Before commit 780db2071a(blk-mq: decouble blk-mq freezing
from generic bypassing), the dying flag is checked before
entering queue, and Tejun converts the checking into .mq_freeze_depth,
and assumes the counter is increased just after dying flag
is set. Unfortunately we doesn't do that in blk_set_queue_dying().
This patch calls blk_freeze_queue_start() in blk_set_queue_dying(),
so that we can block new I/O coming once the queue is set as dying.
Given blk_set_queue_dying() is always called in remove path
of block device, and queue will be cleaned up later, we don't
need to worry about undoing the counter.
Cc: Tejun Heo <tj@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
As the .q_usage_counter is used by both legacy and
mq path, we need to block new I/O if queue becomes
dead in blk_queue_enter().
So rename it and we can use this function in both
paths.
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Without the barrier, reading DEAD flag of .q_usage_counter
and reading .mq_freeze_depth may be reordered, then the
following wait_event_interruptible() may never return.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This patch adds comment on two races related with
timeout handler:
- requeue from queue busy vs. timeout
- rq free & reallocation vs. timeout
Both the races themselves and current solution aren't
explicit enough, so add comments on them.
Cc: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
This patch kills the warning reported on powerpc_pseries,
and actually we don't need the initialization.
After merging the md tree, today's linux-next build (powerpc
pseries_le_defconfig) produced this warning:
drivers/md/raid1.c: In function 'raid1d':
drivers/md/raid1.c:2172:9: warning: 'page_len$' may be used uninitialized in this function [-Wmaybe-uninitialized]
if (memcmp(page_address(ppages[j]),
^
drivers/md/raid1.c:2160:7: note: 'page_len$' was declared here
int page_len[RESYNC_PAGES];
^
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
All reshape I/O share pages from 1st copy device, so just use that pages
for avoiding direct access to bvec table in handle_reshape_read_error.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now one page array is allocated for each resync bio, and we can
retrieve page from this table directly.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now we allocate one page array for managing resync pages, instead
of using bio's vec table to do that, and the old way is very hacky
and won't work any more if multipage bvec is enabled.
The introduced cost is that we need to allocate (128 + 16) * copies
bytes per r10_bio, and it is fine because the inflight r10_bio for
resync shouldn't be much, as pointed by Shaohua.
Also bio_reset() in raid10_sync_request() and reshape_request()
are removed because all bios are freshly new now in these functions
and not necessary to reset any more.
This patch can be thought as cleanup too.
Suggested-by: Shaohua Li <shli@kernel.org>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
reshape read request is a bit special and requires one extra
bio which isn't allocated from r10buf_pool.
Refactor the .bi_end_io for read reshape, so that we can use
raid10's resync page mangement approach easily in the following
patches.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
This patch improve handling of write behind in the following ways:
- introduce behind master bio to hold all write behind pages
- fast clone bios from behind master bio
- avoid to change bvec table directly
- use bio_copy_data() and make code more clean
Suggested-by: Shaohua Li <shli@fb.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The 'offset' local variable can't be changed inside the loop, so
move it out.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Turns out we can use bio_copy_data in raid1's write behind,
and we can make alloc_behind_pages() more clean/efficient,
but we need to partial version of bio_copy_data().
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now one page array is allocated for each resync bio, and we can
retrieve page from this table directly.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now we allocate one page array for managing resync pages, instead
of using bio's vec table to do that, and the old way is very hacky
and won't work any more if multipage bvec is enabled.
The introduced cost is that we need to allocate (128 + 16) * raid_disks
bytes per r1_bio, and it is fine because the inflight r1_bio for
resync shouldn't be much, as pointed by Shaohua.
Also the bio_reset() in raid1_sync_request() is removed because
all bios are freshly new now and not necessary to reset any more.
This patch can be thought as a cleanup too
Suggested-by: Shaohua Li <shli@kernel.org>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
This patch gets each page's reference of each bio for resync,
then r1buf_pool_free() gets simplified a lot.
The same policy has been taken in raid10's buf pool allocation/free
too.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Now resync I/O use bio's bec table to manage pages,
this way is very hacky, and may not work any more
once multipage bvec is introduced.
