cache
loop: loop_set_status_from_info() check before assignment
ublk: remove check IO_URING_F_SQE128 in ublk_ch_uring_cmd
block: remove more NULL checks after bdev_get_queue()
blk-mq: Reorder fields in 'struct blk_mq_tag_set'
block: fix scan partition for exclusively open device again
block: Revert "block: Do not reread partition table on exclusively open device"
sed-opal: add support flag for SUM in status ioctl
|
|
The block layer might merge together discard requests up until the
max_discard_segments limit is hit, but blk_insert_cloned_request checks
the segment count against max_segments regardless of the req op. This
can result in errors like the following when discards are issued through
a DM device and max_discard_segments exceeds max_segments for the queue
of the chosen underlying device.
blk_insert_cloned_request: over max segments limit. (256 > 129)
Fix this by looking at the req_op and enforcing the appropriate segment
limit - max_discard_segments for REQ_OP_DISCARDs and max_segments for
everything else.
Signed-off-by: Uday Shankar <ushankar@purestorage.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20230301000655.48112-1-ushankar@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Pull block updates from Jens Axboe:
- NVMe updates via Christoph:
- Small improvements to the logging functionality (Amit Engel)
- Authentication cleanups (Hannes Reinecke)
- Cleanup and optimize the DMA mapping cod in the PCIe driver
(Keith Busch)
- Work around the command effects for Format NVM (Keith Busch)
- Misc cleanups (Keith Busch, Christoph Hellwig)
- Fix and cleanup freeing single sgl (Keith Busch)
- MD updates via Song:
- Fix a rare crash during the takeover process
- Don't update recovery_cp when curr_resync is ACTIVE
- Free writes_pending in md_stop
- Change active_io to percpu
- Updates to drbd, inching us closer to unifying the out-of-tree driver
with the in-tree one (Andreas, Christoph, Lars, Robert)
- BFQ update adding support for multi-actuator drives (Paolo, Federico,
Davide)
- Make brd compliant with REQ_NOWAIT (me)
- Fix for IOPOLL and queue entering, fixing stalled IO waiting on
timeouts (me)
- Fix for REQ_NOWAIT with multiple bios (me)
- Fix memory leak in blktrace cleanup (Greg)
- Clean up sbitmap and fix a potential hang (Kemeng)
- Clean up some bits in BFQ, and fix a bug in the request injection
(Kemeng)
- Clean up the request allocation and issue code, and fix some bugs
related to that (Kemeng)
- ublk updates and fixes:
- Add support for unprivileged ublk (Ming)
- Improve device deletion handling (Ming)
- Misc (Liu, Ziyang)
- s390 dasd fixes (Alexander, Qiheng)
- Improve utility of request caching and fixes (Anuj, Xiao)
- zoned cleanups (Pankaj)
- More constification for kobjs (Thomas)
- blk-iocost cleanups (Yu)
- Remove bio splitting from drivers that don't need it (Christoph)
- Switch blk-cgroups to use struct gendisk. Some of this is now
incomplete as select late reverts were done. (Christoph)
- Add bvec initialization helpers, and convert callers to use that
rather than open-coding it (Christoph)
- Misc fixes and cleanups (Jinke, Keith, Arnd, Bart, Li, Martin,
Matthew, Ulf, Zhong)
* tag 'for-6.3/block-2023-02-16' of git://git.kernel.dk/linux: (169 commits)
brd: use radix_tree_maybe_preload instead of radix_tree_preload
block: use proper return value from bio_failfast()
block: bio-integrity: Copy flags when bio_integrity_payload is cloned
block: Fix io statistics for cgroup in throttle path
brd: mark as nowait compatible
brd: check for REQ_NOWAIT and set correct page allocation mask
brd: return 0/-error from brd_insert_page()
block: sync mixed merged request's failfast with 1st bio's
Revert "blk-cgroup: pin the gendisk in struct blkcg_gq"
Revert "blk-cgroup: pass a gendisk to blkg_lookup"
Revert "blk-cgroup: delay blk-cgroup initialization until add_disk"
Revert "blk-cgroup: delay calling blkcg_exit_disk until disk_release"
Revert "blk-cgroup: move the cgroup information to struct gendisk"
nvme-pci: remove iod use_sgls
nvme-pci: fix freeing single sgl
block: ublk: check IO buffer based on flag need_get_data
s390/dasd: Fix potential memleak in dasd_eckd_init()
s390/dasd: sort out physical vs virtual pointers usage
block: Remove the ALLOC_CACHE_SLACK constant
block: make kobj_type structures constant
...
