Joe Thornber [Thu, 27 Nov 2014 12:26:46 +0000 (12:26 +0000)]
dm cache: dirty flag was mistakenly being cleared when promoting via overwrite
If the incoming bio is a WRITE and completely covers a block then we
don't bother to do any copying for a promotion operation. Once this is
done the cache block and origin block will be different, so we need to
set it to 'dirty'.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Joe Thornber [Thu, 27 Nov 2014 12:21:08 +0000 (12:21 +0000)]
dm cache: only use overwrite optimisation for promotion when in writeback mode
Overwrite causes the cache block and origin blocks to diverge, which
is only allowed in writeback mode.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Joe Thornber [Wed, 26 Nov 2014 16:07:50 +0000 (16:07 +0000)]
dm cache: discard block size must be a multiple of cache block size
Otherwise the cache blocks may span two discard blocks, which we don't
handle when doing the discard lookup.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Tue, 25 Nov 2014 13:14:57 +0000 (13:14 +0000)]
dm cache: fix a harmless race when working out if a block is discarded
It is more correct to hold the cell before checking the discard state.
These flags are only used as hints to the policy so this change will
have negligable effect.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 24 Nov 2014 14:06:22 +0000 (14:06 +0000)]
dm cache: when reloading a discard bitset allow for a different discard block size
The discard block size can change if the origin changes size or if an
old DM cache is upgraded from using a discard block size that was equal
to cache block size.
To fix this an extent of discarded blocks is established for the purpose
of translating the old discard block size to the new in-core discard
block size and set bits. The old (potentially huge) discard bitset is
left ondisk until it is re-written using the new in-core information on
the next successful DM cache shutdown.
Fixes: 7ae34e777896 ("dm cache: improve discard support")
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 24 Nov 2014 14:05:16 +0000 (14:05 +0000)]
dm cache: fix some issues with the new discard range support
Commit
7ae34e777 ("dm cache: improve discard support") needed to also:
- discontinue having DM core split the discard bios on cache block
boundaries
- calculate the cache's discard_nr_blocks relative to the determined
discard_block_size rather than using oblock_to_dblock()
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 24 Nov 2014 14:08:57 +0000 (14:08 +0000)]
dm array: if resizing the array is a noop set the new root to the old one
This could've been quite bad (to return success but not update the new
root to point at the old) but in practice the only known consumer of the
dm array code is the DM cache target. And the DM cache target passes in
the same old root to array_resize() anyway.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Eric Dumazet [Sun, 23 Nov 2014 17:34:29 +0000 (09:34 -0800)]
dm: use rcu_dereference_protected instead of rcu_dereference
rcu_dereference() should be used in sections protected by rcu_read_lock.
For writers, holding some kind of mutex or lock,
rcu_dereference_protected() is the way to go, adding explicit lockdep
bits.
In __unbind(), we are the last user of this mapped device, so can use
the constant '1' instead of a lockdep_is_held(), not consistent with
other uses of rcu_dereference_protected() which use md->suspend_lock
mutex.
Reported-by: Kirill A. Shutemov <kirill@shutemov.name>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: 33423974bfc1 ("dm: Use rcu_dereference() for accessing rcu pointer")
Cc: Pranith Kumar <bobby.prani@gmail.com>
[snitzer: allow lines longer than 80 columns, refine subject]
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Thu, 20 Nov 2014 23:07:43 +0000 (18:07 -0500)]
dm thin: fix pool_io_hints to avoid looking at max_hw_sectors
Simplify the pool_io_hints code that works to establish a max_sectors
value that is a power-of-2 factor of the thin-pool's blocksize. The
biggest associated improvement is that the DM thin-pool is no longer
concerning itself with the data device's max_hw_sectors when adjusting
max_sectors.
This fixes the relative fragility of the original "dm thin: adjust
max_sectors_kb based on thinp blocksize" commit that only became
apparent when testing was performed using a DM thin-pool ontop of a
virtio_blk device. One proposed upstream patch detailed the problems
inherent in virtio_blk: https://lkml.org/lkml/2014/11/20/611
So even though virtio_blk incorrectly set its max_hw_sectors it actually
helped make it clear that we need DM thinp to be tolerant of any future
Linux driver that incorrectly sets max_hw_sectors.