So introduce helpers and new data structure for
managing resync I/O pages more cleanly.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Both raid1 and raid10 share common resync
block size and page count, so move them into md.h.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
All bio_add_page() is for adding one page into resync bio,
which is big enough to hold RESYNC_PAGES pages, and
the current bio_add_page() doesn't check queue limit any more,
so it won't fail at all.
remove unused label (shaohua)
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
It is obviously that hctx->cpumask is per hctx, and both
share same lifetime, so this patch moves freeing of hctx->cpumask
into release handler of hctx's kobject.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Tested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
This patch removes kobject_put() over hctx in __blk_mq_unregister_dev(),
and trys to keep lifetime consistent between hctx and hctx's kobject.
Now blk_mq_sysfs_register() and blk_mq_sysfs_unregister() become
totally symmetrical, and kobject's refcounter drops to zero just
when the hctx is freed.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Tested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
Currently from kobject view, both q->mq_kobj and ctx->kobj can
be released during one cycle of blk_mq_register_dev() and
blk_mq_unregister_dev(). Actually, sw queue's lifetime is
same with its request queue's, which is covered by request_queue->kobj.
So we don't need to call kobject_put() for the two kinds of
kobject in __blk_mq_unregister_dev(), instead we do that
in release handler of request queue.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Tested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
Firstly bio_clone_mddev() is used in raid normal I/O and isn't
in resync I/O path.
Secondly all the direct access to bvec table in raid happens on
resync I/O except for write behind of raid1, in which we still
use bio_clone() for allocating new bvec table.
So this patch replaces bio_clone() with bio_clone_fast()
in bio_clone_mddev().
Also kill bio_clone_mddev() and call bio_clone_fast() directly, as
suggested by Christoph Hellwig.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
mddev is never NULL and neither is ->bio_set, so
remove the check.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Write behind need to replace pages in bio's bvecs, and we have
to clone a fresh bio with new bvec table, so use the introduced
bio_clone_bioset_partial() for it.
For other bio_clone_mddev() cases, we will use fast clone since
they don't need to touch bvec table.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The current behaviour is to fall back to allocate
bio from 'fs_bio_set', that isn't a correct way
because it might cause deadlock.
So this patch simply return failure if mddev->bio_set
can't be created.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
md still need bio clone(not the fast version) for behind write,
and it is more efficient to use bio_clone_bioset_partial().
The idea is simple and just copy the bvecs range specified from
parameters.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Shaohua Li <shli@fb.com>
The check on bio->bi_vcnt doesn't make sense in erase_end_io().
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Also this patch simplify the code a bit.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Always bio_add_page() is the standard and preferred way to
do the task.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Instead we use standard iterator way to do that.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
When the bio is full, bio_add_pc_page() will return zero,
so use this information tell when the bio is full.
Also replace access to .bi_vcnt for pr_debug() with bio_segments().
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@fb.com>
For a non-cloned bio, bio_add_page() only returns failure when
the io vec table is full, but in that case, bio->bi_vcnt can't
be zero at all.
So remove the impossible failure handling.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Some drivers often use external bvec table, so introduce
this helper for this case. It is always safe to access the
bio->bi_io_vec in this way for this case.
After converting to this usage, it will becomes a bit easier
to evaluate the remaining direct access to bio->bi_io_vec,
so it can help to prepare for the following multipage bvec
support.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Fixed up the new O_DIRECT cases.
Signed-off-by: Jens Axboe <axboe@fb.com>
Use bio_add_page(), the standard interface for adding a page to a bio,
rather than open-coding the same.
It should be noted that the 'clone' bio that is allocated using
bio_alloc_bioset(), in crypt_alloc_buffer(), does _not_ set the
bio's BIO_CLONED flag. As such, bio_add_page()'s early return for true
bio clones (those with BIO_CLONED set) isn't applicable.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Firstly we have mature bvec/bio iterator helper for iterate each
page in one bio, not necessary to reinvent a wheel to do that.
Secondly the coming multipage bvecs requires this patch.
Also add comments about the direct access to bvec table.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Avoid accessing .bi_vcnt directly, because the bio can be split from
block layer and .bi_vcnt should never have been used here.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
This patch fixes one issue reported by Kent, which can
be triggered in bcachefs over sata disk. Actually it
is a generic issue in block flush vs. blk-tag.
