New wait_event{_interruptible}_lock_irq{_cmd} macros added. This commit
moves the private wait_event_lock_irq() macro from MD to regular wait
includes, introduces new macro wait_event_lock_irq_cmd() instead of using
the old method with omitting cmd parameter which is ugly and makes a use
of new macros in the MD. It also introduces the _interruptible_ variant.
The use of new interface is when one have a special lock to protect data
structures used in the condition, or one also needs to invoke "cmd"
before putting it to sleep.
All new macros are expected to be called with the lock taken. The lock
is released before sleep and is reacquired afterwards. We will leave the
macro with the lock held.
Note to DM: IMO this should also fix theoretical race on waitqueue while
using simultaneously wait_event_lock_irq() and wait_event() because of
lack of locking around current state setting and wait queue removal.
Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Cc: Neil Brown <neilb@suse.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull md updates from NeilBrown:
- "discard" support, some dm-raid improvements and other assorted bits
and pieces.
* tag 'md-3.7' of git://neil.brown.name/md: (29 commits)
md: refine reporting of resync/reshape delays.
md/raid5: be careful not to resize_stripes too big.
md: make sure manual changes to recovery checkpoint are saved.
md/raid10: use correct limit variable
md: writing to sync_action should clear the read-auto state.
Subject: [PATCH] md:change resync_mismatches to atomic64_t to avoid races
md/raid5: make sure to_read and to_write never go negative.
md: When RAID5 is dirty, force reconstruct-write instead of read-modify-write.
md/raid5: protect debug message against NULL derefernce.
md/raid5: add some missing locking in handle_failed_stripe.
MD: raid5 avoid unnecessary zero page for trim
MD: raid5 trim support
md/bitmap:Don't use IS_ERR to judge alloc_page().
md/raid1: Don't release reference to device while handling read error.
raid: replace list_for_each_continue_rcu with new interface
add further __init annotations to crypto/xor.c
DM RAID: Fix for "sync" directive ineffectiveness
DM RAID: Fix comparison of index and quantity for "rebuild" parameter
DM RAID: Add rebuild capability for RAID10
DM RAID: Move 'rebuild' checking code to its own function
...
When a RAID5 is reshaping, conf->raid_disks is increased
before mddev->delta_disks becomes zero.
This can result in check_reshape calling resize_stripes with a
number that is too large. This particularly happens
when md_check_recovery calls ->check_reshape().
If we use ->previous_raid_disks, we don't risk this.
Signed-off-by: NeilBrown <neilb@suse.de>
Now that multiple threads can handle stripes, it is safer to
use an atomic64_t for resync_mismatches, to avoid update races.
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
to_read and to_write are part of the result of analysing
a stripe before handling it.
Their use is to avoid some loops and tests if the values are
known to be zero. Thus it is not a problem if they are a
little bit larger than they should be.
So decrementing them in handle_failed_stripe serves little value, and
due to races it could cause some loops to be skipped incorrectly.
So remove those decrements.
Reported-by: "Jianpeng Ma" <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
The pr_debug in add_stripe_bio could race with something
changing *bip, so it is best to hold the lock until
after the pr_debug.
Reported-by: "Jianpeng Ma" <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
We really should hold the stripe_lock while accessing
'toread' else we could race with add_stripe_bio and corrupt
a list.
Reported-by: "Jianpeng Ma" <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
We want to avoid zero discarded dev page, because it's useless for discard.
But if we don't zero it, another read/write hit such page in the cache and will
get inconsistent data.
To avoid zero the page, we don't set R5_UPTODATE flag after construction is
done. In this way, discard write request is still issued and finished, but read
will not hit the page. If the stripe gets accessed soon, we need reread the
stripe, but since the chance is low, the reread isn't a big deal.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Discard for raid4/5/6 has limitation. If discard request size is
small, we do discard for one disk, but we need calculate parity and
write parity disk. To correctly calculate parity, zero_after_discard
must be guaranteed. Even it's true, we need do discard for one disk
but write another disks, which makes the parity disks wear out
fast. This doesn't make sense. So an efficient discard for raid4/5/6
should discard all data disks and parity disks, which requires the
write pattern to be (A, A+chunk_size, A+chunk_size*2...). If A's size
is smaller than chunk_size, such pattern is almost impossible in
practice. So in this patch, I only handle the case that A's size
equals to chunk_size. That is discard request should be aligned to
stripe size and its size is multiple of stripe size.