We only need to be concerned with modifying the thin-pool device's
max_sectors limit if it is smaller than the thin-pool's blocksize. In
this case the value of max_sectors does become a limiting factor when
upper layers (e.g. filesystems) construct their bios. But if the
hardware can support IOs larger than the thin-pool's blocksize the user
is encouraged to adjust the thin-pool's data device's max_sectors
accordingly -- doing so will enable the thin-pool to inherit the
established user-defined max_sectors.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Wed, 29 Oct 2014 00:58:45 +0000 (20:58 -0400)]
dm thin: suspend/resume active thin devices when reloading thin-pool
Before this change it was expected that userspace would first suspend
all active thin devices, reload/resize the thin-pool target, then resume
all active thin devices. Now the thin-pool suspend/resume will trigger
the suspend/resume of all active thins via appropriate calls to
dm_internal_suspend and dm_internal_resume.
Store the mapped_device for each thin device in struct thin_c to make
these calls possible.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Mike Snitzer [Tue, 28 Oct 2014 22:34:52 +0000 (18:34 -0400)]
dm: enhance internal suspend and resume interface
Rename dm_internal_{suspend,resume} to dm_internal_{suspend,resume}_fast
-- dm-stats will continue using these methods to avoid all the extra
suspend/resume logic that is not needed in order to quickly flush IO.
Introduce dm_internal_suspend_noflush() variant that actually calls the
mapped_device's target callbacks -- otherwise target-specific hooks are
avoided (e.g. dm-thin's thin_presuspend and thin_postsuspend). Common
code between dm_internal_{suspend_noflush,resume} and
dm_{suspend,resume} was factored out as __dm_{suspend,resume}.
Update dm_internal_{suspend_noflush,resume} to always take and release
the mapped_device's suspend_lock. Also update dm_{suspend,resume} to be
aware of potential for DM_INTERNAL_SUSPEND_FLAG to be set and respond
accordingly by interruptibly waiting for the DM_INTERNAL_SUSPEND_FLAG to
be cleared. Add lockdep annotation to dm_suspend() and dm_resume().
The existing DM_SUSPEND_FLAG remains unchanged.
DM_INTERNAL_SUSPEND_FLAG is set by dm_internal_suspend_noflush() and
cleared by dm_internal_resume().
Both DM_SUSPEND_FLAG and DM_INTERNAL_SUSPEND_FLAG may be set if a device
was already suspended when dm_internal_suspend_noflush() was called --
this can be thought of as a "nested suspend". A "nested suspend" can
occur with legacy userspace dm-thin code that might suspend all active
thin volumes before suspending the pool for resize.
But otherwise, in the normal dm-thin-pool suspend case moving forward:
the thin-pool will have DM_SUSPEND_FLAG set and all active thins from
that thin-pool will have DM_INTERNAL_SUSPEND_FLAG set.
Also add DM_INTERNAL_SUSPEND_FLAG to status report. This new
DM_INTERNAL_SUSPEND_FLAG state is being reported to assist with
debugging (e.g. 'dmsetup info' will report an internally suspended
device accordingly).
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Mike Snitzer [Fri, 7 Nov 2014 20:09:46 +0000 (15:09 -0500)]
dm thin: do not allow thin device activation while pool is suspended
Otherwise IO could be issued to the pool while it is suspended.
Care was taken to properly interlock between the thin and thin-pool
targets when accessing the pool's 'suspended' flag. The thin_ctr will
not add a new thin device to the pool's active_thins list if the pool is
susepended.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Mike Snitzer [Wed, 29 Oct 2014 00:13:31 +0000 (20:13 -0400)]
dm: add presuspend_undo hook to target_type
The DM thin-pool target now must undo the changes performed during
pool_presuspend() so introduce presuspend_undo hook in target_type.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Joe Thornber <ejt@redhat.com>
Mike Snitzer [Sun, 16 Nov 2014 19:21:47 +0000 (14:21 -0500)]
dm: return earlier from dm_blk_ioctl if target doesn't implement .ioctl
No point checking if the device is suspended if the current target
doesn't even implement .ioctl
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Fri, 7 Nov 2014 20:27:56 +0000 (15:27 -0500)]
dm thin: remove stale 'trim' message in block comment above pool_message
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mikulas Patocka [Wed, 5 Nov 2014 22:00:13 +0000 (17:00 -0500)]
dm thin: fix a race in thin_dtr
As long as struct thin_c is in the list, anyone can grab a reference of
it. Consequently, we must wait for the reference count to drop to zero
*after* we remove the structure from the list, not before.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Tue, 11 Nov 2014 11:58:32 +0000 (11:58 +0000)]
dm cache: emit a warning message if there are a lot of cache blocks
Loading and saving millions of block mappings takes time. We may as
well explain what's going on, and encourage people to use a larger
cache block size.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Thu, 6 Nov 2014 10:18:04 +0000 (10:18 +0000)]
dm cache: improve discard support
Safely allow the discard blocksize to be larger than the cache blocksize
by using the bio prison's range locking support. This also improves
discard performance considerly because larger discards are issued to the
dm-cache device. The discard blocksize was always intended to be
greater than the cache blocksize. But until now it wasn't implemented
safely.
