My workload is a raid5 which had 16 disks. And used our filesystem to
write using direct-io mode.
I used the blktrace to find those message:
8,16 0 6647 2.453665504 2579 M W 7493152 + 8 [md0_raid5]
8,16 0 6648 2.453672411 2579 Q W 7493160 + 8 [md0_raid5]
8,16 0 6649 2.453672606 2579 M W 7493160 + 8 [md0_raid5]
8,16 0 6650 2.453679255 2579 Q W 7493168 + 8 [md0_raid5]
8,16 0 6651 2.453679441 2579 M W 7493168 + 8 [md0_raid5]
8,16 0 6652 2.453685948 2579 Q W 7493176 + 8 [md0_raid5]
8,16 0 6653 2.453686149 2579 M W 7493176 + 8 [md0_raid5]
8,16 0 6654 2.453693074 2579 Q W 7493184 + 8 [md0_raid5]
8,16 0 6655 2.453693254 2579 M W 7493184 + 8 [md0_raid5]
8,16 0 6656 2.453704290 2579 Q W 7493192 + 8 [md0_raid5]
8,16 0 6657 2.453704482 2579 M W 7493192 + 8 [md0_raid5]
8,16 0 6658 2.453715016 2579 Q W 7493200 + 8 [md0_raid5]
8,16 0 6659 2.453715247 2579 M W 7493200 + 8 [md0_raid5]
8,16 0 6660 2.453721730 2579 Q W 7493208 + 8 [md0_raid5]
8,16 0 6661 2.453721974 2579 M W 7493208 + 8 [md0_raid5]
8,16 0 6662 2.453728202 2579 Q W 7493216 + 8 [md0_raid5]
8,16 0 6663 2.453728436 2579 M W 7493216 + 8 [md0_raid5]
8,16 0 6664 2.453734782 2579 Q W 7493224 + 8 [md0_raid5]
8,16 0 6665 2.453735019 2579 M W 7493224 + 8 [md0_raid5]
8,16 0 6666 2.453741401 2579 Q W 7493232 + 8 [md0_raid5]
8,16 0 6667 2.453741632 2579 M W 7493232 + 8 [md0_raid5]
8,16 0 6668 2.453748148 2579 Q W 7493240 + 8 [md0_raid5]
8,16 0 6669 2.453748386 2579 M W 7493240 + 8 [md0_raid5]
8,16 0 6670 2.453851843 2579 I W 7493144 + 104 [md0_raid5]
8,16 0 0 2.453853661 0 m N cfq2579 insert_request
8,16 0 6671 2.453854064 2579 I W 7493120 + 24 [md0_raid5]
8,16 0 0 2.453854439 0 m N cfq2579 insert_request
8,16 0 6672 2.453854793 2579 U N [md0_raid5] 2
8,16 0 0 2.453855513 0 m N cfq2579 Not idling.st->count:1
8,16 0 0 2.453855927 0 m N cfq2579 dispatch_insert
8,16 0 0 2.453861771 0 m N cfq2579 dispatched a request
8,16 0 0 2.453862248 0 m N cfq2579 activate rq,drv=1
8,16 0 6673 2.453862332 2579 D W 7493120 + 24 [md0_raid5]
8,16 0 0 2.453865957 0 m N cfq2579 Not idling.st->count:1
8,16 0 0 2.453866269 0 m N cfq2579 dispatch_insert
8,16 0 0 2.453866707 0 m N cfq2579 dispatched a request
8,16 0 0 2.453867061 0 m N cfq2579 activate rq,drv=2
8,16 0 6674 2.453867145 2579 D W 7493144 + 104 [md0_raid5]
8,16 0 6675 2.454147608 0 C W 7493120 + 24 [0]
8,16 0 0 2.454149357 0 m N cfq2579 complete rqnoidle 0
8,16 0 6676 2.454791505 0 C W 7493144 + 104 [0]
8,16 0 0 2.454794803 0 m N cfq2579 complete rqnoidle 0
8,16 0 0 2.454795160 0 m N cfq schedule dispatch
From above messages,we can find rq[W 7493144 + 104] and rq[W
7493120 + 24] do not merge.
Because the bio order is:
8,16 0 6638 2.453619407 2579 Q W 7493144 + 8 [md0_raid5]
8,16 0 6639 2.453620460 2579 G W 7493144 + 8 [md0_raid5]
8,16 0 6640 2.453639311 2579 Q W 7493120 + 8 [md0_raid5]
8,16 0 6641 2.453639842 2579 G W 7493120 + 8 [md0_raid5]
The bio(7493144) first and bio(7493120) later.So the subsequent
bios will be divided into two parts.
When flushing plug-list,because elv_attempt_insert_merge only support
backmerge,not supporting frontmerge.
So rq[7493120 + 24] can't merge with rq[7493144 + 104].
From my test,i found those situation can count 25% in our system.
Using this patch, there is no this situation.
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
CC:Shaohua Li <shli@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull block IO update from Jens Axboe:
"Core block IO bits for 3.7. Not a huge round this time, it contains:
- First series from Kent cleaning up and generalizing bio allocation
and freeing.
- WRITE_SAME support from Martin.
- Mikulas patches to prevent O_DIRECT crashes when someone changes
the block size of a device.
- Make bio_split() work on data-less bio's (like trim/discards).
- A few other minor fixups."
Fixed up silent semantic mis-merge as per Mikulas Patocka and Andrew
Morton. It is due to the VM no longer using a prio-tree (see commit
6b2dbba8b6ac: "mm: replace vma prio_tree with an interval tree").
So make set_blocksize() use mapping_mapped() instead of open-coding the
internal VM knowledge that has changed.
* 'for-3.7/core' of git://git.kernel.dk/linux-block: (26 commits)
block: makes bio_split support bio without data
scatterlist: refactor the sg_nents
scatterlist: add sg_nents
fs: fix include/percpu-rwsem.h export error
percpu-rw-semaphore: fix documentation typos
fs/block_dev.c:1644:5: sparse: symbol 'blkdev_mmap' was not declared
blockdev: turn a rw semaphore into a percpu rw semaphore
Fix a crash when block device is read and block size is changed at the same time
block: fix request_queue->flags initialization
block: lift the initial queue bypass mode on blk_register_queue() instead of blk_init_allocated_queue()
block: ioctl to zero block ranges
block: Make blkdev_issue_zeroout use WRITE SAME
block: Implement support for WRITE SAME
block: Consolidate command flag and queue limit checks for merges
block: Clean up special command handling logic
block/blk-tag.c: Remove useless kfree
block: remove the duplicated setting for congestion_threshold
block: reject invalid queue attribute values
block: Add bio_clone_bioset(), bio_clone_kmalloc()
block: Consolidate bio_alloc_bioset(), bio_kmalloc()
...
Pull workqueue changes from Tejun Heo:
"This is workqueue updates for v3.7-rc1. A lot of activities this
round including considerable API and behavior cleanups.
* delayed_work combines a timer and a work item. The handling of the
timer part has always been a bit clunky leading to confusing
cancelation API with weird corner-case behaviors. delayed_work is
updated to use new IRQ safe timer and cancelation now works as
expected.
* Another deficiency of delayed_work was lack of the counterpart of
mod_timer() which led to cancel+queue combinations or open-coded
timer+work usages. mod_delayed_work[_on]() are added.
These two delayed_work changes make delayed_work provide interface
and behave like timer which is executed with process context.
* A work item could be executed concurrently on multiple CPUs, which
is rather unintuitive and made flush_work() behavior confusing and
half-broken under certain circumstances. This problem doesn't
exist for non-reentrant workqueues. While non-reentrancy check
isn't free, the overhead is incurred only when a work item bounces
across different CPUs and even in simulated pathological scenario
the overhead isn't too high.
All workqueues are made non-reentrant. This removes the
distinction between flush_[delayed_]work() and
flush_[delayed_]_work_sync(). The former is now as strong as the
latter and the specified work item is guaranteed to have finished
execution of any previous queueing on return.
* In addition to the various bug fixes, Lai redid and simplified CPU
hotplug handling significantly.
* Joonsoo introduced system_highpri_wq and used it during CPU
hotplug.
There are two merge commits - one to pull in IRQ safe timer from
tip/timers/core and the other to pull in CPU hotplug fixes from
wq/for-3.6-fixes as Lai's hotplug restructuring depended on them."
Fixed a number of trivial conflicts, but the more interesting conflicts
were silent ones where the deprecated interfaces had been used by new
code in the merge window, and thus didn't cause any real data conflicts.
Tejun pointed out a few of them, I fixed a couple more.