Cc: Christoph Hellwig <hch@infradead.org>
Reported-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The local variable of 'devname' in scsi_report_lun_scan() isn't used any
more, so remove it.
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Currently q->mq_ops is used widely to decide if the queue
is mq or not, so we should set the 'flag' asap so that both
block core and drivers can get the correct mq info.
For example, commit 868f2f0b720(blk-mq: dynamic h/w context count)
moves the hctx's initialization before setting q->mq_ops in
blk_mq_init_allocated_queue(), then cause blk_alloc_flush_queue()
to think the queue is non-mq and don't allocate command size
for the per-hctx flush rq.
This patches should fix the problem reported by Sasha.
Cc: Keith Busch <keith.busch@intel.com>
Reported-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Fixes: 868f2f0b72 ("blk-mq: dynamic h/w context count")
Signed-off-by: Jens Axboe <axboe@fb.com>
After commit e36f62042880(block: split bios to maxpossible length),
bio can be splitted in the middle of a vector entry, then it
is easy to split out one bio which size isn't aligned with block
size, especially when the block size is bigger than 512.
This patch fixes the issue by making the max io size aligned
to logical block size.
Fixes: e36f62042880(block: split bios to maxpossible length)
Reported-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Cc: Keith Busch <keith.busch@intel.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Both htab_map_update_elem() and htab_map_delete_elem() can be
called from eBPF program, and they may be in kernel hot path,
so it isn't efficient to use a per-hashtable lock in this two
helpers.
The per-hashtable spinlock is used for protecting bucket's
hlist, and per-bucket lock is just enough. This patch converts
the per-hashtable lock into per-bucket spinlock, so that
contention can be decreased a lot.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The spinlock is just used for protecting the per-bucket
hlist, so it isn't needed for selecting bucket.
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preparing for removing global per-hashtable lock, so
the counter need to be defined as aotmic_t first.
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Percpu refcount is the perfect match for partition's case,
and the conversion is quite straight.
With the convertion, one pair of atomic inc/dec can be saved
for accounting block I/O, which is run in hot path of block I/O.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
So the helper can be used in both generic partition
case and part0 case.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Now blk_cleanup_queue() can be called before calling
del_gendisk()[1], inside which hctx->ctxs is touched
from blk_mq_unregister_hctx(), but the variable has
been freed by blk_cleanup_queue() at that time.
So this patch moves freeing of hctx->ctxs into queue's
release handler for fixing the oops reported by Stefan.
[1], 6cd18e711d (block: destroy bdi before blockdev is
unregistered)
Reported-by: Stefan Seyfried <stefan.seyfried@googlemail.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: stable@vger.kernel.org (v4.0)
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
For cloned bio, bio->bi_vcnt can't be used at all, and we
have resort to bio_segments() to figure out how many
segment there are in the bio.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The problem is introduced by commit 764f612c6c3c231b(blk-merge:
don't compute bi_phys_segments from bi_vcnt for cloned bio),
and merge is needed if number of current segment isn't less than
max segments.
Strictly speaking, bio->bi_vcnt shouldn't be used here since
it may not be accurate in cases of both cloned bio or bio cloned
from, but bio_segments() is a bit expensive, and bi_vcnt is still
the biggest number, so the approach should work.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This way will become consistent with non-mq case, also
avoid to update rq->deadline twice for mq.
The comment said: "We do this early, to ensure we are on
the right CPU.", but no percpu stuff is used in blk_add_timer(),
so it isn't necessary. Even when inserting from plug list, there
is no such guarantee at all.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
blk_mq_tag_to_rq() needs to be able to tell if it should return
the original request, or the flush request if we are doing a flush
sequence. Clear the flush tag when IO completes for a flush, since
that is what we are comparing against.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
'struct blk_mq_ctx' is __percpu, so add the annotation
and fix the sparse warning reported from Fengguang:
[block:for-linus 2/3] block/blk-mq.h:75:16: sparse: incorrect
type in initializer (different address spaces)
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
blk_mq_put_ctx() has to be called before io_schedule() in
bt_get().
