Arrays personalities such as 'raid0' and 'linear' have no redundancy,
and so marking them as 'clean' or 'dirty' is not meaningful.
So always allow write requests without requiring a superblock update.
Such arrays types are detected by ->sync_request being NULL. If it is
not possible to send a sync request we don't need a 'dirty' flag because
all a dirty flag does is trigger some sync_requests.
Signed-off-by: Neil Brown <neilb@suse.de>
There is a possible race in md_probe. If two threads call md_probe
for the same device, then one could exit (having checked that
->gendisk exists) before the other has called kobject_init_and_add,
thus returning an incomplete kobj which will cause problems when
we try to add children to it.
So extend the range of protection of disks_mutex slightly to
avoid this possibility.
Signed-off-by: Neil Brown <neilb@suse.de>
This makes it possible to just resync a small part of an array.
e.g. if a drive reports that it has questionable sectors,
a 'repair' of just the region covering those sectors will
cause them to be read and, if there is an error, re-written
with correct data.
Signed-off-by: Neil Brown <neilb@suse.de>
When an array is degraded, bits in the write-intent bitmap are not
cleared, so that if the missing device is re-added, it can be synced
by only updated those parts of the device that have changed since
it was removed.
The enable this a 'events_cleared' value is stored. It is the event
counter for the array the last time that any bits were cleared.
Sometimes - if a device disappears from an array while it is 'clean' -
the events_cleared value gets updated incorrectly (there are subtle
ordering issues between updateing events in the main metadata and the
bitmap metadata) resulting in the missing device appearing to require
a full resync when it is re-added.
With this patch, we update events_cleared precisely when we are about
to clear a bit in the bitmap. We record events_cleared when we clear
the bit internally, and copy that to the superblock which is written
out before the bit on storage. This makes it more "obviously correct".
We also need to update events_cleared when the event_count is going
backwards (as happens on a dirty->clean transition of a non-degraded
array).
Thanks to Mike Snitzer for identifying this problem and testing early
"fixes".
Cc: "Mike Snitzer" <snitzer@gmail.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Turn calls to bi->bi_end_io() into bio_endio(). Apparently bio_endio does
exactly the same error processing as is hardcoded at these places.
bio_endio() avoids recursion (or will soon), so it should be used.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Neil Brown <neilb@suse.de>
From: "Nikanth Karthikesan" <knikanth@novell.com>
Correct disk numbering problem check.
Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>
Signed-off-by: Neil Brown <neilb@suse.de>
md_probe can fail (e.g. alloc_disk could fail) without
returning an error (as it alway returns NULL).
So when we call mddev_find immediately afterwards, we need
to check that md_probe actually succeeded. This means checking
that mdev->gendisk is non-NULL.
cc: <stable@kernel.org>
Cc: Dave Jones <davej@redhat.com>
Signed-off-by: Neil Brown <neilb@suse.de>
We shouldn't acknowledge that a stripe has been expanded (When
reshaping a raid5 by adding a device) until the moved data has
actually been written out. However we are currently
acknowledging (by calling md_done_sync) when the POST_XOR
is complete and before the write.
So track in s.locked whether there are pending writes, and don't
call md_done_sync yet if there are.
Note: we all set R5_LOCKED on devices which are are about to
read from. This probably isn't technically necessary, but is
usually done when writing a block, and justifies the use of
s.locked here.
This bug can lead to a crash if an array is stopped while an reshape
is in progress.
Cc: <stable@kernel.org>
Signed-off-by: Neil Brown <neilb@suse.de>
If, while assembling an array, we find a device which is not fully
in-sync with the array, it is important to set the "fullsync" flags.
This is an exact analog to the setting of this flag in hot_add_disk
methods.
Currently, only v1.x metadata supports having devices in an array
which are not fully in-sync (it keep track of how in sync they are).
The 'fullsync' flag only makes a difference when a write-intent bitmap
is being used. In this case it tells recovery to ignore the bitmap
and recovery all blocks.
This fix is already in place for raid1, but not raid5/6 or raid10.
So without this fix, a raid1 ir raid4/5/6 array with version 1.x
metadata and a write intent bitmaps, that is stopped in the middle
of a recovery, will appear to complete the recovery instantly
after it is reassembled, but the recovery will not be correct.
If you might have an array like that, issueing
echo repair > /sys/block/mdXX/md/sync_action
will make sure recovery completes properly.