* 'for-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq: (46 commits)
workqueue: remove spurious WARN_ON_ONCE(in_irq()) from try_to_grab_pending()
workqueue: use cwq_set_max_active() helper for workqueue_set_max_active()
workqueue: introduce cwq_set_max_active() helper for thaw_workqueues()
workqueue: remove @delayed from cwq_dec_nr_in_flight()
workqueue: fix possible stall on try_to_grab_pending() of a delayed work item
workqueue: use hotcpu_notifier() for workqueue_cpu_down_callback()
workqueue: use __cpuinit instead of __devinit for cpu callbacks
workqueue: rename manager_mutex to assoc_mutex
workqueue: WORKER_REBIND is no longer necessary for idle rebinding
workqueue: WORKER_REBIND is no longer necessary for busy rebinding
workqueue: reimplement idle worker rebinding
workqueue: deprecate __cancel_delayed_work()
workqueue: reimplement cancel_delayed_work() using try_to_grab_pending()
workqueue: use mod_delayed_work() instead of __cancel + queue
workqueue: use irqsafe timer for delayed_work
workqueue: clean up delayed_work initializers and add missing one
workqueue: make deferrable delayed_work initializer names consistent
workqueue: cosmetic whitespace updates for macro definitions
workqueue: deprecate system_nrt[_freezable]_wq
workqueue: deprecate flush[_delayed]_work_sync()
...
A queue newly allocated with blk_alloc_queue_node() has only
QUEUE_FLAG_BYPASS set. For request-based drivers,
blk_init_allocated_queue() is called and q->queue_flags is overwritten
with QUEUE_FLAG_DEFAULT which doesn't include BYPASS even though the
initial bypass is still in effect.
In blk_init_allocated_queue(), or QUEUE_FLAG_DEFAULT to q->queue_flags
instead of overwriting.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: stable@vger.kernel.org
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
b82d4b197c ("blkcg: make request_queue bypassing on allocation") made
request_queues bypassed on allocation to avoid switching on and off
bypass mode on a queue being initialized. Some drivers allocate and
then destroy a lot of queues without fully initializing them and
incurring bypass latency overhead on each of them could add upto
significant overhead.
Unfortunately, blk_init_allocated_queue() is never used by queues of
bio-based drivers, which means that all bio-based driver queues are in
bypass mode even after initialization and registration complete
successfully.
Due to the limited way request_queues are used by bio drivers, this
problem is hidden pretty well but it shows up when blk-throttle is
used in combination with a bio-based driver. Trying to configure
(echoing to cgroupfs file) blk-throttle for a bio-based driver hangs
indefinitely in blkg_conf_prep() waiting for bypass mode to end.
This patch moves the initial blk_queue_bypass_end() call from
blk_init_allocated_queue() to blk_register_queue() which is called for
any userland-visible queues regardless of its type.
I believe this is correct because I don't think there is any block
driver which needs or wants working elevator and blk-cgroup on a queue
which isn't visible to userland. If there are such users, we need a
different solution.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Joseph Glanville <joseph.glanville@orionvm.com.au>
Cc: stable@vger.kernel.org
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The WRITE SAME command supported on some SCSI devices allows the same
block to be efficiently replicated throughout a block range. Only a
single logical block is transferred from the host and the storage device
writes the same data to all blocks described by the I/O.
This patch implements support for WRITE SAME in the block layer. The
blkdev_issue_write_same() function can be used by filesystems and block
drivers to replicate a buffer across a block range. This can be used to
efficiently initialize software RAID devices, etc.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
- blk_check_merge_flags() verifies that cmd_flags / bi_rw are
compatible. This function is called for both req-req and req-bio
merging.
- blk_rq_get_max_sectors() and blk_queue_get_max_sectors() can be used
to query the maximum sector count for a given request or queue. The
calls will return the right value from the queue limits given the
type of command (RW, discard, write same, etc.)
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Remove special-casing of non-rw fs style requests (discard). The nomerge
flags are consolidated in blk_types.h, and rq_mergeable() and
bio_mergeable() have been modified to use them.
bio_is_rw() is used in place of bio_has_data() a few places. This is
done to to distinguish true reads and writes from other fs type requests
that carry a payload (e.g. write same).
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Before call the blk_queue_congestion_threshold(),
the blk_queue_congestion_threshold() is already called at blk_queue_make_rquest().
Because this code is the duplicated, it has removed.
Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Previously, there was bio_clone() but it only allocated from the fs bio
set; as a result various users were open coding it and using
__bio_clone().
This changes bio_clone() to become bio_clone_bioset(), and then we add
bio_clone() and bio_clone_kmalloc() as wrappers around it, making use of
the functionality the last patch adedd.
This will also help in a later patch changing how bio cloning works.
Signed-off-by: Kent Overstreet <koverstreet@google.com>
CC: Jens Axboe <axboe@kernel.dk>
CC: NeilBrown <neilb@suse.de>
CC: Alasdair Kergon <agk@redhat.com>
CC: Boaz Harrosh <bharrosh@panasas.com>
CC: Jeff Garzik <jeff@garzik.org>
Acked-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now that we've got generic code for freeing bios allocated from bio
pools, this isn't needed anymore.
This patch also makes bio_free() static, since without bi_destructor
there should be no need for it to be called anywhere else.
bio_free() is now only called from bio_put, so we can refactor those a
bit - move some code from bio_put() to bio_free() and kill the redundant
bio->bi_next = NULL.
v5: Switch to BIO_KMALLOC_POOL ((void *)~0), per Boaz
v6: BIO_KMALLOC_POOL now NULL, drop bio_free's EXPORT_SYMBOL
v7: No #define BIO_KMALLOC_POOL anymore
Signed-off-by: Kent Overstreet <koverstreet@google.com>
CC: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now that bios keep track of where they were allocated from,
bio_integrity_alloc_bioset() becomes redundant.
Remove bio_integrity_alloc_bioset() and drop bio_set argument from the
related functions and make them use bio->bi_pool.
Signed-off-by: Kent Overstreet <koverstreet@google.com>
CC: Jens Axboe <axboe@kernel.dk>
CC: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When performing a cable pull test w/ active stress I/O using fio over
a dual port Intel 82599 FCoE CNA, w/ 256LUNs on one port and about 32LUNs
on the other, it is observed that the system becomes not usable due to
scsi-ml being busy printing the error messages for all the failing commands.
I don't believe this problem is specific to FCoE and these commands are
anyway failing due to link being down (DID_NO_CONNECT), just rate-limit
the messages here to solve this issue.
v2->v1: use __ratelimit() as Tomas Henzl mentioned as the proper way for
rate-limit per function. However, in this case, the failed i/o gets to
blk_end_request_err() and then blk_update_request(), which also has to
be rate-limited, as added in the v2 of this patch.
v3-v2: resolved conflict to apply on current 3.6-rc3 upstream tip.
Signed-off-by: Yi Zou <yi.zou@intel.com>
Cc: www.Open-FCoE.org <devel@open-fcoe.org>
Cc: Tomas Henzl <thenzl@redhat.com>
Cc: <linux-scsi@vger.kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now that cancel_delayed_work() can be safely called from IRQ handlers,
there's no reason to use __cancel_delayed_work(). Use
cancel_delayed_work() instead of __cancel_delayed_work() and mark the
latter deprecated.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: Jiri Kosina <jkosina@suse.cz>
Cc: Roland Dreier <roland@kernel.org>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Now that mod_delayed_work() is safe to call from IRQ handlers,
__cancel_delayed_work() followed by queue_delayed_work() can be
replaced with mod_delayed_work().
Most conversions are straight-forward except for the following.
* net/core/link_watch.c: linkwatch_schedule_work() was doing a quite
elaborate dancing around its delayed_work. Collapse it such that
linkwatch_work is queued for immediate execution if LW_URGENT and
existing timer is kept otherwise.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
This will allow md/raid to know why the unplug was called,
and will be able to act according - if !from_schedule it
is safe to perform tasks which could themselves schedule.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
MD raid1 prepares to dispatch request in unplug callback. If make_request in
low level queue also uses unplug callback to dispatch request, the low level
queue's unplug callback will not be called. Recheck the callback list helps
this case.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Both md and umem has similar code for getting notified on an
blk_finish_plug event.
Centralize this code in block/ and allow each driver to
provide its distinctive difference.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, request_queue has one request_list to allocate requests
from regardless of blkcg of the IO being issued. When the unified
request pool is used up, cfq proportional IO limits become meaningless
- whoever grabs the next request being freed wins the race regardless
of the configured weights.
This can be easily demonstrated by creating a blkio cgroup w/ very low
weight, put a program which can issue a lot of random direct IOs there
and running a sequential IO from a different cgroup. As soon as the
request pool is used up, the sequential IO bandwidth crashes.
This patch implements per-blkg request_list. Each blkg has its own
request_list and any IO allocates its request from the matching blkg
making blkcgs completely isolated in terms of request allocation.
* Root blkcg uses the request_list embedded in each request_queue,
which was renamed to @q->root_rl from @q->rq. While making blkcg rl
handling a bit harier, this enables avoiding most overhead for root
blkcg.
* Queue fullness is properly per request_list but bdi isn't blkcg
aware yet, so congestion state currently just follows the root
blkcg. As writeback isn't aware of blkcg yet, this works okay for
async congestion but readahead may get the wrong signals. It's
better than blkcg completely collapsing with shared request_list but
needs to be improved with future changes.
* After this change, each block cgroup gets a full request pool making
resource consumption of each cgroup higher. This makes allowing
non-root users to create cgroups less desirable; however, note that
allowing non-root users to directly manage cgroups is already
severely broken regardless of this patch - each block cgroup
consumes kernel memory and skews IO weight (IO weights are not
hierarchical).
v2: queue-sysfs.txt updated and patch description udpated as suggested
by Vivek.
v3: blk_get_rl() wasn't checking error return from
blkg_lookup_create() and may cause oops on lookup failure. Fix it
by falling back to root_rl on blkg lookup failures. This problem
was spotted by Rakesh Iyer <rni@google.com>.
v4: Updated to accomodate 458f27a982 "block: Avoid missed wakeup in
request waitqueue". blk_drain_queue() now wakes up waiters on all
blkg->rl on the target queue.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Request allocation is about to be made per-blkg meaning that there'll
be multiple request lists.