Also, by safely restoring the ability to have discard blocksize larger
than cache blocksize we're able to significantly reduce the memory used
for the cache's discard bitset. Before, with a small discard blocksize,
the discard bitset could get quite large because its size is a function
of the discard blocksize and the origin device's size. For example,
previously, using a 32KB cache blocksize with a 40TB origin resulted in
1280MB of incore memory use for the discard bitset! Now, the discard
blocksize is scaled up accordingly to ensure the discard bitset is
capped at 2**14 bits, or 16KB.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Thu, 6 Nov 2014 14:38:01 +0000 (14:38 +0000)]
dm cache: revert "prevent corruption caused by discard_block_size > cache_block_size"
This reverts commit
d132cc6d9e92424bb9d4fd35f5bd0e55d583f4be because we
actually do want to allow the discard blocksize to be larger than the
cache blocksize. Further dm-cache discard changes will make this
possible.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Fri, 7 Nov 2014 14:47:07 +0000 (14:47 +0000)]
dm cache: revert "remove remainder of distinct discard block size"
This reverts commit
64ab346a360a4b15c28fb8531918d4a01f4eabd9 because we
actually do want to allow the discard blocksize to be larger than the
cache blocksize. Further dm-cache discard changes will make this
possible.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Wed, 17 Sep 2014 09:17:39 +0000 (10:17 +0100)]
dm bio prison: introduce support for locking ranges of blocks
Ranges will be placed in the same cell if they overlap.
Range locking is a prerequisite for more efficient multi-block discard
support in both the cache and thin-provisioning targets.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Thu, 30 Oct 2014 14:02:01 +0000 (10:02 -0400)]
dm cache policy mq: simplify ability to promote sequential IO to the cache
Before, if the user wanted sequential IO to be promoted to the cache
they'd have to set sequential_threshold to some nebulous large value.
Now, the user may easily disable sequential IO detection (and sequential
IO's implicit bypass of the cache) by setting sequential_threshold to 0.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Wed, 22 Oct 2014 13:30:58 +0000 (14:30 +0100)]
dm cache policy mq: tweak algorithm that decides when to promote a block
Rather than maintaining a separate promote_threshold variable that we
periodically update we now use the hit count of the oldest clean
block. Also add a fudge factor to discourage demoting dirty blocks.
With some tests this has a sizeable difference, because the old code
was too eager to demote blocks. For example, device-mapper-test-suite's
git_extract_cache_quick test goes from taking 190 seconds, to 142
(linear on spindle takes 250).
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Hannes Reinecke [Wed, 5 Nov 2014 13:35:50 +0000 (14:35 +0100)]
dm: do not call dm_sync_table() when creating new devices
When creating new devices dm_sync_table() calls
synchronize_rcu_expedited(), causing _all_ pending RCU pointers to be
flushed. This causes a latency overhead that is especially noticeable
when creating lots of devices.