Since we can only handle request with specific alignment and size (or
part of the request fitting stripes), we can't guarantee
zero_after_discard even zero_after_discard is true in low level
drives.
The block layer doesn't send down correctly aligned requests even
correct discard alignment is set, so I must filter out.
For raid4/5/6 parity calculation, if data is 0, parity is 0. So if
zero_after_discard is true for all disks, data is consistent after
discard. Otherwise, data might be lost. Let's consider a scenario:
discard a stripe, write data to one disk and write parity disk. The
stripe could be still inconsistent till then depending on using data
from other data disks or parity disks to calculate new parity. If the
disk is broken, we can't restore it. So in this patch, we only enable
discard support if all disks have zero_after_discard.
If discard fails in one disk, we face the similar inconsistent issue
above. The patch will make discard follow the same path as normal
write request. If discard fails, a resync will be scheduled to make
the data consistent. This isn't good to have extra writes, but data
consistency is important.
If a subsequent read/write request hits raid5 cache of a discarded
stripe, the discarded dev page should have zero filled, so the data is
consistent. This patch will always zero dev page for discarded request
stripe. This isn't optimal because discard request doesn't need such
payload. Next patch will avoid it.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Change the thread parameter, so the thread can carry extra info. Next patch
will use it.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
commit b17459c050
raid5: add a per-stripe lock
added a spin_lock to the 'stripe_head' struct.
Unfortunately there are two places where this struct is allocated
but the spin lock was only initialised in one of them.
So add the missing spin_lock_init.
Signed-off-by: NeilBrown <neilb@suse.de>
When a replacement device becomes active, we mark the device that it
replaces as 'faulty' so that it can subsequently get removed.
However 'calc_degraded' only pays attention to the primary device, not
the replacement, so the array appears to become degraded, which is
wrong.
So teach 'calc_degraded' to consider any replacement if a primary
device is faulty.
This is suitable for -stable as an incorrect 'degraded' value can
confuse md and could lead to data corruption.
This is only relevant for 3.3 and later.
Cc: stable@vger.kernel.org
Reported-by: Robin Hill <robin@robinhill.me.uk>
Reported-by: John Drescher <drescherjm@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
This reverts commit 895e3c5c58.
While this patch seemed like a good idea and did help some workloads,
it hurts other workloads.
Large sequential O_DIRECT writes were faster,
Small random O_DIRECT writes were slower.
Other changes (batching RAID5 writes) have improved the sequential
writes using a different mechanism, so the net result of this patch
is definitely negative. So revert it.
Reported-by: Shaohua Li <shli@kernel.org>
Tested-by: Jianpeng Ma <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Pull additional md update from NeilBrown:
"This contains a few patches that depend on plugging changes in the
block layer so needed to wait for those.
It also contains a Kconfig fix for the new RAID10 support in dm-raid."
* tag 'md-3.6' of git://neil.brown.name/md:
md/dm-raid: DM_RAID should select MD_RAID10
md/raid1: submit IO from originating thread instead of md thread.
raid5: raid5d handle stripe in batch way
raid5: make_request use batch stripe release
Let raid5d handle stripe in batch way to reduce conf->device_lock locking.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
make_request() does stripe release for every stripe and the stripe usually has
count 1, which makes previous release_stripe() optimization not work. In my
test, this release_stripe() becomes the heaviest pleace to take
conf->device_lock after previous patches applied.
Below patch makes stripe release batch. All the stripes will be released in
unplug. The STRIPE_ON_UNPLUG_LIST bit is to protect concurrent access stripe
lru.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Pull block driver changes from Jens Axboe:
- Making the plugging support for drivers a bit more sane from Neil.
This supersedes the plugging change from Shaohua as well.
- The usual round of drbd updates.
- Using a tail add instead of a head add in the request completion for
ndb, making us find the most completed request more quickly.
- A few floppy changes, getting rid of a duplicated flag and also
running the floppy init async (since it takes forever in boot terms)
from Andi.