* Make queue full state per request_list. blk_*queue_full() functions
are renamed to blk_*rl_full() and takes @rl instead of @q.
* Rename blk_init_free_list() to blk_init_rl() and make it take @rl
instead of @q. Also add @gfp_mask parameter.
* Add blk_exit_rl() instead of destroying rl directly from
blk_release_queue().
* Add request_list->q and make request alloc/free functions -
blk_free_request(), [__]freed_request(), __get_request() - take @rl
instead of @q.
This patch doesn't introduce any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add q->nr_rqs[] which currently behaves the same as q->rq.count[] and
move q->rq.elvpriv to q->nr_rqs_elvpriv. blk_drain_queue() is updated
to use q->nr_rqs[] instead of q->rq.count[].
These counters separates queue-wide request statistics from the
request list and allow implementation of per-queue request allocation.
While at it, properly indent fields of struct request_list.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Block layer very lazy allocation of ioc. It waits until the moment
ioc is absolutely necessary; unfortunately, that time could be inside
queue lock and __get_request() performs unlock - try alloc - retry
dancing.
Just allocate it up-front on entry to block layer. We're not saving
the rain forest by deferring it to the last possible moment and
complicating things unnecessarily.
This patch is to prepare for further updates to request allocation
path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, there are two request allocation functions - get_request()
and get_request_wait(). The former tries to allocate a request once
and the latter keeps retrying until it succeeds. The latter wraps the
former and keeps retrying until allocation succeeds.
The combination of two functions deliver fallible non-wait allocation,
fallible wait allocation and unfailing wait allocation. However,
given that forward progress is guaranteed, fallible wait allocation
isn't all that useful and in fact nobody uses it.
This patch simplifies the interface as follows.
* get_request() is renamed to __get_request() and is only used by the
wrapper function.
* get_request_wait() is renamed to get_request(). It now takes
@gfp_mask and retries iff it contains %__GFP_WAIT.
This patch doesn't introduce any functional change and is to prepare
for further updates to request allocation path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
mempool_create_node() currently assumes %GFP_KERNEL. Its only user,
blk_init_free_list(), is about to be updated to use other allocation
flags - add @gfp_mask argument to the function.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Commit 777eb1bf15 disconnects externally
supplied queue_lock before blk_drain_queue(). Switching the lock would
introduce lock unbalance because theads which have taken the external
lock might unlock the internal lock in the during the queue drain. This
patch mitigate this by disconnecting the lock after the queue draining
since queue draining makes a lot of request_queue users go away.
However, please note, this patch only makes the problem less likely to
happen. Anyone who still holds a ref might try to issue a new request on
a dead queue after the blk_cleanup_queue() finishes draining, the lock
unbalance might still happen in this case.
=====================================
[ BUG: bad unlock balance detected! ]
3.4.0+ #288 Not tainted
-------------------------------------
fio/17706 is trying to release lock (&(&q->__queue_lock)->rlock) at:
[<ffffffff81329372>] blk_queue_bio+0x2a2/0x380
but there are no more locks to release!
other info that might help us debug this:
1 lock held by fio/17706:
#0: (&(&vblk->lock)->rlock){......}, at: [<ffffffff81327f1a>]
get_request_wait+0x19a/0x250
stack backtrace:
Pid: 17706, comm: fio Not tainted 3.4.0+ #288
Call Trace:
[<ffffffff81329372>] ? blk_queue_bio+0x2a2/0x380
[<ffffffff810dea49>] print_unlock_inbalance_bug+0xf9/0x100
[<ffffffff810dfe4f>] lock_release_non_nested+0x1df/0x330
[<ffffffff811dae24>] ? dio_bio_end_aio+0x34/0xc0
[<ffffffff811d6935>] ? bio_check_pages_dirty+0x85/0xe0
[<ffffffff811daea1>] ? dio_bio_end_aio+0xb1/0xc0
[<ffffffff81329372>] ? blk_queue_bio+0x2a2/0x380
[<ffffffff81329372>] ? blk_queue_bio+0x2a2/0x380
[<ffffffff810e0079>] lock_release+0xd9/0x250
[<ffffffff81a74553>] _raw_spin_unlock_irq+0x23/0x40
[<ffffffff81329372>] blk_queue_bio+0x2a2/0x380
[<ffffffff81328faa>] generic_make_request+0xca/0x100
[<ffffffff81329056>] submit_bio+0x76/0xf0
[<ffffffff8115470c>] ? set_page_dirty_lock+0x3c/0x60
[<ffffffff811d69e1>] ? bio_set_pages_dirty+0x51/0x70
[<ffffffff811dd1a8>] do_blockdev_direct_IO+0xbf8/0xee0
[<ffffffff811d8620>] ? blkdev_get_block+0x80/0x80
[<ffffffff811dd4e5>] __blockdev_direct_IO+0x55/0x60
[<ffffffff811d8620>] ? blkdev_get_block+0x80/0x80
[<ffffffff811d92e7>] blkdev_direct_IO+0x57/0x60
[<ffffffff811d8620>] ? blkdev_get_block+0x80/0x80
[<ffffffff8114c6ae>] generic_file_aio_read+0x70e/0x760
[<ffffffff810df7c5>] ? __lock_acquire+0x215/0x5a0
[<ffffffff811e9924>] ? aio_run_iocb+0x54/0x1a0
[<ffffffff8114bfa0>] ? grab_cache_page_nowait+0xc0/0xc0
[<ffffffff811e82cc>] aio_rw_vect_retry+0x7c/0x1e0
[<ffffffff811e8250>] ? aio_fsync+0x30/0x30
[<ffffffff811e9936>] aio_run_iocb+0x66/0x1a0
[<ffffffff811ea9b0>] do_io_submit+0x6f0/0xb80
[<ffffffff8134de2e>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[<ffffffff811eae50>] sys_io_submit+0x10/0x20
[<ffffffff81a7c9e9>] system_call_fastpath+0x16/0x1b
Changes since v2: Update commit log to explain how the code is still
broken even if we delay the lock switching after the drain.
Changes since v1: Update commit log as Tejun suggested.
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Asias He <asias@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
After hot-unplug a stressed disk, I found that rl->wait[] is not empty
while rl->count[] is empty and there are theads still sleeping on
get_request after the queue cleanup. With simple debug code, I found
there are exactly nr_sleep - nr_wakeup of theads in D state. So there
are missed wakeup.
$ dmesg | grep nr_sleep
[ 52.917115] ---> nr_sleep=1046, nr_wakeup=873, delta=173
$ vmstat 1
1 173 0 712640 24292 96172 0 0 0 0 419 757 0 0 0 100 0
To quote Tejun:
Ah, okay, freed_request() wakes up single waiter with the assumption
that after the wakeup there will at least be one successful allocation
which in turn will continue the wakeup chain until the wait list is
empty - ie. waiter wakeup is dependent on successful request
allocation happening after each wakeup. With queue marked dead, any
woken up waiter fails the allocation path, so the wakeup chaining is
lost and we're left with hung waiters. What we need is wake_up_all()
after drain completion.
This patch fixes the missed wakeup by waking up all the theads which
are sleeping on wait queue after queue drain.
Changes in v2: Drop waitqueue_active() optimization
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Asias He <asias@redhat.com>
Fixed a bug by me, where stacked devices would oops on calling
blk_drain_queue() since ->rq.wait[] do not get initialized unless
it's a full queue setup.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)
iQEcBAABAgAGBQJPnb50AAoJEHm+PkMAQRiGAE0H/A4zFZIUGmF3miKPDYmejmrZ
oVDYxVAu6JHjHWhu8E3VsinvyVscowjV8dr15eSaQzmDmRkSHAnUQ+dB7Di7jLC2
MNopxsWjwyZ8zvvr3rFR76kjbWKk/1GYytnf7GPZLbJQzd51om2V/TY/6qkwiDSX
U8Tt7ihSgHAezefqEmWp2X/1pxDCEt+VFyn9vWpkhgdfM1iuzF39MbxSZAgqDQ/9
JJrBHFXhArqJguhENwL7OdDzkYqkdzlGtS0xgeY7qio2CzSXxZXK4svT6FFGA8Za
xlAaIvzslDniv3vR2ZKd6wzUwFHuynX222hNim3QMaYdXm012M+Nn1ufKYGFxI0=
=4d4w
-----END PGP SIGNATURE-----
Merge tag 'v3.4-rc5' into for-3.5/core
The core branch is behind driver commits that we want to build
on for 3.5, hence I'm pulling in a later -rc.
Linux 3.4-rc5
Conflicts:
Documentation/feature-removal-schedule.txt
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Request allocation is mempool backed to guarantee forward progress
under memory pressure; unfortunately, this property got broken while
adding elvpriv data. Failures during elvpriv allocation, including
ioc and icq creation failures, currently make get_request() fail as
whole. There's no forward progress guarantee for these allocations -
they may fail indefinitely under memory pressure stalling IO and
deadlocking the system.