And all of this is pointless as there are no old maps to be
disconnected, and hence no stale pointers which would need to be
cleared up.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Pranith Kumar [Tue, 28 Oct 2014 22:09:57 +0000 (15:09 -0700)]
dm: sparse: Annotate field with __rcu for checking
Annotate the map field with __rcu since this is a rcu pointer which is checked
by sparse.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Pranith Kumar [Tue, 28 Oct 2014 22:09:56 +0000 (15:09 -0700)]
dm: Use rcu_dereference() for accessing rcu pointer
The map field in 'struct mapped_device' is an rcu pointer. Use rcu_dereference()
while accessing it.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Sun, 19 Oct 2014 11:52:44 +0000 (07:52 -0400)]
dm thin: refactor requeue_io to eliminate spinlock bouncing
Also refactor some other bio_list erroring helpers.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Sun, 19 Oct 2014 12:23:09 +0000 (08:23 -0400)]
dm thin: optimize retry_bios_on_resume
Eliminate redundant should_error_unserviceable_bio check and error
loop.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Fri, 10 Oct 2014 15:42:10 +0000 (16:42 +0100)]
dm thin: sort the deferred cells
Sort the cells in logical block order before processing each cell in
process_thin_deferred_cells(). This significantly improves the ondisk
layout on rotational storage, whereby improving read performance.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Wed, 15 Oct 2014 13:46:58 +0000 (14:46 +0100)]
dm thin: direct dispatch when breaking sharing
This use of direct submission in process_shared_bio() reduces latency
for submitting bios in the shared cell by avoiding adding those bios to
the deferred list and waiting for the next iteration of the worker.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Fri, 10 Oct 2014 14:27:16 +0000 (15:27 +0100)]
dm thin: remap the bios in a cell immediately
This use of direct submission in process_prepared_mapping() reduces
latency for submitting bios in a cell by avoiding adding those bios to
the deferred list and waiting for the next iteration of the worker.
But this direct submission exposes the potential for a race between
releasing a cell and incrementing deferred set. Fix this by introducing
dm_cell_visit_release() and refactoring inc_remap_and_issue_cell()
accordingly.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Fri, 10 Oct 2014 12:43:14 +0000 (13:43 +0100)]
dm thin: defer whole cells rather than individual bios
This avoids dropping the cell, so increases the probability that other
bios will collect within the cell, rather than being passed individually
to the worker.
Also add required process_cell and process_discard_cell error handling
wrappers and set associated pool-mode function pointers accordingly.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Thu, 9 Oct 2014 23:20:21 +0000 (19:20 -0400)]
dm thin: factor out remap_and_issue_overwrite
Purely cleanup of duplicated code, no functional change.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Fri, 12 Sep 2014 10:34:01 +0000 (11:34 +0100)]
dm thin: performance improvement to discard processing
When processing a discard bio, if the block is already quiesced do the
discard immediately rather than adding the mapping to a list for the
next iteration of the worker thread.
Discarding a fully provisioned 100G thin volume with 64k block size goes
from 860s to 95s with this change.
Clearly there's something wrong with the worker architecture, more
investigation needed.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Thu, 9 Oct 2014 19:24:12 +0000 (15:24 -0400)]
dm thin: implement thin_merge
Introduce thin_merge so that any additional constraints from the data
volume may be taken into account when determing the maximum number of
sectors that can be issued relative to the specified logical offset.
This is particularly important if/when the data volume is layered ontop
of a more sophisticated device (e.g. dm-raid or some other DM target).
Reviewed-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Thu, 9 Oct 2014 23:32:22 +0000 (19:32 -0400)]
dm: improve documentation and code clarity in dm_merge_bvec
These code changes do not introduce a functional change.
But bio_add_page() will never attempt to build up a bio larger than
queue_max_sectors(). Similarly, bio_get_nr_vecs() is also bound by
queue_max_sectors(). Therefore, there is no point in allowing
dm_merge_bvec() to answer "how many sectors can a bio have at this
offset?" with anything larger than queue_max_sectors(). Using
queue_max_sectors() rather than BIO_MAX_SECTORS serves to more
accurately convey the limits that are being imposed.
Also, use unlikely() to clarify the fact that the defensive code in
dm_merge_bvec() relative to max_size going negative shouldn't ever
happen -- if it does happen there is a bug in the block layer for
requesting larger than dm_merge_bvec()'s initial response for a given
offset. Also, update a comment in dm_merge_bvec() relative to
max_hw_sectors_kb. And fix empty newline whitespace.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Mike Snitzer [Thu, 9 Oct 2014 22:43:25 +0000 (18:43 -0400)]
dm thin: adjust max_sectors_kb based on thinp blocksize
Allows for filesystems to submit bios that are a factor of the thinp
blocksize, improving dm-thinp efficiency (particularly when the data
volume is RAID).