This patch fixes the problem by taking similar approach from
percpu_ida allocation for the situation.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The percpu counter is only used for blk-mq, so move
its allocation and free inside blk-mq, and don't
allocate it for legacy queue device.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
blk_mq_exit_hw_queues() and blk_mq_free_hw_queues()
are introduced to avoid code duplication.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
When there isn't enough vring descriptor for adding to vq,
blk-mq will be put as stopped state until some of pending
descriptors are completed & freed.
Unfortunately, the vq's interrupt may come just before
blk-mq's BLK_MQ_S_STOPPED flag is set, so the blk-mq will
still be kept as stopped even though lots of descriptors
are completed and freed in the interrupt handler. The worst
case is that all pending descriptors are freed in the
interrupt handler, and the queue is kept as stopped forever.
This patch fixes the problem by starting/stopping blk-mq
with holding vq_lock.
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
Conflicts:
drivers/block/virtio_blk.c
hctx->ctx_map should have been freed inside blk_mq_free_queue().
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Access to tgt->req_vq is strictly serialized by spin_lock
of tgt->tgt_lock, so the ACCESS_ONCE() isn't necessary.
smp_read_barrier_depends() in virtscsi_req_done was introduced
to order reading req_vq and decreasing tgt->reqs, but it isn't
needed now because req_vq is read from
scsi->req_vqs[vq->index - VIRTIO_SCSI_VQ_BASE] instead of
tgt->req_vq, so remove the unnecessary barrier.
Also remove related comment about the barrier.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
When there isn't enough vring descriptor for adding to vq,
blk-mq will be put as stopped state until some of pending
descriptors are completed & freed.
Unfortunately, the vq's interrupt may come just before
blk-mq's BLK_MQ_S_STOPPED flag is set, so the blk-mq will
still be kept as stopped even though lots of descriptors
are completed and freed in the interrupt handler. The worst
case is that all pending descriptors are freed in the
interrupt handler, and the queue is kept as stopped forever.
This patch fixes the problem by starting/stopping blk-mq
with holding vq_lock.
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Both nr_cache and nr_tags arn't needed for bitmap tag anymore.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The selected tag should be selected at random between 0 and
(depth - 1) with probability 1/depth, instead between 0 and
(depth - 2) with probability 1/(depth - 1).
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The barrier isn't necessary because both atomic_dec_and_test()
and wake_up() implicate one barrier.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The unlock memory barrier need to order access to req in free
path and clearing tag bit, otherwise either request free path
may see a allocated request, or initialized request in allocate
path might be modified by the ongoing free path.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
entry(cmd->ll_list) may belong to new request once end_cmd()
returns, so fix the bug with the patch.
Without the change, it is easy to observe oops when
doing null_blk(timer) test.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The patch basically reverts the patch of(blk-mq:
initialize request on allocation) in Jens's tree(already
in -next), and only initialize req->q in allocation
for two reasons:
- presumed cache hotness on completion
- blk_rq_tagged(rq) depends on reset of req->mq_ctx
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
type of set->tags is struct blk_mq_tags **.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Commit 8146875de7 (arm, kvm: Fix CPU hotplug callback registration)
holds the lock before calling the two functions:
kvm_vgic_hyp_init()
kvm_timer_hyp_init()
and both the two functions are calling register_cpu_notifier()
to register cpu notifier, so cause double lock on cpu_add_remove_lock.
Considered that both two functions are only called inside
kvm_arch_init() with holding cpu_add_remove_lock, so simply use
__register_cpu_notifier() to fix the problem.
Fixes: 8146875de7 (arm, kvm: Fix CPU hotplug callback registration)
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
__percpu_counter_add() may be called in softirq/hardirq handler (such
as, blk_mq_queue_exit() is typically called in hardirq/softirq handler),
so we need to call this_cpu_add()(irq safe helper) to update percpu
counter, otherwise counts may be lost.
This fixes the problem that 'rmmod null_blk' hangs in blk_cleanup_queue()
because of miscounting of request_queue->mq_usage_counter.
This patch is the v1 of previous one of "lib/percpu_counter.c:
disable local irq when updating percpu couter", and takes Andrew's
approach which may be more efficient for ARCHs(x86, s390) that
have optimized this_cpu_add().