This patch updates get_request() such that elvpriv allocation failure
doesn't make the whole function fail. If elvpriv allocation fails,
the allocation is degraded into !ELVPRIV. This will force the request
to ELEVATOR_INSERT_BACK disturbing scheduling but elvpriv alloc
failures should be rare (nothing is per-request) and anything is
better than deadlocking.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Allocation failure handling in get_request() is about to be updated.
To ease the update, collapse blk_alloc_request() into get_request().
This patch doesn't introduce any functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
With the previous change to guarantee bypass visiblity for RCU read
lock regions, entering bypass mode involves non-trivial overhead and
future changes are scheduled to make use of bypass mode during init
path. Combined it may end up adding noticeable delay during boot.
This patch makes request_queue start its life in bypass mode, which is
ended on queue init completion at the end of
blk_init_allocated_queue(), and updates blk_queue_bypass_start() such
that draining and RCU synchronization are performed only when the
queue actually enters bypass mode.
This avoids unnecessarily switching in and out of bypass mode during
init avoiding the overhead and any nasty surprises which may step from
leaving bypass mode on half-initialized queues.
The boot time overhead was pointed out by Vivek.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, blkg_lookup() doesn't check @q bypass state. This patch
updates blk_queue_bypass_start() to do synchronize_rcu() before
returning and updates blkg_lookup() to check blk_queue_bypass() and
return %NULL if bypassing. This ensures blkg_lookup() returns %NULL
if @q is bypassing.
This is to guarantee that nobody is accessing policy data while @q is
bypassing, which is necessary to allow replacing blkio_cgroup->pd[] in
place on policy [de]activation.
v2: Added more comments explaining bypass guarantees as suggested by
Vivek.
v3: Added more comments explaining why there's no synchronize_rcu() in
blk_cleanup_queue() as suggested by Vivek.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We do auto block plug flush to reduce latency, the threshold is 16
requests. This works well if the task is accessing one or two drives.
The problem is if the task is accessing a raid 0 device and the raid
disk number is big, say 8 or 16, 16/8 = 2 or 16/16=1, we will have
heavy lock contention.
This patch makes the threshold per-disk based. The latency should be
still ok accessing one or two drives. The setup with application
accessing a lot of drives in the meantime uaually is big machine,
avoiding lock contention is more important, because any contention
will actually increase latency.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We should use the GFP flags that the caller specified instead of picking
our own. All the callers specify GFP_KERNEL so this doesn't make a
difference to how the kernel runs, it's just a cleanup.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IO scheduling and cgroup are tied to the issuing task via io_context
and cgroup of %current. Unfortunately, there are cases where IOs need
to be routed via a different task which makes scheduling and cgroup
limit enforcement applied completely incorrectly.
For example, all bios delayed by blk-throttle end up being issued by a
delayed work item and get assigned the io_context of the worker task
which happens to serve the work item and dumped to the default block
cgroup. This is double confusing as bios which aren't delayed end up
in the correct cgroup and makes using blk-throttle and cfq propio
together impossible.
Any code which punts IO issuing to another task is affected which is
getting more and more common (e.g. btrfs). As both io_context and
cgroup are firmly tied to task including userland visible APIs to
manipulate them, it makes a lot of sense to match up tasks to bios.
This patch implements bio_associate_current() which associates the
specified bio with %current. The bio will record the associated ioc
and blkcg at that point and block layer will use the recorded ones
regardless of which task actually ends up issuing the bio. bio
release puts the associated ioc and blkcg.
It grabs and remembers ioc and blkcg instead of the task itself
because task may already be dead by the time the bio is issued making
ioc and blkcg inaccessible and those are all block layer cares about.
elevator_set_req_fn() is updated such that the bio elvdata is being
allocated for is available to the elevator.
This doesn't update block cgroup policies yet. Further patches will
implement the support.
-v2: #ifdef CONFIG_BLK_CGROUP added around bio->bi_ioc dereference in
rq_ioc() to fix build breakage.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Kent Overstreet <koverstreet@google.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Make the following interface updates to prepare for future ioc related
changes.
* create_io_context() returning ioc only works for %current because it
doesn't increment ref on the ioc. Drop @task parameter from it and
always assume %current.
* Make create_io_context_slowpath() return 0 or -errno and rename it
to create_task_io_context().
* Make ioc_create_icq() take @ioc as parameter instead of assuming
that of %current. The caller, get_request(), is updated to create
ioc explicitly and then pass it into ioc_create_icq().
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
get_request() is structured a bit unusually in that failure path is
inlined in the usual flow with goto labels atop and inside it.
Relocate the error path to the end of the function.
This is to prepare for icq handling changes in get_request() and
doesn't introduce any behavior change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, blkg is per cgroup-queue-policy combination. This is
unnatural and leads to various convolutions in partially used
duplicate fields in blkg, config / stat access, and general management
of blkgs.
This patch make blkg's per cgroup-queue and let them serve all
policies. blkgs are now created and destroyed by blkcg core proper.
This will allow further consolidation of common management logic into
blkcg core and API with better defined semantics and layering.
As a transitional step to untangle blkg management, elvswitch and
policy [de]registration, all blkgs except the root blkg are being shot
down during elvswitch and bypass. This patch adds blkg_root_update()
to update root blkg in place on policy change. This is hacky and racy
but should be good enough as interim step until we get locking
simplified and switch over to proper in-place update for all blkgs.
-v2: Root blkgs need to be updated on elvswitch too and blkg_alloc()
comment wasn't updated according to the function change. Fixed.
Both pointed out by Vivek.
-v3: v2 updated blkg_destroy_all() to invoke update_root_blkg_pd() for
all policies. This freed root pd during elvswitch before the
last queue finished exiting and led to oops. Directly invoke
update_root_blkg_pd() only on BLKIO_POLICY_PROP from
cfq_exit_queue(). This also is closer to what will be done with
proper in-place blkg update. Reported by Vivek.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, specific policy implementations are responsible for
maintaining list and number of blkgs. This duplicates code
unnecessarily, and hinders factoring common code and providing blkcg
API with better defined semantics.
After this patch, request_queue hosts list heads and counters and blkg
has list nodes for both policies. This patch only relocates the
necessary fields and the next patch will actually move management code
into blkcg core.
Note that request_queue->blkg_list[] and ->nr_blkgs[] are hardcoded to
have 2 elements. This is to avoid include dependency and will be
removed by the next patch.
This patch doesn't introduce any behavior change.
-v2: Now unnecessary conditional on CONFIG_BLK_CGROUP_MODULE removed
as pointed out by Vivek.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently block core calls directly into blk-throttle for init, drain
and exit. This patch adds blkcg_{init|drain|exit}_queue() which wraps
the blk-throttle functions. This is to give more control and
visiblity to blkcg core layer for proper layering. Further patches
will add logic common to blkcg policies to the functions.
While at it, collapse blk_throtl_release() into blk_throtl_exit().
There's no reason to keep them separate.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For root blkg, blk_throtl_init() was using throtl_alloc_tg()
explicitly and cfq_init_queue() was manually initializing embedded
cfqd->root_group, adding unnecessarily different code paths to blkg
handling.
Make both use the usual blkio_group get functions - throtl_get_tg()
and cfq_get_cfqg() - for the root blkio_group too. Note that
blk_throtl_init() callsite is pushed downwards in
blk_alloc_queue_node() so that @q is sufficiently initialized for
throtl_get_tg().
This simplifies root blkg handling noticeably for cfq and will allow
further modularization of blkcg API.
-v2: Vivek pointed out that using cfq_get_cfqg() won't work if
CONFIG_CFQ_GROUP_IOSCHED is disabled. Fix it by factoring out
initialization of base part of cfqg into cfq_init_cfqg_base() and
alloc/init/free explicitly if !CONFIG_CFQ_GROUP_IOSCHED.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Extend queue bypassing such that dying queue is always bypassing and
blk-throttle is drained on bypass. With blkcg policies updated to
test blk_queue_bypass() instead of blk_queue_dead(), this ensures that
no bio or request is held by or going through blkcg policies on a
bypassing queue.
This will be used to implement blkg cleanup on elevator switches and
policy changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rename and extend elv_queisce_start/end() to
blk_queue_bypass_start/end() which are exported and supports nesting
via @q->bypass_depth. Also add blk_queue_bypass() to test bypass
state.
This will be further extended and used for blkio_group management.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, blk_cleanup_queue() doesn't call elv_drain_elevator() if
q->elevator doesn't exist; however, bio based drivers don't have
elevator initialized but can still use blk-throttle. This patch moves
q->elevator test inside blk_drain_queue() such that only
elv_drain_elevator() is skipped if !q->elevator.
-v2: loop can have registered queue which has NULL request_fn. Make
sure we don't call into __blk_run_queue() in such cases.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Vivek Goyal <vgoyal@redhat.com>
Fold in bug fix from Vivek.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Plug merge calls two elevator callbacks outside queue lock -
elevator_allow_merge_fn() and elevator_bio_merged_fn(). Although
attempt_plug_merge() suggests that elevator is guaranteed to be there
through the existing request on the plug list, nothing prevents plug
merge from calling into dying or initializing elevator.