Cc: <stable@kernel.org>
Signed-off-by: Neil Brown <neilb@suse.de>
If a block is computed (rather than read) then a check/repair operation
may be lead to believe that the data on disk is correct, when infact it
isn't. So only compute blocks for failed devices.
This issue has been around since at least 2.6.12, but has become harder to
hit in recent kernels since most reads bypass the cache.
echo repair > /sys/block/mdN/md/sync_action will set the parity blocks to the
correct state.
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If an array was created with --assume-clean we will oops when trying to
set ->resync_max.
Fix this by initializing ->recovery_wait in mddev_find.
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
During the initial array synchronization process there is a window between
when a prexor operation is scheduled to a specific stripe and when it
completes for a sync_request to be scheduled to the same stripe. When
this happens the prexor completes and the stripe is unconditionally marked
"insync", effectively canceling the sync_request for the stripe. Prior to
2.6.23 this was not a problem because the prexor operation was done under
sh->lock. The effect in older kernels being that the prexor would still
erroneously mark the stripe "insync", but sync_request would be held off
and re-mark the stripe as "!in_sync".
Change the write completion logic to not mark the stripe "in_sync" if a
prexor was performed. The effect of the change is to sometimes not set
STRIPE_INSYNC. The worst this can do is cause the resync to stall waiting
for STRIPE_INSYNC to be set. If this were happening, then STRIPE_SYNCING
would be set and handle_issuing_new_read_requests would cause all
available blocks to eventually be read, at which point prexor would never
be used on that stripe any more and STRIPE_INSYNC would eventually be set.
echo repair > /sys/block/mdN/md/sync_action will correct arrays that may
have lost this race.
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When we get any IO error during a recovery (rebuilding a spare), we abort
the recovery and restart it.
For RAID6 (and multi-drive RAID1) it may not be best to restart at the
beginning: when multiple failures can be tolerated, the recovery may be
able to continue and re-doing all that has already been done doesn't make
sense.
We already have the infrastructure to record where a recovery is up to
and restart from there, but it is not being used properly.
This is because:
- We sometimes abort with MD_RECOVERY_ERR rather than just MD_RECOVERY_INTR,
which causes the recovery not be be checkpointed.
- We remove spares and then re-added them which loses important state
information.
The distinction between MD_RECOVERY_ERR and MD_RECOVERY_INTR really isn't
needed. If there is an error, the relevant drive will be marked as
Faulty, and that is enough to ensure correct handling of the error. So we
first remove MD_RECOVERY_ERR, changing some of the uses of it to
MD_RECOVERY_INTR.
Then we cause the attempt to remove a non-faulty device from an array to
fail (unless recovery is impossible as the array is too degraded). Then
when remove_and_add_spares attempts to remove the devices on which
recovery can continue, it will fail, they will remain in place, and
recovery will continue on them as desired.
Issue: If we are halfway through rebuilding a spare and another drive
fails, and a new spare is immediately available, do we want to:
1/ complete the current rebuild, then go back and rebuild the new spare or
2/ restart the rebuild from the start and rebuild both devices in
parallel.
Both options can be argued for. The code currently takes option 2 as
a/ this requires least code change
b/ this results in a minimally-degraded array in minimal time.
Cc: "Eivind Sarto" <ivan@kasenna.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In some configurations, a raid6 resync can be limited by CPU speed
(Calculating P and Q and moving data) rather than by device speed. In
these cases there is nothing to be gained byt serialising resync of arrays
that share a device, and doing the resync in parallel can provide benefit.
So add a sysfs tunable to flag an array as being allowed to resync in
parallel with other arrays that use (a different part of) the same device.
Signed-off-by: Bernd Schubert <bs@q-leap.de>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This additional notification to 'array_state' is needed to allow the
monitor application to learn about stop events via sysfs. The
sysfs_notify("sync_action") call that comes at the end of do_md_stop()
(via md_new_event) is insufficient since the 'sync_action' attribute has
been removed by this point.
(Seems like a sysfs-notify-on-removal patch is a better fix. Currently
removal updates the event count but does not wake up waiters)
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When an array enters write pending, 'array_state' changes, so we must be
sure to sysfs_notify.
Also, when waiting for user-space to acknowledge 'write-pending' by
marking the metadata as dirty, we don't want to wait for MD_CHANGE_DEVS to
be cleared as that might not happen. So explicity test for the bits that
we are really interested in.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When performing a "recovery" or "check" pass on a RAID1 array, we read
from each device and possible, if there is a difference or a read error,
write back to some devices.