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Shaohua Li <shli@fusionio.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fan Du <fan.du@windriver.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When queue_mode is NULL_Q_MQ and null_blk is being removed,
blk_cleanup_queue() isn't called to cleanup queue, so the queue
allocated won't be freed.
This patch calls blk_cleanup_queue() for MQ to drain all pending
requests first and release the reference counter of queue kobject, then
blk_mq_free_queue() will be called in queue kobject's release handler
when queue kobject's reference counter drops to zero.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
blk_rq_init() is called in req's complete handler to initialize
the request, so the members of start_time and start_time_ns might
become inaccurate when it is allocated in future.
The patch initializes the two members in blk_mq_rq_ctx_init() to
fix the problem.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_mq_free_queue() is called from release handler of
queue kobject, so it needn't be called from drivers.
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This patch moves synchronization on mq->delay_work
from blk_mq_free_queue() to blk_sync_queue(), so that
blk_sync_queue can work on mq.
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_mq_drain_queue() is introduced so that we can drain
mq queue inside blk_cleanup_queue().
Also don't accept new requests any more if queue is marked
as dying.
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Actually CONFIG_PAGE_OFFSET isn't same with PAGE_OFFSET, so
it isn't easy to figue out PAGE_OFFSET defined in header
file from scripts.
Because CONFIG_PAGE_OFFSET may not be defined in some ARCHs(
64bit ARCH), or defined as bogus value in !MMU case, so
this patch only applys the filter on ARM when CONFIG_PAGE_OFFSET
is defined as the original problem is only on ARM.
Cc: <stable@vger.kernel.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Fixes: f6537f2f0e
Singed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If accounting is on, we will do the IO completion accounting after
we have freed the request. Fix that by moving it sooner instead.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This patch uses CONFIG_PAGE_OFFSET to filter symbols which
are not in kernel address space because these symbols are
generally for generating code purpose and can't be run at
kernel mode, so we needn't keep them in /proc/kallsyms.
For example, on ARM there are some symbols which may be
linked in relocatable code section, then perf can't parse
symbols any more from /proc/kallsyms, this patch fixes the
problem (introduced b9b32bf70f)
Cc: Russell King <linux@arm.linux.org.uk>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@vger.kernel.org
Currently flush_dcache_page() thinks pages as non-mapped if
mapping_mapped(mapping) return false. This approach is very
coase:
- mmap on part of file may cause all pages backed on
the file being thought as mmaped
- file-backed pages aren't mapped into user space actually
if the memory mmaped on the file isn't accessed
This patch uses page_mapped() to decide if the page has been
mapped.
From the attached test code, I find there is much performance
improvement(>25%) when accessing page caches via read under this
situations, so memcpy benefits a lot from not flushing cache
under this situation.
No. read time without the patch No. read time with the patch
================================================================
No. 0, time 22615636 us No. 0, time 22014717 us
No. 1, time 4387851 us No. 1, time 3113184 us
No. 2, time 4276535 us No. 2, time 3005244 us
No. 3, time 4259821 us No. 3, time 3001565 us
No. 4, time 4263811 us No. 4, time 3002748 us
No. 5, time 4258486 us No. 5, time 3004104 us
No. 6, time 4253009 us No. 6, time 3002188 us
No. 7, time 4262809 us No. 7, time 2998196 us
No. 8, time 4264525 us No. 8, time 3007255 us
No. 9, time 4267795 us No. 9, time 3005094 us
1), No.0. is to read the file from storage device, and others are
to read the file from page caches basically.
2), file size is 512M, and is on ext4 over usb mass storage.
3), the test is done on Pandaboard.