Also set io_min to max_sectors_kb if it is a factor of the thinp
blocksize.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 6 Oct 2014 14:45:59 +0000 (15:45 +0100)]
dm thin: throttle incoming IO
Throttle IO based on the time it's taking the worker to do one loop.
There were reports of hung task timeouts occuring and it was observed
that the excessively long avgqu-sz (as reported by iostat) was
contributing to these hung tasks.
Throttling definitely helps dm-thinp perform better under heavy IO load
(without being detremental by being overzealous). It reduces avgqu-sz
drastically, e.g.: from 60K to ~6K, and even as low as 150 once metadata
is cached by bufio, when dirty_ratio=5, dirty_background_ratio=2. And
avgqu-sz stays at or below 30K even with dirty_ratio=20,
dirty_background_ratio=10.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 6 Oct 2014 14:28:30 +0000 (15:28 +0100)]
dm thin: prefetch missing metadata pages
Prefetch metadata at the start of the worker thread and then again every
128th bio processed from the deferred list.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 6 Oct 2014 14:27:26 +0000 (15:27 +0100)]
dm transaction manager: add support for prefetching blocks of metadata
Introduce the dm_tm_issue_prefetches interface. If you're using a
non-blocking clone the tm will build up a list of requested blocks that
weren't in core. dm_tm_issue_prefetches will request those blocks to be
prefetched.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 6 Oct 2014 14:24:55 +0000 (15:24 +0100)]
dm thin metadata: change dm_thin_find_block to allow blocking, but not issuing, IO
This change is a prerequisite for allowing metadata to be prefetched.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 6 Oct 2014 20:30:06 +0000 (16:30 -0400)]
dm bio prison: switch to using a red black tree
Previously it was using a fixed sized hash table. There are times
when very many concurrent cells are held (such as when processing a very
large discard). When this happens the hash table performance becomes
very poor.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Thu, 9 Oct 2014 10:10:25 +0000 (11:10 +0100)]
dm bufio: evict buffers that are past the max age but retain some buffers
These changes help keep metadata backed by dm-bufio in-core longer which
fixes reports of metadata churn in the face of heavy random IO workloads.
Before, bufio evicted all buffers older than DM_BUFIO_DEFAULT_AGE_SECS.
Having a device (e.g. dm-thinp or dm-cache) lose all metadata just
because associated buffers had been idle for some time is unfriendly.
Now, the user may now configure the number of bytes that bufio retains
using the 'retain_bytes' module parameter. The default is 256K.
Also, the DM_BUFIO_WORK_TIMER_SECS and DM_BUFIO_DEFAULT_AGE_SECS
defaults were quite low so increase them (to 30 and 300 respectively).
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 6 Oct 2014 12:48:51 +0000 (13:48 +0100)]
dm bufio: switch from a huge hash table to an rbtree
Converting over to using an rbtree eliminates a fixed 8MB allocation
from vmalloc space for the hash table.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Joe Thornber [Mon, 10 Nov 2014 15:03:24 +0000 (15:03 +0000)]
dm btree: fix a recursion depth bug in btree walking code
The walk code was using a 'ro_spine' to hold it's locked btree nodes.
But this data structure is designed for the rolling lock scheme, and
as such automatically unlocks blocks that are two steps up the call
chain. This is not suitable for the simple recursive walk algorithm,
which retraces its steps.
This code is only used by the persistent array code, which in turn is
only used by dm-cache. In order to trigger it you need to have a
mapping tree that is more than 2 levels deep; which equates to 8-16
million cache blocks. For instance a 4T ssd with a very small block
size of 32k only just triggers this bug.
The fix just places the locked blocks on the stack, and stops using
the ro_spine altogether.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Joe Thornber [Fri, 10 Oct 2014 08:41:09 +0000 (09:41 +0100)]
dm thin: grab a virtual cell before looking up the mapping
Avoids normal IO racing with discard.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Heinz Mauelshagen [Wed, 29 Oct 2014 18:02:27 +0000 (19:02 +0100)]
dm raid: fix inaccessible superblocks causing oops in configure_discard_support
Commit
48cf06bc5f ("dm raid: add discard support for RAID levels 4, 5
and 6") did not properly handle missing metadata device(s). A failing
read of the superblock causes the metadata and data devices to be
removed from the dev array in struct raid_set, setting references to
both devices to NULL. configure_discard_support() nonetheless tries to
access the data dev unconditionally causing an oops.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Heinz Mauelshagen [Fri, 17 Oct 2014 11:38:50 +0000 (13:38 +0200)]
dm raid: ensure superblock's size matches device's logical block size
The dm-raid superblock (struct dm_raid_superblock) is padded to 512
bytes and that size is being used to read it in from the metadata
device into one preallocated page.