We use the same 'bio' for both read and write, resetting various fields
between the two operations.
We forgot to reset bv_offset and bv_len however. These are often left
unchanged, but in the case where there is an IO error one or two sectors
into a page, they are changed.
This results in correctable errors not being corrected properly. It does
not result in any data corruption.
Cc: "Fairbanks, David" <David.Fairbanks@stratus.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Last night we had scsi problems and a hardware raid unit was offlined
during heavy i/o. While this happened we got for about 3 minutes a huge
number messages like these
Apr 12 03:36:07 pfs1n14 kernel: [197510.696595] raid5:md7: read error not correctable (sector 2993096568 on sdj2).
I guess the high error rate is responsible for not scheduling other events
- during this time the system was not pingable and in the end also other
devices run into scsi command timeouts causing problems on these unrelated
devices as well.
Signed-off-by: Bernd Schubert <bernd-schubert@gmx.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Kill the trivial and rather pointless file_path wrapper around d_path.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It is possible to add a write-intent bitmap to an active array, or remove
the bitmap that is there.
When we do with the 'quiesce' the array, which causes make_request to
block in "wait_barrier()".
However we are sampling the value of "mddev->bitmap" before the
wait_barrier call, and using it afterwards. This can result in using a
bitmap structure that has been freed.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As setting and clearing queue flags now requires that we hold a spinlock
on the queue, and as blk_queue_stack_limits is called without that lock,
get the lock inside blk_queue_stack_limits.
For blk_queue_stack_limits to be able to find the right lock, each md
personality needs to set q->queue_lock to point to the appropriate lock.
Those personalities which didn't previously use a spin_lock, us
q->__queue_lock. So always initialise that lock when allocated.
With this in place, setting/clearing of the QUEUE_FLAG_PLUGGED bit will no
longer cause warnings as it will be clear that the proper lock is held.
Thanks to Dan Williams for review and fixing the silly bugs.
Signed-off-by: NeilBrown <neilb@suse.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Alistair John Strachan <alistair@devzero.co.uk>
Cc: Nick Piggin <npiggin@suse.de>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Jacek Luczak <difrost.kernel@gmail.com>
Cc: Prakash Punnoor <prakash@punnoor.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
commit bd2ab67030 "md: close a livelock window
in handle_parity_checks5" introduced a bug in handling 'repair' operations.
After a repair operation completes we clear the state bits tracking this
operation. However, they are cleared too early and this results in the code
deciding to re-run the parity check operation. Since we have done the repair
in memory the second check does not find a mismatch and thus does not do a
writeback.
Test results:
$ echo repair > /sys/block/md0/md/sync_action
$ cat /sys/block/md0/md/mismatch_cnt
51072
$ echo repair > /sys/block/md0/md/sync_action
$ cat /sys/block/md0/md/mismatch_cnt
0
(also fix incorrect indentation)
Cc: <stable@kernel.org>
Tested-by: George Spelvin <linux@horizon.com>
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/md/raid10.c:889:17: warning: Using plain integer as NULL pointer
drivers/media/video/cx18/cx18-driver.c:616:12: warning: Using plain integer as NULL pointer
sound/oss/kahlua.c:70:12: warning: Using plain integer as NULL pointer
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Allows a userspace metadata handler to take action upon detecting a device
failure.
Based on an original patch by Neil Brown.
Changes:
-added blocked_wait waitqueue to rdev
-don't qualify Blocked with Faulty always let userspace block writes
-added md_wait_for_blocked_rdev to wait for the block device to be clear, if
userspace misses the notification another one is sent every 5 seconds
-set MD_RECOVERY_NEEDED after clearing "blocked"
-kill DoBlock flag, just test mddev->external
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Found when trying to reassemble an active externally managed array. Without
this check we hit the more noisy "sysfs duplicate" warning in the later call
to kobject_add.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When setting an array to 'readonly' or to 'active' via sysfs, we must make the
appropriate set_disk_ro call too.
Also when switching to "read_auto" (which is like readonly, but blocks on the
first write so that metadata can be marked 'dirty') we need to be more careful
about what state we are changing from.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
'safemode' relates to marking an array as 'clean' if there has been no write
traffic for a while (a couple of seconds), to reduce the chance of the array
being found dirty on reboot.