* 'for-3.6/drivers' of git://git.kernel.dk/linux-block:
floppy: remove duplicated flag FD_RAW_NEED_DISK
blk: pass from_schedule to non-request unplug functions.
block: stack unplug
blk: centralize non-request unplug handling.
md: remove plug_cnt feature of plugging.
block/nbd: micro-optimization in nbd request completion
drbd: announce FLUSH/FUA capability to upper layers
drbd: fix max_bio_size to be unsigned
drbd: flush drbd work queue before invalidate/invalidate remote
drbd: fix potential access after free
drbd: call local-io-error handler early
drbd: do not reset rs_pending_cnt too early
drbd: reset congestion information before reporting it in /proc/drbd
drbd: report congestion if we are waiting for some userland callback
drbd: differentiate between normal and forced detach
drbd: cleanup, remove two unused global flags
floppy: Run floppy initialization asynchronous
This seemed like a good idea at the time, but after further thought I
cannot see it making a difference other than very occasionally and
testing to try to exercise the case it is most likely to help did not
show any performance difference by removing it.
So remove the counting of active plugs and allow 'pending writes' to
be activated at any time, not just when no plugs are active.
This is only relevant when there is a write-intent bitmap, and the
updating of the bitmap will likely introduce enough delay that
the single-threading of bitmap updates will be enough to collect large
numbers of updates together.
Removing this will make it easier to centralise the unplug code, and
will clear the other for other unplug enhancements which have a
measurable effect.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
'sync' writes set both REQ_SYNC and REQ_NOIDLE.
O_DIRECT writes set REQ_SYNC but not REQ_NOIDLE.
We currently assume that a REQ_SYNC request will not be followed by
more requests and so set STRIPE_PREREAD_ACTIVE to expedite the
request.
This is appropriate for sync requests, but not for O_DIRECT requests.
So make the setting of STRIPE_PREREAD_ACTIVE conditional on REQ_NOIDLE
rather than REQ_SYNC. This is consistent with the documented meaning
of REQ_NOIDLE:
__REQ_NOIDLE, /* don't anticipate more IO after this one */
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Because bios will merge at block-layer,so bios-error may caused by other
bio which be merged into to the same request.
Using this flag,it will find exactly error-sector and not do redundant
operation like re-write and re-read.
V0->V1:Using REQ_FLUSH instead REQ_NOMERGE avoid bio merging at block
layer.
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Add a per-stripe lock to protect stripe specific data. The purpose is to reduce
lock contention of conf->device_lock.
stripe ->toread, ->towrite are protected by per-stripe lock. Accessing bio
list of the stripe is always serialized by this lock, so adding bio to the
lists (add_stripe_bio()) and removing bio from the lists (like
ops_run_biofill()) not race.
If bio in ->read, ->written ... list are not shared by multiple stripes, we
don't need any lock to protect ->read, ->written, because STRIPE_ACTIVE will
protect them. If the bio are shared, there are two protections:
1. bi_phys_segments acts as a reference count
2. traverse the list uses r5_next_bio, which makes traverse never access bio
not belonging to the stripe
Let's have an example:
| stripe1 | stripe2 | stripe3 |
...bio1......|bio2|bio3|....bio4.....
stripe2 has 4 bios, when it's finished, it will decrement bi_phys_segments for
all bios, but only end_bio for bio2 and bio3. bio1->bi_next still points to
bio2, but this doesn't matter. When stripe1 is finished, it will not touch bio2
because of r5_next_bio check. Next time stripe1 will end_bio for bio1 and
stripe3 will end_bio bio4.
before add_stripe_bio() addes a bio to a stripe, we already increament the bio
bi_phys_segments, so don't worry other stripes release the bio.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Neil pointed out the bitmap write optimization in handle_stripe_clean_event()
is unnecessary, because the chance one stripe gets written twice in the mean
time is rare. We can always do a bitmap_startwrite when a write request is
added to a stripe and bitmap_endwrite after write request is done. Delete the
optimization. With it, we can delete some cases of device_lock.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Raid5 overrides bio->bi_phys_segments, accessing it is with device_lock hold,
which is unnecessary, We can make it lockless actually.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
release_stripe() is a place conf->device_lock is heavily contended. We take the
lock even stripe count isn't 1, which isn't required.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Lukas Czerner
Linus Torvalds
NeilBrown
Jianpeng Ma
Alexander Lyakas
Shaohua Li