For regular merges, bypass ensures elvpriv count to reach zero, which
in turn prevents merges as all !ELVPRIV requests get REQ_SOFTBARRIER
from forced back insertion. Plug merge doesn't check ELVPRIV, and, as
the requests haven't gone through elevator insertion yet, it doesn't
have SOFTBARRIER set allowing merges on a bypassed queue.
This, for example, leads to the following crash during elevator
switch.
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
IP: [<ffffffff813b34e9>] cfq_allow_merge+0x49/0xa0
PGD 112cbc067 PUD 115d5c067 PMD 0
Oops: 0000 [#1] PREEMPT SMP
CPU 1
Modules linked in: deadline_iosched
Pid: 819, comm: dd Not tainted 3.3.0-rc2-work+ #76 Bochs Bochs
RIP: 0010:[<ffffffff813b34e9>] [<ffffffff813b34e9>] cfq_allow_merge+0x49/0xa0
RSP: 0018:ffff8801143a38f8 EFLAGS: 00010297
RAX: 0000000000000000 RBX: ffff88011817ce28 RCX: ffff880116eb6cc0
RDX: 0000000000000000 RSI: ffff880118056e20 RDI: ffff8801199512f8
RBP: ffff8801143a3908 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000000 R12: ffff880118195708
R13: ffff880118052aa0 R14: ffff8801143a3d50 R15: ffff880118195708
FS: 00007f19f82cb700(0000) GS:ffff88011fc80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000008 CR3: 0000000112c6a000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process dd (pid: 819, threadinfo ffff8801143a2000, task ffff880116eb6cc0)
Stack:
ffff88011817ce28 ffff880118195708 ffff8801143a3928 ffffffff81391bba
ffff88011817ce28 ffff880118195708 ffff8801143a3948 ffffffff81391bf1
ffff88011817ce28 0000000000000000 ffff8801143a39a8 ffffffff81398e3e
Call Trace:
[<ffffffff81391bba>] elv_rq_merge_ok+0x4a/0x60
[<ffffffff81391bf1>] elv_try_merge+0x21/0x40
[<ffffffff81398e3e>] blk_queue_bio+0x8e/0x390
[<ffffffff81396a5a>] generic_make_request+0xca/0x100
[<ffffffff81396b04>] submit_bio+0x74/0x100
[<ffffffff811d45c2>] __blockdev_direct_IO+0x1ce2/0x3450
[<ffffffff811d0dc7>] blkdev_direct_IO+0x57/0x60
[<ffffffff811460b5>] generic_file_aio_read+0x6d5/0x760
[<ffffffff811986b2>] do_sync_read+0xe2/0x120
[<ffffffff81199345>] vfs_read+0xc5/0x180
[<ffffffff81199501>] sys_read+0x51/0x90
[<ffffffff81aeac12>] system_call_fastpath+0x16/0x1b
There are multiple ways to fix this including making plug merge check
ELVPRIV; however,
* Calling into elevator outside queue lock is confusing and
error-prone.
* Requests on plug list aren't known to the elevator. They aren't on
the elevator yet, so there's no elevator specific state to update.
* Given the nature of plug merges - collecting bio's for the same
purpose from the same issuer - elevator specific restrictions aren't
applicable.
So, simply don't call into elevator methods from plug merge by moving
elv_bio_merged() from bio_attempt_*_merge() to blk_queue_bio(), and
using blk_try_merge() in attempt_plug_merge().
This is based on Jens' patch to skip elevator_allow_merge_fn() from
plug merge.
Note that this makes per-cgroup merged stats skip plug merging.
Signed-off-by: Tejun Heo <tj@kernel.org>
LKML-Reference: <4F16F3CA.90904@kernel.dk>
Original-patch-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_rq_merge_ok() is the elevator-neutral part of merge eligibility
test. blk_try_merge() determines merge direction and expects the
caller to have tested elv_rq_merge_ok() previously.
elv_rq_merge_ok() now wraps blk_rq_merge_ok() and then calls
elv_iosched_allow_merge(). elv_try_merge() is removed and the two
callers are updated to call elv_rq_merge_ok() explicitly followed by
blk_try_merge(). While at it, make rq_merge_ok() functions return
bool.
This is to prepare for plug merge update and doesn't introduce any
behavior change.
This is based on Jens' patch to skip elevator_allow_merge_fn() from
plug merge.
Signed-off-by: Tejun Heo <tj@kernel.org>
LKML-Reference: <4F16F3CA.90904@kernel.dk>
Original-patch-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
put_io_context() performed a complex trylock dancing to avoid
deferring ioc release to workqueue. It was also broken on UP because
trylock was always assumed to succeed which resulted in unbalanced
preemption count.
While there are ways to fix the UP breakage, even the most
pathological microbench (forced ioc allocation and tight fork/exit
loop) fails to show any appreciable performance benefit of the
optimization. Strip it out. If there turns out to be workloads which
are affected by this change, simpler optimization from the discussion
thread can be applied later.
Signed-off-by: Tejun Heo <tj@kernel.org>
LKML-Reference: <1328514611.21268.66.camel@sli10-conroe>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
* 'for-3.3/core' of git://git.kernel.dk/linux-block: (37 commits)
Revert "block: recursive merge requests"
block: Stop using macro stubs for the bio data integrity calls
blockdev: convert some macros to static inlines
fs: remove unneeded plug in mpage_readpages()
block: Add BLKROTATIONAL ioctl
block: Introduce blk_set_stacking_limits function
block: remove WARN_ON_ONCE() in exit_io_context()
block: an exiting task should be allowed to create io_context
block: ioc_cgroup_changed() needs to be exported
block: recursive merge requests
block, cfq: fix empty queue crash caused by request merge
block, cfq: move icq creation and rq->elv.icq association to block core
block, cfq: restructure io_cq creation path for io_context interface cleanup
block, cfq: move io_cq exit/release to blk-ioc.c
block, cfq: move icq cache management to block core
block, cfq: move io_cq lookup to blk-ioc.c
block, cfq: move cfqd->icq_list to request_queue and add request->elv.icq
block, cfq: reorganize cfq_io_context into generic and cfq specific parts
block: remove elevator_queue->ops
block: reorder elevator switch sequence
...
Fix up conflicts in:
- block/blk-cgroup.c
Switch from can_attach_task to can_attach
- block/cfq-iosched.c
conflict with now removed cic index changes (we now use q->id instead)
While probing, fd sets up queue, probes hardware and tears down the
queue if probing fails. In the process, blk_drain_queue() kicks the
queue which failed to finish initialization and fd is unhappy about
that.
floppy0: no floppy controllers found
------------[ cut here ]------------
WARNING: at drivers/block/floppy.c:2929 do_fd_request+0xbf/0xd0()
Hardware name: To Be Filled By O.E.M.
VFS: do_fd_request called on non-open device
Modules linked in:
Pid: 1, comm: swapper Not tainted 3.2.0-rc4-00077-g5983fe2 #2
Call Trace:
[<ffffffff81039a6a>] warn_slowpath_common+0x7a/0xb0
[<ffffffff81039b41>] warn_slowpath_fmt+0x41/0x50
[<ffffffff813d657f>] do_fd_request+0xbf/0xd0
[<ffffffff81322b95>] blk_drain_queue+0x65/0x80
[<ffffffff81322c93>] blk_cleanup_queue+0xe3/0x1a0
[<ffffffff818a809d>] floppy_init+0xdeb/0xe28
[<ffffffff818a72b2>] ? daring+0x6b/0x6b
[<ffffffff810002af>] do_one_initcall+0x3f/0x170
[<ffffffff81884b34>] kernel_init+0x9d/0x11e
[<ffffffff810317c2>] ? schedule_tail+0x22/0xa0
[<ffffffff815dbb14>] kernel_thread_helper+0x4/0x10
[<ffffffff81884a97>] ? start_kernel+0x2be/0x2be
[<ffffffff815dbb10>] ? gs_change+0xb/0xb
Avoid it by making blk_drain_queue() kick queue iff dispatch queue has
something on it.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Ralf Hildebrandt <Ralf.Hildebrandt@charite.de>
Reported-by: Wu Fengguang <fengguang.wu@intel.com>
Tested-by: Sergei Trofimovich <slyich@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now block layer knows everything necessary to create and associate
icq's with requests. Move ioc_create_icq() to blk-ioc.c and update
get_request() such that, if elevator_type->icq_size is set, requests
are automatically associated with their matching icq's before
elv_set_request(). io_context reference is also managed by block core
on request alloc/free.
* Only ioprio/cgroup changed handling remains from cfq_get_cic().
Collapsed into cfq_set_request().
* This removes queue kicking on icq allocation failure (for now). As
icq allocation failure is rare and the only effect of queue kicking
achieved was possibily accelerating queue processing, this change
shouldn't be noticeable.
There is a larger underlying problem. Unlike request allocation,
icq allocation is not guaranteed to succeed eventually after
retries. The number of icq is unbound and thus mempool can't be the
solution either. This effectively adds allocation dependency on
memory free path and thus possibility of deadlock.
This usually wouldn't happen because icq allocation is not a hot
path and, even when the condition triggers, it's highly unlikely
that none of the writeback workers already has icq.