|
|
kernel test robot complains about a type mismatch:
block/blk-merge.c:984:42: sparse: expected restricted blk_opf_t const [usertype] ff
block/blk-merge.c:984:42: sparse: got unsigned int
block/blk-merge.c:1010:42: sparse: sparse: incorrect type in initializer (different base types) @@ expected restricted blk_opf_t const [usertype] ff @@ got unsigned int @@
block/blk-merge.c:1010:42: sparse: expected restricted blk_opf_t const [usertype] ff
block/blk-merge.c:1010:42: sparse: got unsigned int
because bio_failfast() is return an unsigned int rather than the
appropriate blk_opt_f type. Fix it up.
Fixes: 3ce6a115980c ("block: sync mixed merged request's failfast with 1st bio's")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202302170743.GXypM9Rt-lkp@intel.com/
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
We support mixed merge for requests/bios with different fastfail
settings. When request fails, each time we only handle the portion
with same failfast setting, then bios with failfast can be failed
immediately, and bios without failfast can be retried.
The idea is pretty good, but the current implementation has several
defects:
1) initially RA bio doesn't set failfast, however bio merge code
doesn't consider this point, and just check its failfast setting for
deciding if mixed merge is required. Fix this issue by adding helper
of bio_failfast().
2) when merging bio to request front, if this request is mixed
merged, we have to sync request's faifast setting with 1st bio's
failfast. Fix it by calling blk_update_mixed_merge().
3) when merging bio to request back, if this request is mixed
merged, we have to mark the bio as failfast, because blk_update_request
simply updates request failfast with 1st bio's failfast. Fix
it by calling blk_update_mixed_merge().
Fixes one normal EXT4 READ IO failure issue, because it is observed
that the normal READ IO is merged with RA IO, and the mixed merged
request has different failfast setting with 1st bio's, so finally
the normal READ IO doesn't get retried.
Cc: Tejun Heo <tj@kernel.org>
Fixes: 80a761fd33cf ("block: implement mixed merge of different failfast requests")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20230209125527.667004-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
bio_split_rw can be used by file systems to split and incoming write
bio into multiple bios fitting the hardware limit for use as ZONE_APPEND
bios. Export it for initial use in btrfs.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
If we split a bio marked with REQ_NOWAIT, then we can trigger spurious
EAGAIN if constituent parts of that split bio end up failing request
allocations. Parts will complete just fine, but just a single failure
in one of the chained bios will yield an EAGAIN final result for the
parent bio.
Return EAGAIN early if we end up needing to split such a bio, which
allows for saner recovery handling.