unsigned int sum = 0;
unsigned long sum_val = 0;
static unsigned long tv_diff(struct timeval *tv1, struct timeval *tv2)
{
return (tv2->tv_sec - tv1->tv_sec) * 1000000 +
(tv2->tv_usec - tv1->tv_usec);
}
int main(int argc, char *argv[])
{
char *mbuf, fbuf;
int fd;
int i;
unsigned long page_size, size;
struct stat stat;
struct timeval t1, t2;
unsigned char *rbuf = malloc(32 * page_size);
if (!rbuf) {
printf(" %sn", "malloc failed");
exit(-1);
}
page_size = getpagesize();
fd = open(argv[1], O_RDWR);
assert(fd >= 0);
fstat(fd, &stat);
size = stat.st_size;
printf("%s: file %s, size %lu, page size %lun",
argv[0],
argv[1], size, page_size);
gettimeofday(&t1, NULL);
mbuf = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
if (!mbuf) {
printf(" %sn", "mmap failed");
exit(-1);
}
for (i = 0 ; i < size ; i += (page_size * 32)) {
int rcnt;
lseek(fd, i, SEEK_SET);
rcnt = read(fd, rbuf, page_size * 32);
if (rcnt != page_size * 32) {
printf("%s: read faildn", __func__);
exit(-1);
}
}
free(rbuf);
munmap(mbuf, size);
gettimeofday(&t2, NULL);
printf("tread mmaped time: %luusn", tv_diff(&t1, &t2));
close(fd);
}
Cc: Michel Lespinasse <walken@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Nicolas Pitre <nicolas.pitre@linaro.org>
Reviewed-by: Will Deacon <will.deacon@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
It is common for one sg to include many pages, so mark all these
pages as clean to avoid unnecessary flushing on them in
set_pte_at() or update_mmu_cache().
The patch might improve loading performance of applciation code a bit.
On the below test code to read file(~1GByte size) from usb mass storage
disk to buffer created with mmap(PROT_READ | PROT_EXEC) on
Pandaboard, average ~1% improvement can be observed with the patch on
10 times test.
unsigned int sum = 0;
static unsigned long tv_diff(struct timeval *tv1, struct timeval *tv2)
{
return (tv2->tv_sec - tv1->tv_sec) * 1000000 + (tv2->tv_usec - tv1->tv_usec);
}
int main(int argc, char *argv[])
{
char *mbuffer;
int fd;
int i;
unsigned long page_size, size;
struct stat stat;
struct timeval t1, t2;
page_size = getpagesize();
fd = open(argv[1], O_RDONLY);
assert(fd >= 0);
fstat(fd, &stat);
size = stat.st_size;
printf("%s: file %s, file size %lu, page size %lun", argv[0],
read_filename, size, page_size);
gettimeofday(&t1, NULL);
mbuffer = mmap(NULL, size, PROT_READ | PROT_EXEC, MAP_SHARED, fd, 0);
for (i = 0 ; i < size ; i += page_size)
sum += mbuffer[i];
munmap(mbuffer, page_size);
gettimeofday(&t2, NULL);
printf("tread mmaped time: %luusn", tv_diff(&t1, &t2));
close(fd);
}
Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Commit 14318efb(ARM: 7587/1: implement optimized percpu variable access)
introduces arm's __my_cpu_offset to optimize percpu vaiable access,
which really works well on hackbench, but will cause __my_cpu_offset
to return garbage value before it is initialized in cpu_init() called
by setup_arch, so accessing percpu variable before setup_arch may cause
kernel hang. But generic __my_cpu_offset always returns zero before
percpu area is brought up, and won't hang kernel.
So the patch tries to clear __my_cpu_offset on boot CPU early
to avoid boot hang.
At least now percpu variable is accessed by lockdep before
setup_arch(), and enabling CONFIG_LOCK_STAT or CONFIG_DEBUG_LOCKDEP
can trigger kernel hang.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
This patch documents the firmware cache mechanism so that
users of request_firmware() know that it can be called
safely inside device's suspend and resume callback, and
the device's firmware needn't be cached any more by individual
driver itself to deal with firmware loss during system resume.
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
device_del can happen anytime, so once it happens,
the devres of the device will be freed inside device_del, but
drivers can't know it has been deleted and may still add
resources into the device, so memory leak is caused.
This patch moves the devres_release_all to fix the problem.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The line below in intr_complete isn't needed,
memset(urb->transfer_buffer, 0, urb->transfer_buffer_length);
so just remove it.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove declaration for intr_complete so that ctags may be happy to
decrease duplicated symbols, also decrease one line code.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The flag of EVENT_DEV_WAKING is not used any more, so just remove it.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
usb_device is parent device of usb_interface in the view of driver
model, so its reference count is always held during .probe/.disconnect
of usb_interface instance.
This patch just removes the unnecessay usb_get/put_dev.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ming Lei