However, this is still possible especially if elevator is being
switched under high memory pressure, so we better get it fixed.
Probably the only solution is just bypassing elevator and appending
to dispatch queue on any elevator allocation failure.
* Comment added to explain how icq's are managed and synchronized.
This completes cleanup of io_context interface.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Most of icq management is about to be moved out of cfq into blk-ioc.
This patch prepares for it.
* Move cfqd->icq_list to request_queue->icq_list
* Make request explicitly point to icq instead of through elevator
private data. ->elevator_private[3] is replaced with sub struct elv
which contains icq pointer and priv[2]. cfq is updated accordingly.
* Meaningless clearing of ->elevator_private[0] removed from
elv_set_request(). At that point in code, the field was guaranteed
to be %NULL anyway.
This patch doesn't introduce any functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When called under queue_lock, current_io_context() triggers lockdep
warning if it hits allocation path. This is because io_context
installation is protected by task_lock which is not IRQ safe, so it
triggers irq-unsafe-lock -> irq -> irq-safe-lock -> irq-unsafe-lock
deadlock warning.
Given the restriction, accessor + creator rolled into one doesn't work
too well. Drop current_io_context() and let the users access
task->io_context directly inside queue_lock combined with explicit
creation using create_io_context().
Future ioc updates will further consolidate ioc access and the create
interface will be unexported.
While at it, relocate ioc internal interface declarations in blk.h and
add section comments before and after.
This patch does not introduce functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
* blk_get_queue() is peculiar in that it returns 0 on success and 1 on
failure instead of 0 / -errno or boolean. Update it such that it
returns %true on success and %false on failure.
* Make sure the caller checks for the return value.
* Separate out __blk_get_queue() which doesn't check whether @q is
dead and put it in blk.h. This will be used later.
This patch doesn't introduce any functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
cfq allocates per-queue id using ida and uses it to index cic radix
tree from io_context. Move it to q->id and allocate on queue init and
free on queue release. This simplifies cfq a bit and will allow for
further improvements of io context life-cycle management.
This patch doesn't introduce any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_insert_cloned_request(), blk_execute_rq_nowait() and
blk_flush_plug_list() either didn't check whether the queue was dead
or did it without holding queue_lock. Update them so that dead state
is checked while holding queue_lock.
AFAICS, this plugs all holes (requeue doesn't matter as the request is
transitioning atomically from in_flight to queued).
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When trying to drain all requests, blk_drain_queue() checked only
q->rq.count[]; however, this only tracks REQ_ALLOCED requests. This
patch updates blk_drain_queue() such that it looks at all the counters
and queues so that request_queue is actually empty on completion.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are a number of QUEUE_FLAG_DEAD tests. Add blk_queue_dead()
macro and use it.
This patch doesn't introduce any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The only user left for blk_insert_request() is sx8 and it can be
trivially switched to use blk_execute_rq_nowait() - special requests
aren't included in io stat and sx8 doesn't use block layer tagging.
Switch sx8 and kill blk_insert_requeset().
This patch doesn't introduce any functional difference.
Only compile tested.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
struct request_queue is allocated with __GFP_ZERO so its "node" field is
zero before initialization. This causes an oops if node 0 is offline in
the page allocator because its zonelists are not initialized. From Dave
Young's dmesg:
SRAT: Node 1 PXM 2 0-d0000000
SRAT: Node 1 PXM 2 100000000-330000000
SRAT: Node 0 PXM 1 330000000-630000000
Initmem setup node 1 0000000000000000-000000000affb000
...
Built 1 zonelists in Node order, mobility grouping on.
...
BUG: unable to handle kernel paging request at 0000000000001c08
IP: [<ffffffff8111c355>] __alloc_pages_nodemask+0xb5/0x870
and __alloc_pages_nodemask+0xb5 translates to a NULL pointer on
zonelist->_zonerefs.
The fix is to initialize q->node at the time of allocation so the correct
node is passed to the slab allocator later.
Since blk_init_allocated_queue_node() is no longer needed, merge it with
blk_init_allocated_queue().
[rientjes@google.com: changelog, initializing q->node]
Cc: stable@vger.kernel.org [2.6.37+]
Reported-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Tested-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
After flush plug list, the list has no request, so we need to add a
trace_block_plug().
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Reviewed-by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
get_request_wait() could sleep and flush the plug list. If the list is
already flushed, don't flush again.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Reviewed-by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_cleanup_queue() may be called before elevator is set up on a
queue which triggers the following oops.
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: [<ffffffff8125a69c>] elv_drain_elevator+0x1c/0x70
...
Pid: 830, comm: kworker/0:2 Not tainted 3.1.0-next-20111025_64+ #1590
Bochs Bochs
RIP: 0010:[<ffffffff8125a69c>] [<ffffffff8125a69c>] elv_drain_elevator+0x1c/0x70
...
Call Trace:
[<ffffffff8125da92>] blk_drain_queue+0x42/0x70
[<ffffffff8125db90>] blk_cleanup_queue+0xd0/0x1c0
[<ffffffff81469640>] md_free+0x50/0x70
[<ffffffff8126f43b>] kobject_release+0x8b/0x1d0
[<ffffffff81270d56>] kref_put+0x36/0xa0
[<ffffffff8126f2b7>] kobject_put+0x27/0x60
[<ffffffff814693af>] mddev_delayed_delete+0x2f/0x40
[<ffffffff81083450>] process_one_work+0x100/0x3b0
[<ffffffff8108527f>] worker_thread+0x15f/0x3a0
[<ffffffff81089937>] kthread+0x87/0x90
[<ffffffff81621834>] kernel_thread_helper+0x4/0x10
Fix it by making blk_cleanup_queue() check whether q->elevator is set
up before invoking blk_drain_queue.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A dm-multipath user reported[1] a problem when trying to boot
a kernel with commit 4853abaae7
(block: fix flush machinery for stacking drivers with differring
flush flags) applied. It turns out that an empty flush request
can be sent into blk_insert_flush. When the BUG_ON was fixed
to allow for this, I/O on the underlying device would stall. The
reason is that blk_insert_cloned_request does not kick the queue.
In the aforementioned commit, I had added a special case to
kick the queue if data was sent down but the queue flags did
not require a flush. A better solution is to push the queue
kick up into blk_insert_cloned_request.
This patch, along with a follow-on which fixes the BUG_ON, fixes
the issue reported.
[1] http://www.redhat.com/archives/dm-devel/2011-September/msg00154.html
Reported-by: Christophe Saout <christophe@saout.de>
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Acked-by: Tejun Heo <tj@kernel.org>
Stable note: 3.1
Cc: stable@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
bio originally has the functionality to set the complete cpu, but
it is broken.
Chirstoph said that "This code is unused, and from the all the
discussions lately pretty obviously broken. The only thing keeping
it serves is creating more confusion and possibly more bugs."
And Jens replied with "We can kill bio_set_completion_cpu(). I'm fine
with leaving cpu control to the request based drivers, they are the
only ones that can toggle the setting anyway".
So this patch tries to remove all the work of controling complete cpu
from a bio.
Cc: Shaohua Li <shaohua.li@intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
request_queue is refcounted but actually depdends on lifetime
management from the queue owner - on blk_cleanup_queue(), block layer
expects that there's no request passing through request_queue and no
new one will.
This is fundamentally broken. The queue owner (e.g. SCSI layer)
doesn't have a way to know whether there are other active users before
calling blk_cleanup_queue() and other users (e.g. bsg) don't have any
guarantee that the queue is and would stay valid while it's holding a
reference.
With delay added in blk_queue_bio() before queue_lock is grabbed, the
following oops can be easily triggered when a device is removed with
in-flight IOs.
sd 0:0:1:0: [sdb] Stopping disk
ata1.01: disabled
general protection fault: 0000 [#1] PREEMPT SMP
CPU 2
Modules linked in:
Pid: 648, comm: test_rawio Not tainted 3.1.0-rc3-work+ #56 Bochs Bochs
RIP: 0010:[<ffffffff8137d651>] [<ffffffff8137d651>] elv_rqhash_find+0x61/0x100
...
Process test_rawio (pid: 648, threadinfo ffff880019efa000, task ffff880019ef8a80)
...
Call Trace:
[<ffffffff8137d774>] elv_merge+0x84/0xe0
[<ffffffff81385b54>] blk_queue_bio+0xf4/0x400
[<ffffffff813838ea>] generic_make_request+0xca/0x100
[<ffffffff81383994>] submit_bio+0x74/0x100
[<ffffffff811c53ec>] dio_bio_submit+0xbc/0xc0
[<ffffffff811c610e>] __blockdev_direct_IO+0x92e/0xb40
[<ffffffff811c39f7>] blkdev_direct_IO+0x57/0x60
[<ffffffff8113b1c5>] generic_file_aio_read+0x6d5/0x760
[<ffffffff8118c1ca>] do_sync_read+0xda/0x120
[<ffffffff8118ce55>] vfs_read+0xc5/0x180
[<ffffffff8118cfaa>] sys_pread64+0x9a/0xb0
[<ffffffff81afaf6b>] system_call_fastpath+0x16/0x1b
This happens because blk_queue_cleanup() destroys the queue and
elevator whether IOs are in progress or not and DEAD tests are
sprinkled in the request processing path without proper
synchronization.