Cc: stable@vger.kernel.org # 5.15+
Link: https://github.com/axboe/liburing/issues/766
Reported-by: Michael Kelley <mikelley@microsoft.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
This can't happen right now, but in preparation for allowing
bio_split_to_limits() returning NULL if it ended the bio, check for it
in all the callers.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
This patch removes a conditional jump from get_max_segment_size(). The
x86-64 assembler code for this function without this patch is as follows:
206 return min_not_zero(mask - offset + 1,
0x0000000000000118 <+72>: not %rax
0x000000000000011b <+75>: and 0x8(%r10),%rax
0x000000000000011f <+79>: add $0x1,%rax
0x0000000000000123 <+83>: je 0x138 <bvec_split_segs+104>
0x0000000000000125 <+85>: cmp %rdx,%rax
0x0000000000000128 <+88>: mov %rdx,%r12
0x000000000000012b <+91>: cmovbe %rax,%r12
0x000000000000012f <+95>: test %rdx,%rdx
0x0000000000000132 <+98>: mov %eax,%edx
0x0000000000000134 <+100>: cmovne %r12d,%edx
With this patch applied:
206 return min(mask - offset, (unsigned long)lim->max_segment_size - 1) + 1;
0x000000000000003f <+63>: mov 0x28(%rdi),%ebp
0x0000000000000042 <+66>: not %rax
0x0000000000000045 <+69>: and 0x8(%rdi),%rax
0x0000000000000049 <+73>: sub $0x1,%rbp
0x000000000000004d <+77>: cmp %rbp,%rax
0x0000000000000050 <+80>: cmova %rbp,%rax
0x0000000000000054 <+84>: add $0x1,%eax
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Keith Busch <kbusch@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20221025191755.1711437-4-bvanassche@acm.org
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Document which functions do not modify the queue limits.
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Keith Busch <kbusch@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20221025191755.1711437-3-bvanassche@acm.org
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Allow using the splitting helpers on just a queue_limits instead of
a full request_queue structure. This will eventually allow file systems
or remapping drivers to split REQ_OP_ZONE_APPEND bios based on limits
calculated as the minimum common capabilities over multiple devices.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20220727162300.3089193-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Move this helper into the only file where it is used.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20220727162300.3089193-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Prepare for reusing blk_bio_segment_split for (file system controlled)
splits of REQ_OP_ZONE_APPEND bios by letting the caller control the
maximum size of the bio.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20220727162300.3089193-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Only non-passthrough requests are split by the block layer and use the
->bio_split bio_set. Move it from the request_queue to the gendisk.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20220727162300.3089193-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The double indirect bio leads to somewhat suboptimal code generation.
Instead return the (original or split) bio, and make sure the
request_queue arguments to the lower level helpers is passed after the
bio to avoid constant reshuffling of the argument passing registers.
Also give it and the helpers used to implement it more descriptive names.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220727162300.3089193-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Pull block updates from Jens Axboe:
- Improve the type checking of request flags (Bart)
- Ensure queue mapping for a single queues always picks the right queue
(Bart)
- Sanitize the io priority handling (Jan)
- rq-qos race fix (Jinke)
- Reserved tags handling improvements (John)
- Separate memory alignment from file/disk offset aligment for O_DIRECT
(Keith)
- Add new ublk driver, userspace block driver using io_uring for
communication with the userspace backend (Ming)
- Use try_cmpxchg() to cleanup the code in various spots (Uros)
- Finally remove bdevname() (Christoph)
- Clean up the zoned device handling (Christoph)
- Clean up independent access range support (Christoph)
- Clean up and improve block sysfs handling (Christoph)
- Clean up and improve teardown of block devices.
This turns the usual two step process into something that is simpler
to implement and handle in block drivers (Christoph)
- Clean up chunk size handling (Christoph)
- Misc cleanups and fixes (Bart, Bo, Dan, GuoYong, Jason, Keith, Liu,
Ming, Sebastian, Yang, Ying)
* tag 'for-5.20/block-2022-07-29' of git://git.kernel.dk/linux-block: (178 commits)
ublk_drv: fix double shift bug
ublk_drv: make sure that correct flags(features) returned to userspace
ublk_drv: fix error handling of ublk_add_dev
ublk_drv: fix lockdep warning
block: remove __blk_get_queue
block: call blk_mq_exit_queue from disk_release for never added disks
blk-mq: fix error handling in __blk_mq_alloc_disk
ublk: defer disk allocation
ublk: rewrite ublk_ctrl_get_queue_affinity to not rely on hctx->cpumask
ublk: fold __ublk_create_dev into ublk_ctrl_add_dev
ublk: cleanup ublk_ctrl_uring_cmd
ublk: simplify ublk_ch_open and ublk_ch_release
ublk: remove the empty open and release block device operations
ublk: remove UBLK_IO_F_PREFLUSH
ublk: add a MAINTAINERS entry
block: don't allow the same type rq_qos add more than once
mmc: fix disk/queue leak in case of adding disk failure
ublk_drv: fix an IS_ERR() vs NULL check
ublk: remove UBLK_IO_F_INTEGRITY
ublk_drv: remove unneeded semicolon
...