->safemode is set to '1' when there have been no write for a while, and it
gets set to '0' when the superblock is updates with the 'clean' flag set.
This requires a few fixes for 'external' metadata:
- When an array is set to 'clean' via sysfs, 'safemode' must be cleared.
- when we write to an array that has 'safemode' set (there must have been
some delay in updating the metadata), we need to clear safemode.
- Don't try to update external metadata in md_check_recovery for safemode
transitions - it won't work.
Also, don't try to support "immediate safe mode" (safemode==2) for external
metadata, it cannot really work (the safemode timeout can be set very low if
this is really needed).
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
I keep finding problems where an mddev gets reused and some fields has a value
from a previous usage that confuses the new usage. So clear all fields that
could possible need clearing when calling do_md_stop.
Also initialise the 'level' of a new array to LEVEL_NONE (which isn't 0).
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All the metadata update processing for external metadata is on in user-space
or through the sysfs interfaces, so make "md_update_sb" a no-op in that case.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
rdev->mddev is no longer valid upon return from entry->store() when the
'remove' command is given.
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
block: Skip I/O merges when disabled
block: add large command support
block: replace sizeof(rq->cmd) with BLK_MAX_CDB
ide: use blk_rq_init() to initialize the request
block: use blk_rq_init() to initialize the request
block: rename and export rq_init()
block: no need to initialize rq->cmd with blk_get_request
block: no need to initialize rq->cmd in prepare_flush_fn hook
block/blk-barrier.c:blk_ordered_cur_seq() mustn't be inline
block/elevator.c:elv_rq_merge_ok() mustn't be inline
block: make queue flags non-atomic
block: add dma alignment and padding support to blk_rq_map_kern
unexport blk_max_pfn
ps3disk: Remove superfluous cast
block: make rq_init() do a full memset()
relay: fix splice problem
Use proc_create()/proc_create_data() to make sure that ->proc_fops and ->data
be setup before gluing PDE to main tree.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Peter Osterlund <petero2@telia.com>
Cc: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Cc: Neil Brown <neilb@suse.de>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Cc: Bjorn Helgaas <bjorn.helgaas@hp.com>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
blk_get_request initializes rq->cmd (rq_init does) so the users don't
need to do that.
The purpose of this patch is to remove sizeof(rq->cmd) and &rq->cmd,
as a preparation for large command support, which changes rq->cmd from
the static array to a pointer. sizeof(rq->cmd) will not make sense and
&rq->cmd won't work.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Alasdair G Kergon <agk@redhat.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
We can save some atomic ops in the IO path, if we clearly define
the rules of how to modify the queue flags.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
The functions time_before, time_before_eq, time_after, and time_after_eq
are more robust for comparing jiffies against other values.
A simplified version of the semantic patch making this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@ change_compare_np @
expression E;
@@
(
- jiffies <= E
+ time_before_eq(jiffies,E)
|
- jiffies >= E
+ time_after_eq(jiffies,E)
|
- jiffies < E
+ time_before(jiffies,E)
|
- jiffies > E
+ time_after(jiffies,E)
)
@ include depends on change_compare_np @
@@
#include <linux/jiffies.h>
@ no_include depends on !include && change_compare_np @
@@
#include <linux/...>
+ #include <linux/jiffies.h>
// </smpl>
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Julia Lawall <julia@diku.dk>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
MD drivers use one printk() call to print 2 log messages and the second line
may be prefixed by a TAB character. It may also output a trailing space
before newline. klogd (I think) turns the TAB character into the 2 characters
'^I' when logging to a file. This looks ugly.
Instead of a leading TAB to indicate continuation, prefix both output lines
with 'raid:' or similar. Also remove any trailing space in the vicinity of
the affected code and consistently end the sentences with a period.
Signed-off-by: Nick Andrew <nick@nick-andrew.net>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
strict_strtoul handles the open-coded sanity checks in
raid5_store_stripe_cache_size and raid5_store_preread_threshold
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Improve write performance by preventing the delayed_list from dumping all its
stripes onto the handle_list in one shot. Delayed stripes are now further
delayed by being held on the 'hold_list'. The 'hold_list' is bypassed when:
* a STRIPE_IO_STARTED stripe is found at the head of 'handle_list'
* 'handle_list' is empty and i/o is being done to satisfy full stripe-width
write requests
* 'bypass_count' is less than 'bypass_threshold'. By default the threshold
is 1, i.e. every other stripe handled is a preread stripe provided the
top two conditions are false.