Reading or writing this on a 512-byte sector device works fine but on
a 4096-byte sector device this fails.
Set the dm-raid superblock's size to the logical block size of the
metadata device, because IO at that size is guaranteed too work. Also
add a size check to avoid silent partial metadata loss in case the
superblock should ever grow past the logical block size or PAGE_SIZE.
[includes pointer math fix from Dan Carpenter]
Reported-by: "Liuhua Wang" <lwang@suse.com>
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Mikulas Patocka [Thu, 16 Oct 2014 18:45:20 +0000 (14:45 -0400)]
dm bufio: change __GFP_IO to __GFP_FS in shrinker callbacks
The shrinker uses gfp flags to indicate what kind of operation can the
driver wait for. If __GFP_IO flag is present, the driver can wait for
block I/O operations, if __GFP_FS flag is present, the driver can wait on
operations involving the filesystem.
dm-bufio tested for __GFP_IO. However, dm-bufio can run on a loop block
device that makes calls into the filesystem. If __GFP_IO is present and
__GFP_FS isn't, dm-bufio could still block on filesystem operations if it
runs on a loop block device.
The change from __GFP_IO to __GFP_FS supposedly fixes one observed (though
unreproducible) deadlock involving dm-bufio and loop device.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Pavitra Kumar [Fri, 10 Oct 2014 15:19:46 +0000 (15:19 +0000)]
dm stripe: fix potential for leak in stripe_ctr error path
Fix a potential struct stripe_c leak that would occur if the
chunk_size exceeded the maximum allowed by dm_set_target_max_io_len
(UINT_MAX). However, in practice there is no possibility of this
occuring given that chunk_size is of type uint32_t. But it is good to
fix this to future-proof in case dm_set_target_max_io_len's
implementation were to change.
Signed-off-by: Pavitra Kumar <pavitrak@nvidia.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Alexey Khoroshilov [Wed, 1 Oct 2014 20:58:35 +0000 (22:58 +0200)]
dm log userspace: fix memory leak in dm_ulog_tfr_init failure path
If cn_add_callback() fails in dm_ulog_tfr_init(), it does not
deallocate prealloced memory but calls cn_del_callback().
Found by Linux Driver Verification project (linuxtesting.org).
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Reviewed-by: Jonathan Brassow <jbrassow@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Mikulas Patocka [Wed, 1 Oct 2014 17:29:48 +0000 (13:29 -0400)]
dm bufio: when done scanning return from __scan immediately
When __scan frees the required number of buffer entries that the
shrinker requested (nr_to_scan becomes zero) it must return. Before
this fix the __scan code exited only the inner loop and continued in the
outer loop -- which could result in reduced performance due to extra
buffers being freed (e.g. unnecessarily evicted thinp metadata needing
to be synchronously re-read into bufio's cache).
Also, move dm_bufio_cond_resched to __scan's inner loop, so that
iterating the bufio client's lru lists doesn't result in scheduling
latency.
Reported-by: Joe Thornber <thornber@redhat.com>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org # 3.2+
Joe Thornber [Tue, 30 Sep 2014 08:32:46 +0000 (09:32 +0100)]
dm bufio: update last_accessed when relinking a buffer
The 'last_accessed' member of the dm_buffer structure was only set when
the the buffer was created. This led to each buffer being discarded
after dm_bufio_max_age time even if it was used recently. In practice
this resulted in all thinp metadata being evicted soon after being read
-- this is particularly problematic for metadata intensive workloads
like multithreaded small random IO.
'last_accessed' is now updated each time the buffer is moved to the head
of the LRU list, so the buffer is now properly discarded if it was not
used in dm_bufio_max_age time.