|
|
Use the new blk_opf_t type for arguments and variables that represent
request flags or a bitwise combination of a request operation and
request flags. Rename the function arguments and also a structure member
that hold a request operation and flags from 'rw' into 'opf'.
This patch does not change any functionality.
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20220714180729.1065367-7-bvanassche@acm.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Improve static type checking by changing the type of the value returned by
req_op() and bio_op() from unsigned int into enum req_op. Insert
'default: break;' in switch statements on the enum req_op type to prevent
that the compiler warns about these switch statements.
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Tim Waugh <tim@cyberelk.net>
Cc: Alasdair Kergon <agk@redhat.com>
Cc: Mike Snitzer <snitzer@kernel.org>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20220714180729.1065367-5-bvanassche@acm.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The commit 513616843d73 ("block: remove superfluous calls to
blkcg_bio_issue_init") has removed blkcg_bio_issue_init from
__bio_clone since submit_bio will override ->bi_issue.
However, __blk_queue_split is called after blkcg_bio_issue_init
(see blk_mq_submit_bio) in submit_bio. In this case, the
->bi_issue is 0. Fix it.
Fixes: 513616843d73 ("block: remove superfluous calls to blkcg_bio_issue_init")
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Link: https://lore.kernel.org/r/20220713140226.68135-1-songmuchun@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Drop the unused q argument, and invert the check to move the exception
into a branch and the regular path as the normal return.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20220706070350.1703384-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Now that blk_max_size_offset has a single caller left, fold it into that
and clean up the naming convention for the local variables there.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Pankaj Raghav <p.raghav@samsung.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20220614090934.570632-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
get_max_io_size has a very odd choice of variables names and
initialization patterns. Switch to more descriptive names and more
clear initialization of them.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220614090934.570632-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
blk_rq_get_max_sectors always uses q->limits.chunk_sectors as the
chunk_sectors argument, and already checks for max_sectors through the
call to blk_queue_get_max_sectors. That means much of
blk_max_size_offset is not needed and open coding it simplifies the code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20220614090934.570632-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Individual bv_len's may not be a sector size.
Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20220610195830.3574005-7-kbusch@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Pull NVMe write streams removal from Jens Axboe:
"This removes the write streams support in NVMe. No vendor ever really
shipped working support for this, and they are not interested in
supporting it.
With the NVMe support gone, we have nothing in the tree that supports
this. Remove passing around of the hints.
The only discussion point in this patchset imho is the fact that the
file specific write hint setting/getting fcntl helpers will now return
-1/EINVAL like they did before we supported write hints. No known
applications use these functions, I only know of one prototype that I
help do for RocksDB, and that's not used. That said, with a change
like this, it's always a bit controversial. Alternatively, we could
just make them return 0 and pretend it worked. It's placement based
hints after all"
* tag 'for-5.18/write-streams-2022-03-18' of git://git.kernel.dk/linux-block:
fs: remove fs.f_write_hint
fs: remove kiocb.ki_hint
block: remove the per-bio/request write hint
nvme: remove support or stream based temperature hint
|
|
Pull SCSI updates from James Bottomley:
"This series consists of the usual driver updates (qla2xxx, pm8001,
libsas, smartpqi, scsi_debug, lpfc, iscsi, mpi3mr) plus minor updates
and bug fixes.