Benchmark data:
System: 2x Xeon 5150, 4x SATA, mem=1GB
Baseline: 2.6.24-rc7
Configuration: mdadm --create /dev/md0 /dev/sd[b-e] -n 4 -l 5 --assume-clean
Test1: dd if=/dev/zero of=/dev/md0 bs=1024k count=2048
* patched: +33% (stripe_cache_size = 256), +25% (stripe_cache_size = 512)
Test2: tiobench --size 2048 --numruns 5 --block 4096 --block 131072 (XFS)
* patched: +13%
* patched + preread_bypass_threshold = 0: +37%
Changes since v1:
* reduce bypass_threshold from (chunk_size / sectors_per_chunk) to (1) and
make it configurable. This defaults to fairness and modest performance
gains out of the box.
Changes since v2:
* [neilb@suse.de]: kill STRIPE_PRIO_HI and preread_needed as they are not
necessary, the important change was clearing STRIPE_DELAYED in
add_stripe_bio and this has been moved out to make_request for the hang
fix.
* [neilb@suse.de]: simplify get_priority_stripe
* [dan.j.williams@intel.com]: reset the bypass_count when ->hold_list is
sampled empty (+11%)
* [dan.j.williams@intel.com]: decrement the bypass_count at the detection
of stripes being naturally promoted off of hold_list +2%. Note, resetting
bypass_count instead of decrementing on these events yields +4% but that is
probably too aggressive.
Changes since v3:
* cosmetic fixups
Tested-by: James W. Laferriere <babydr@baby-dragons.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__FUNCTION__ is gcc-specific, use __func__
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/md/md.c:734:16: warning: Using plain integer as NULL pointer
drivers/md/md.c:1115:16: warning: Using plain integer as NULL pointer
Add some braces to match the else-block as well.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The small patch below:
- Removes the unused md argument from both specific_minor() and next_free_minor()
- Folds kmalloc + memset(0) into a single kzalloc call in alloc_dev()
This has been compile tested on x86.
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@gmail.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
dm_create_error_table() was added in kernel 2.6.18 and never used...
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
void returning functions returned the return value of another void
returning function...
Spotted by sparse.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Remove an avoidable 3ms delay on some dm-raid1 and kcopyd I/O.
It is specified that any submitted bio without BIO_RW_SYNC flag may plug the
queue (i.e. block the requests from being dispatched to the physical device).
The queue is unplugged when the caller calls blk_unplug() function. Usually, the
sequence is that someone calls submit_bh to submit IO on a buffer. The IO plugs
the queue and waits (to be possibly joined with other adjacent bios). Then, when
the caller calls wait_on_buffer(), it unplugs the queue and submits the IOs to
the disk.
This was happenning:
When doing O_SYNC writes, function fsync_buffers_list() submits a list of
bios to dm_raid1, the bios are added to dm_raid1 write queue and kmirrord is
woken up.
fsync_buffers_list() calls wait_on_buffer(). That unplugs the queue, but
there are no bios on the device queue as they are still in the dm_raid1 queue.
wait_on_buffer() starts waiting until the IO is finished.
kmirrord is scheduled, kmirrord takes bios and submits them to the devices.
The submitted bio plugs the harddisk queue but there is no one to unplug it.
(The process that called wait_on_buffer() is already sleeping.)
So there is a 3ms timeout, after which the queues on the harddisks are
unplugged and requests are processed.
This 3ms timeout meant that in certain workloads (e.g. O_SYNC, 8kb writes),
dm-raid1 is 10 times slower than md raid1.
Every time we submit something asynchronously via dm_io, we must unplug the
queue actually to send the request to the device.
This patch adds an unplug call to kmirrord - while processing requests, it keeps
the queue plugged (so that adjacent bios can be merged); when it finishes
processing all the bios, it unplugs the queue to submit the bios.
It also fixes kcopyd which has the same potential problem. All kcopyd requests
are submitted with BIO_RW_SYNC.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
This patch replaces the schedule() in the main kmirrord thread with a timer.
The schedule() could introduce an unwanted delay when work is ready to be
processed.
The code instead calls wake() when there's work to be done immediately, and
delayed_wake() after a failure to give a short delay before retrying.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