Similar problem exists for blk-throtl. On queue cleanup, blk-throtl
is shutdown whether it has requests in it or not. Depending on
timing, it either oopses or throttled bios are lost putting tasks
which are waiting for bio completion into eternal D state.
The way it should work is having the usual clear distinction between
shutdown and release. Shutdown drains all currently pending requests,
marks the queue dead, and performs partial teardown of the now
unnecessary part of the queue. Even after shutdown is complete,
reference holders are still allowed to issue requests to the queue
although they will be immmediately failed. The rest of teardown
happens on release.
This patch makes the following changes to make blk_queue_cleanup()
behave as proper shutdown.
* QUEUE_FLAG_DEAD is now set while holding both q->exit_mutex and
queue_lock.
* Unsynchronized DEAD check in generic_make_request_checks() removed.
This couldn't make any meaningful difference as the queue could die
after the check.
* blk_drain_queue() updated such that it can drain all requests and is
now called during cleanup.
* blk_throtl updated such that it checks DEAD on grabbing queue_lock,
drains all throttled bios during cleanup and free td when queue is
released.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
attempt_plug_merge() accesses elevator without holding queue_lock and
may call into ->elevator_bio_merge_fn(). The elvator is guaranteed to
be valid because it's accessed iff the plugged list has requests and
elevator is never exited with live requests, so as long as the
elevator method can deal with unlocked access, this is safe.
Explain the sync rules around attempt_plug_merge() and drop the
unnecessary @tsk parameter.
This patch doesn't introduce any functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently get_request[_wait]() allocates request whether queue is dead
or not. This patch makes get_request[_wait]() return NULL if @q is
dead. blk_queue_bio() is updated to fail the submitted bio if request
allocation fails. While at it, add docbook comments for
get_request[_wait]().
Note that the current code has rather unclear (there are spurious DEAD
tests scattered around) assumption that the owner of a queue
guarantees that no request travels block layer if the queue is dead
and this patch in itself doesn't change much; however, this will allow
fixing the broken assumption in the next patch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_throtl_bio() and throtl_get_tg() have rather unusual interface.
* throtl_get_tg() returns pointer to a valid tg or ERR_PTR(-ENODEV),
and drops queue_lock in the latter case. Different locking context
depending on return value is error-prone and DEAD state is scheduled
to be protected by queue_lock anyway. Move DEAD check inside
queue_lock and return valid tg or NULL.
* blk_throtl_bio() indicates return status both with its return value
and in/out param **@bio. The former is used to indicate whether
queue is found to be dead during throtl processing. The latter
whether the bio is throttled.
There's no point in returning DEAD check result from
blk_throtl_bio(). The queue can die after blk_throtl_bio() is
finished but before make_request_fn() grabs queue lock.
Make it take *@bio instead and return boolean result indicating
whether the request is throttled or not.
This patch doesn't cause any visible functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Reorganize queue draining related code in preparation of queue exit
changes.
* Factor out actual draining from elv_quiesce_start() to
blk_drain_queue().
* Make elv_quiesce_start/end() responsible for their own locking.
* Replace open-coded ELVSWITCH clearing in elevator_switch() with
elv_quiesce_end().
This patch doesn't cause any visible functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_alloc_request() and freed_request() take different combinations of
REQ_* @flags, @priv and @is_sync when @flags is superset of the latter
two. Make them take @flags only. This cleans up the code a bit and
will ease updating allocation related REQ_* flags.
This patch doesn't introduce any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A kernel crash is observed when a mounted ext3/ext4 filesystem is
physically removed. The problem is that blk_cleanup_queue() frees up
some resources eg by calling elevator_exit(), which are not checked for
in normal operation. So we should rather move these calls to the
destructor function blk_release_queue() as at that point all remaining
references are gone. However, in doing so we have to ensure that any
externally supplied queue_lock is disconnected as the driver might free
up the lock after the call of blk_cleanup_queue(),
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Thus spake Andrew Morton:
"And I have the usual maintainability whine. If someone comes up to
vmscan.c and sees it calling blk_start_plug(), how are they supposed to
work out why that call is there? They go look at the blk_start_plug()
definition and it is undocumented. I think we can do better than this?"
Adapted from the LWN article - http://lwn.net/Articles/438256/ by Jens
Axboe and from an earlier attempt by Shaohua Li to document blk-plug.
[akpm@linux-foundation.org: grammatical and spelling tweaks]
Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de>
Cc: Shaohua Li <shaohua.li@intel.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Andrew Morton <akpm@google.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move all the checks performed on a bio into a new helper, and call it as
soon as bio is submitted even if it is a re-submission from ->make_request.
We explicitly mark the new helper as beeing non-inlined as the stack
usage for printing the block device name in the failure case is quite
high and this a patch where we have to be extremely conservative about
stack usage.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There is very little benefit in allowing to let a ->make_request
instance update the bios device and sector and loop around it in
__generic_make_request when we can archive the same through calling
generic_make_request from the driver and letting the loop in
generic_make_request handle it.
Note that various drivers got the return value from ->make_request and
returned non-zero values for errors.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Avoid the hacks need for request based device mappers currently by simply
exporting the symbol instead of trying to get it through the back door.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Cleaning up the code a little bit. attempt_plug_merge() traverses the plug
list anyway, we can do the request counting there, so stack size is reduced
a little bit.
The motivation here is I suspect if we should count the requests for each
queue (task could handle multiple disks in the meantime), but my test doesn't
show it's worthy doing. If somebody proves we should do it, below change
will make that more easier.
Signed-off-by: Shaohua Li <shli@kernel.org>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Do blk_flush_plug_list() first and then add new request aDo blk_flush_plug_list() first and then add new request aDo blk_flush_plug_list() first and then add new request at the tail. New
request can't be merged to existing requests, but later new requests might
be merged with this new one. If blk_flush_plug_list() is done later, the
merge doesn't happen.
Believe it or not, this fixes a 10% regression running sysbench workload.
Signed-off-by: Shaohua Li <shli@kernel.org>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
* 'for-linus' of git://git.kernel.dk/linux-block: (23 commits)
Revert "cfq: Remove special treatment for metadata rqs."
block: fix flush machinery for stacking drivers with differring flush flags
block: improve rq_affinity placement
blktrace: add FLUSH/FUA support
Move some REQ flags to the common bio/request area
allow blk_flush_policy to return REQ_FSEQ_DATA independent of *FLUSH
xen/blkback: Make description more obvious.
cfq-iosched: Add documentation about idling
block: Make rq_affinity = 1 work as expected
block: swim3: fix unterminated of_device_id table
block/genhd.c: remove useless cast in diskstats_show()
drivers/cdrom/cdrom.c: relax check on dvd manufacturer value
drivers/block/drbd/drbd_nl.c: use bitmap_parse instead of __bitmap_parse
bsg-lib: add module.h include
cfq-iosched: Reduce linked group count upon group destruction
blk-throttle: correctly determine sync bio
loop: fix deadlock when sysfs and LOOP_CLR_FD race against each other
loop: add BLK_DEV_LOOP_MIN_COUNT=%i to allow distros 0 pre-allocated loop devices
loop: add management interface for on-demand device allocation
loop: replace linked list of allocated devices with an idr index
...
Commit ae1b153962, block: reimplement
FLUSH/FUA to support merge, introduced a performance regression when
running any sort of fsyncing workload using dm-multipath and certain
storage (in our case, an HP EVA). The test I ran was fs_mark, and it
dropped from ~800 files/sec on ext4 to ~100 files/sec. It turns out
that dm-multipath always advertised flush+fua support, and passed
commands on down the stack, where those flags used to get stripped off.
The above commit changed that behavior:
static inline struct request *__elv_next_request(struct request_queue *q)
{
struct request *rq;
while (1) {
- while (!list_empty(&q->queue_head)) {
+ if (!list_empty(&q->queue_head)) {
rq = list_entry_rq(q->queue_head.next);
- if (!(rq->cmd_flags & (REQ_FLUSH | REQ_FUA)) ||
- (rq->cmd_flags & REQ_FLUSH_SEQ))
- return rq;
- rq = blk_do_flush(q, rq);
- if (rq)
- return rq;
+ return rq;
}
Note that previously, a command would come in here, have
REQ_FLUSH|REQ_FUA set, and then get handed off to blk_do_flush:
struct request *blk_do_flush(struct request_queue *q, struct request *rq)
{
unsigned int fflags = q->flush_flags; /* may change, cache it */
bool has_flush = fflags & REQ_FLUSH, has_fua = fflags & REQ_FUA;
bool do_preflush = has_flush && (rq->cmd_flags & REQ_FLUSH);
bool do_postflush = has_flush && !has_fua && (rq->cmd_flags &
REQ_FUA);
unsigned skip = 0;
...
if (blk_rq_sectors(rq) && !do_preflush && !do_postflush) {
rq->cmd_flags &= ~REQ_FLUSH;
if (!has_fua)
rq->cmd_flags &= ~REQ_FUA;
return rq;
}
So, the flush machinery was bypassed in such cases (q->flush_flags == 0
&& rq->cmd_flags & (REQ_FLUSH|REQ_FUA)).