The high blast radius core update is the removal of write same, which
affects block and several non-SCSI devices. The other big change,
which is more local, is the removal of the SCSI pointer"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (281 commits)
scsi: scsi_ioctl: Drop needless assignment in sg_io()
scsi: bsg: Drop needless assignment in scsi_bsg_sg_io_fn()
scsi: lpfc: Copyright updates for 14.2.0.0 patches
scsi: lpfc: Update lpfc version to 14.2.0.0
scsi: lpfc: SLI path split: Refactor BSG paths
scsi: lpfc: SLI path split: Refactor Abort paths
scsi: lpfc: SLI path split: Refactor SCSI paths
scsi: lpfc: SLI path split: Refactor CT paths
scsi: lpfc: SLI path split: Refactor misc ELS paths
scsi: lpfc: SLI path split: Refactor VMID paths
scsi: lpfc: SLI path split: Refactor FDISC paths
scsi: lpfc: SLI path split: Refactor LS_RJT paths
scsi: lpfc: SLI path split: Refactor LS_ACC paths
scsi: lpfc: SLI path split: Refactor the RSCN/SCR/RDF/EDC/FARPR paths
scsi: lpfc: SLI path split: Refactor PLOGI/PRLI/ADISC/LOGO paths
scsi: lpfc: SLI path split: Refactor base ELS paths and the FLOGI path
scsi: lpfc: SLI path split: Introduce lpfc_prep_wqe
scsi: lpfc: SLI path split: Refactor fast and slow paths to native SLI4
scsi: lpfc: SLI path split: Refactor lpfc_iocbq
scsi: lpfc: Use kcalloc()
...
|
|
blk-iocost and iolatency are cgroup aware rq-qos policies but they didn't
disable merges across different cgroups. This obviously can lead to
accounting and control errors but more importantly to priority inversions -
e.g. an IO which belongs to a higher priority cgroup or IO class may end up
getting throttled incorrectly because it gets merged to an IO issued from a
low priority cgroup.
Fix it by adding blk_cgroup_mergeable() which is called from merge paths and
rejects cross-cgroup and cross-issue_as_root merges.
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: d70675121546 ("block: introduce blk-iolatency io controller")
Cc: stable@vger.kernel.org # v4.19+
Cc: Josef Bacik <jbacik@fb.com>
Link: https://lore.kernel.org/r/Yi/eE/6zFNyWJ+qd@slm.duckdns.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Song reports that a RAID rebuild workload runs much slower recently,
and it is seeing a lot less merging than it did previously. The reason
is that a previous commit reduced the amount of work we do for plug
merging. RAID rebuild interleaves requests between disks, so a last-entry
check in plug merging always misses a merge opportunity since we always
find a different disk than what we are looking for.
Modify the logic such that it's still a one-hit cache, but ensure that
we check enough to find the right target before giving up.
Fixes: d38a9c04c0d5 ("block: only check previous entry for plug merge attempt")
Reported-and-tested-by: Song Liu <song@kernel.org>
Reviewed-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
With the NVMe support for this gone, there are no consumers of these hints
left, so remove them.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220304175556.407719-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
No more users of REQ_OP_WRITE_SAME or drivers implementing it are left,
so remove the infrastructure.
[mkp: fold in and tweak sysfs reporting fix]
Link: https://lore.kernel.org/r/20220209082828.2629273-8-hch@lst.de
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
|
|
Commit 111be8839817 ("block-throttle: avoid double charge") marks bio as
BIO_THROTTLED unconditionally if __blk_throtl_bio() is called on this bio,
then this bio won't be called into __blk_throtl_bio() any more. This way
is to avoid double charge in case of bio splitting. It is reasonable for
read/write throughput limit, but not reasonable for IOPS limit because
block layer provides io accounting against split bio.
Chunguang Xu has already observed this issue and fixed it in commit
4f1e9630afe6 ("blk-throtl: optimize IOPS throttle for large IO scenarios").
However, that patch only covers bio splitting in __blk_queue_split(), and
we have other kind of bio splitting, such as bio_split() &
submit_bio_noacct() and other ways.
This patch tries to fix the issue in one generic way by always charging
the bio for iops limit in blk_throtl_bio(). This way is reasonable:
re-submission & fast-cloned bio is charged if it is submitted to same
disk/queue, and BIO_THROTTLED will be cleared if bio->bi_bdev is changed.
This new approach can get much more smooth/stable iops limit compared with
commit 4f1e9630afe6 ("blk-throtl: optimize IOPS throttle for large IO
scenarios") since that commit can't throttle current split bios actually.
Also this way won't cause new double bio iops charge in
blk_throtl_dispatch_work_fn() in which blk_throtl_bio() won't be called
any more.
Reported-by: Ning Li <lining2020x@163.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Chunguang Xu <brookxu@tencent.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20220216044514.2903784-7-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
There is a 1:1 relationship between request_queues and gendisks now, so
no need for these extra checks.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Not needed, shift it into the source files that need it instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
No needed, shift it into the source files that need it instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
blk_mq_submit_bio has two different plug cases, one that uses full
plugging and a limited plugging one.
The limited plugging case is only used for a corner case that does
not matter in real life:
- no ->commit_rqs (so not NVMe)
- no shared tags (so not SCSI)
- not rotational (so no old disk or floppy driver)
- must have multiple queues (so no eMMC)
Remove the limited merging case and all the related junk to simplify
blk_mq_submit_bio and the functions called from it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123160443.1315598-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
It is obvious that io merge can't be done between two different queues, so
just try to run io merge in case of same queue.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20211102133502.3619184-2-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Convert bdev->bd_disk->queue to bdev_get_queue(), which is faster.
Apparently, there are a few such spots in block that got lost during
rebases.
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Use a singly linked list for the blk_plug. This saves 8 bytes in the
blk_plug struct, and makes for faster list manipulations than doubly
linked lists. As we don't use the doubly linked lists for anything,
singly linked is just fine.
This yields a bump in default (merging enabled) performance from 7.0
to 7.1M IOPS, and ~7.5M IOPS with merging disabled.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Instead of returning the same queue request through a request pointer,
use a boolean to accomplish the same.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The fast path is no splitting needed. Separate the handling into a
check part we can inline, and an out-of-line handling path if we do
need to split.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Unlike the RWF_HIPRI userspace ABI which is intentionally kept vague,
the bio flag is specific to the polling implementation, so rename and
document it properly.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Link: https://lore.kernel.org/r/20211012111226.760968-12-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Currently we scan the entire plug list, which is potentially very
expensive. In an IOPS bound workload, we can drive about 5.6M IOPS with
merging enabled, and profiling shows that the plug merge check is the
(by far) most expensive thing we're doing:
Overhead Command Shared Object Symbol
+ 20.89% io_uring [kernel.vmlinux] [k] blk_attempt_plug_merge
+ 4.98% io_uring [kernel.vmlinux] [k] io_submit_sqes
+ 4.78% io_uring [kernel.vmlinux] [k] blkdev_direct_IO
+ 4.61% io_uring [kernel.vmlinux] [k] blk_mq_submit_bio
Instead of browsing the whole list, just check the previously inserted
entry. That is enough for a naive merge check and will catch most cases,
and for devices that need full merging, the IO scheduler attached to
such devices will do that anyway. The plug merge is meant to be an
inexpensive check to avoid getting a request, but if we repeatedly
scan the list for every single insert, it is very much not a cheap
check.
With this patch, the workload instead runs at ~7.0M IOPS, providing
a 25% improvement. Disabling merging entirely yields another 5%
improvement.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
Yu Kuai
Damien Le Moal
John Garry
Bart Van Assche
Christoph Hellwig
Eric Biggers
Linus Torvalds
Uday Shankar
Jens Axboe
Ming Lei
Muchun Song
Keith Busch
Tejun Heo
Pavel Begunkov