Now, however, we don't get into the flush machinery at all. Instead,
__elv_next_request just hands a request with flush and fua bits set to
the scsi_request_fn, even if the underlying request_queue does not
support flush or fua.
The agreed upon approach is to fix the flush machinery to allow
stacking. While this isn't used in practice (since there is only one
request-based dm target, and that target will now reflect the flush
flags of the underlying device), it does future-proof the solution, and
make it function as designed.
In order to make this work, I had to add a field to the struct request,
inside the flush structure (to store the original req->end_io). Shaohua
had suggested overloading the union with rb_node and completion_data,
but the completion data is used by device mapper and can also be used by
other drivers. So, I didn't see a way around the additional field.
I tested this patch on an HP EVA with both ext4 and xfs, and it recovers
the lost performance. Comments and other testers, as always, are
appreciated.
Cheers,
Jeff
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
init_fault_attr_dentries() is used to export fault_attr via debugfs.
But it can only export it in debugfs root directory.
Per Forlin is working on mmc_fail_request which adds support to inject
data errors after a completed host transfer in MMC subsystem.
The fault_attr for mmc_fail_request should be defined per mmc host and
export it in debugfs directory per mmc host like
/sys/kernel/debug/mmc0/mmc_fail_request.
init_fault_attr_dentries() doesn't help for mmc_fail_request. So this
introduces fault_create_debugfs_attr() which is able to create a
directory in the arbitrary directory and replace
init_fault_attr_dentries().
[akpm@linux-foundation.org: extraneous semicolon, per Randy]
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Tested-by: Per Forlin <per.forlin@linaro.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This changes should_fail_request() to more usable wrapper function of
should_fail(). It can avoid putting #ifdef CONFIG_FAIL_MAKE_REQUEST in
the middle of a function.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
After commit 5757a6d7 introduced an unsafe calling of
smp_processor_id(), with preempt debuggin turned on we spew a lot of:
BUG: using smp_processor_id() in preemptible [00000000] code: kjournald/514
caller is __make_request+0x1b8/0x308
[<c0019f44>] (unwind_backtrace+0x0/0xe8) from [<c024b4cc>] (debug_smp_processor_id+0xbc/0xf0)
[<c024b4cc>] (debug_smp_processor_id+0xbc/0xf0) from [<c0223d14>] (__make_request+0x1b8/0x308)
[<c0223d14>] (__make_request+0x1b8/0x308) from [<c02215ac>] (generic_make_request+0x4dc/0x558)
[<c02215ac>] (generic_make_request+0x4dc/0x558) from [<c022173c>] (submit_bio+0x114/0x138)
[<c022173c>] (submit_bio+0x114/0x138) from [<c011f504>] (submit_bh+0x148/0x16c)
[<c011f504>] (submit_bh+0x148/0x16c) from [<c0121ed8>] (__sync_dirty_buffer+0x88/0xd8)
[<c0121ed8>] (__sync_dirty_buffer+0x88/0xd8) from [<c01aff78>] (journal_commit_transaction+0x1198/0x1688)
[<c01aff78>] (journal_commit_transaction+0x1198/0x1688) from [<c01b4034>] (kjournald+0xb4/0x224)
[<c01b4034>] (kjournald+0xb4/0x224) from [<c0069ea0>] (kthread+0x8c/0x94)
[<c0069ea0>] (kthread+0x8c/0x94) from [<c00137f8>] (kernel_thread_exit+0x0/0x8)
Fix this by just using raw_smp_processor_id(), it's just a hint
after all. There's no pinning of the CPU or accessing per-cpu
structures involved.
Reported-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
* 'for-3.1/core' of git://git.kernel.dk/linux-block: (24 commits)
block: strict rq_affinity
backing-dev: use synchronize_rcu_expedited instead of synchronize_rcu
block: fix patch import error in max_discard_sectors check
block: reorder request_queue to remove 64 bit alignment padding
CFQ: add think time check for group
CFQ: add think time check for service tree
CFQ: move think time check variables to a separate struct
fixlet: Remove fs_excl from struct task.
cfq: Remove special treatment for metadata rqs.
block: document blk_plug list access
block: avoid building too big plug list
compat_ioctl: fix make headers_check regression
block: eliminate potential for infinite loop in blkdev_issue_discard
compat_ioctl: fix warning caused by qemu
block: flush MEDIA_CHANGE from drivers on close(2)
blk-throttle: Make total_nr_queued unsigned
block: Add __attribute__((format(printf...) and fix fallout
fs/partitions/check.c: make local symbols static
block:remove some spare spaces in genhd.c
block:fix the comment error in blkdev.h
...
Some systems benefit from completions always being steered to the strict
requester cpu rather than the looser "per-socket" steering that
blk_cpu_to_group() attempts by default. This is because the first
CPU in the group mask ends up being completely overloaded with work,
while the others (including the original submitter) has power left
to spare.
Allow the strict mode to be set by writing '2' to the sysfs control
file. This is identical to the scheme used for the nomerges file,
where '2' is a more aggressive setting than just being turned on.
echo 2 > /sys/block/<bdev>/queue/rq_affinity
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Roland Dreier <roland@purestorage.com>
Tested-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
USB surprise removal of sr is triggering an oops in
scsi_dispatch_command(). What seems to be happening is that USB is
hanging on to a queue reference until the last close of the upper
device, so the crash is caused by surprise remove of a mounted CD
followed by attempted unmount.
The problem is that USB doesn't issue its final commands as part of
the SCSI teardown path, but on last close when the block queue is long
gone. The long term fix is probably to make sr do the teardown in the
same way as sd (so remove all the lower bits on ejection, but keep the
upper disk alive until last close of user space). However, the
current oops can be simply fixed by not allowing any commands to be
sent to a dead queue.
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
When I test fio script with big I/O depth, I found the total throughput drops
compared to some relative small I/O depth. The reason is the thread accumulates
big requests in its plug list and causes some delays (surely this depends
on CPU speed).
I thought we'd better have a threshold for requests. When a threshold reaches,
this means there is no request merge and queue lock contention isn't severe
when pushing per-task requests to queue, so the main advantages of blk plug
don't exist. We can force a plug list flush in this case.
With this, my test throughput actually increases and almost equals to small
I/O depth. Another side effect is irq off time decreases in blk_flush_plug_list()
for big I/O depth.
The BLK_MAX_REQUEST_COUNT is choosen arbitarily, but 16 is efficiently to
reduce lock contention to me. But I'm open here, 32 is ok in my test too.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Commit 73c1010119 ("block: initial patch for on-stack per-task plugging")
removed calls to elv_bio_merged() when @bio merged with @req. Re-add them.
This in turn will update merged stats in associated group. That
should be safe as long as request has got reference to the blkio_group.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Cc: Divyesh Shah <dpshah@google.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
We don't need them anymore, so kill:
- REQ_ON_PLUG checks in various places
- !rq_mergeable() check in plug merging
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Currently, all the cfq_group or throtl_group allocations happen while
we are holding ->queue_lock and sleeping is not allowed.
Soon, we will move to per cpu stats and also need to allocate the
per group stats. As one can not call alloc_percpu() from atomic
context as it can sleep, we need to drop ->queue_lock, allocate the
group, retake the lock and continue processing.
In throttling code, I check the queue DEAD flag again to make sure
that driver did not call blk_cleanup_queue() in the mean time.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Let's check a scenario:
1. blk_delay_queue(q, SCSI_QUEUE_DELAY);
2. blk_run_queue_async();
the second one will became a noop, because q->delay_work already has
WORK_STRUCT_PENDING_BIT set, so the delayed work will still run after
SCSI_QUEUE_DELAY. But blk_run_queue_async actually hopes the delayed
work runs immediately.
Fix this by doing a cancel on potentially pending delayed work
before queuing an immediate run of the workqueue.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
We don't pass in a 'force_kblockd' anymore, get rid of the
stsale comment.
Reported-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
We are currently using this flag to check whether it's safe
to call into ->request_fn(). If it is set, we punt to kblockd.
But we get a lot of false positives and excessive punts to
kblockd, which hurts performance.
The only real abuser of this infrastructure is SCSI. So export
the async queue run and convert SCSI over to use that. There's
room for improvement in that SCSI need not always use the async
call, but this fixes our performance issue and they can fix that
up in due time.
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
With all drivers and file systems converted, we only have
in-core use of this function. So remove the export.
Reporteed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Instead of overloading __blk_run_queue to force an offload to kblockd
add a new blk_run_queue_async helper to do it explicitly. I've kept
the blk_queue_stopped check for now, but I suspect it's not needed
as the check we do when the workqueue items runs should be enough.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Ezequiel Garcia
Jianpeng Ma
Linus Torvalds
Tejun Heo
Martin K. Petersen
Jaehoon Chung
Kent Overstreet
Yi Zou
NeilBrown
Shaohua Li
Asias He
Jens Axboe
Shaohua Li
Dan Carpenter
Shaohua Li
Mike Snitzer
Jeff Moyer
Tao Ma
Hannes Reinecke
Suresh Jayaraman
Christoph Hellwig
Jens Axboe
Akinobu Mita
Dan Williams
James Bottomley
Luca Tettamanti
Vivek Goyal