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org # v3.2+
Heinz Mauelshagen [Wed, 24 Sep 2014 15:47:19 +0000 (17:47 +0200)]
dm raid: add discard support for RAID levels 4, 5 and 6
In case of RAID levels 4, 5 and 6 we have to verify each RAID members'
ability to zero data on discards to avoid stripe data corruption -- if
discard_zeroes_data is not set for each RAID member discard support must
be disabled. But given the uncertainty of whether or not a RAID member
properly supports zeroing data on discard we require the user to
explicitly allow discard support on RAID levels 4, 5, and 6 by setting
a dm-raid module paramter, e.g.: dm-raid.devices_handle_discard_safely=Y
Otherwise, discards could cause data corruption on RAID4/5/6.
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Heinz Mauelshagen [Wed, 24 Sep 2014 15:47:18 +0000 (17:47 +0200)]
dm raid: add discard support for RAID levels 1 and 10
Discard support is not enabled for RAID levels 4, 5, and 6 at this time
due to concerns about unreliable discard_zeroes_data support on some
hardware. Otherwise, discards could cause stripe data corruption
(classic example of bad apples spoiling the bunch).
Signed-off-by: Heinz Mauelshagen <heinzm@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Benjamin Marzinski [Wed, 13 Aug 2014 18:53:43 +0000 (13:53 -0500)]
dm: allow active and inactive tables to share dm_devs
Until this change, when loading a new DM table, DM core would re-open
all of the devices in the DM table. Now, DM core will avoid redundant
device opens (and closes when destroying the old table) if the old
table already has a device open using the same mode. This is achieved
by managing reference counts on the table_devices that DM core now
stores in the mapped_device structure (rather than in the dm_table
structure). So a mapped_device's active and inactive dm_tables' dm_dev
lists now just point to the dm_devs stored in the mapped_device's
table_devices list.
This improvement in DM core's device reference counting has the
side-effect of fixing a long-standing limitation of the multipath
target: a DM multipath table couldn't include any paths that were unusable
(failed). For example: if all paths have failed and you add a new,
working, path to the table; you can't use it since the table load would
fail due to it still containing failed paths. Now a re-load of a
multipath table can include failed devices and when those devices become
active again they can be used instantly.
The device list code in dm.c isn't a straight copy/paste from the code in
dm-table.c, but it's very close (aside from some variable renames). One
subtle difference is that find_table_device for the tables_devices list
will only match devices with the same name and mode. This is because we
don't want to upgrade a device's mode in the active table when an
inactive table is loaded.
Access to the mapped_device structure's tables_devices list requires a
mutex (tables_devices_lock), so that tables cannot be created and
destroyed concurrently.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Benjamin Marzinski [Wed, 13 Aug 2014 18:53:42 +0000 (13:53 -0500)]
dm mpath: stop queueing IO when no valid paths exist
'queue_io' is set so that IO is queued while paths are being
initialized. Clear queue_io in __choose_pgpath if there are no valid
paths, since there are obviously no paths that can be initialized.
Otherwise IOs to the device will back up.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Junichi Nomura [Fri, 3 Oct 2014 11:55:26 +0000 (11:55 +0000)]
dm: use bioset_create_nobvec()
Since DM core uses bio_clone_fast() for both bio-based and request-based
DM devices there is no need for DM's bioset to have a bvec mempool.
With this patch, on arch with 4KB page for example, memory usage will be
reduced by 64KB for each bio-based DM device and 1MB for each
request-based DM device.
For example, when you create 10,000 bio-based DM devices and 1,000
request-based DM devices, memory usage of biovec under no load is:
# grep biovec /proc/slabinfo
biovec-256 418068 418068 4096 ...
biovec-128 0 0 2048 ...
biovec-64 0 0 1024 ...
biovec-16 0 0 256 ...
With this patch series applied, the usage becomes:
# grep biovec /proc/slabinfo
biovec-256 116 116 4096 ...
biovec-128 0 0 2048 ...
biovec-64 0 0 1024 ...
biovec-16 0 0 256 ...
So 4096 * (418068 - 116) = 1.6GB of memory is saved in this example.
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Junichi Nomura [Fri, 3 Oct 2014 11:55:16 +0000 (11:55 +0000)]
dm: remove nr_iovecs parameter from alloc_tio()
alloc_tio() uses bio_alloc_bioset() to allocate a clone-bio for a bio.
alloc_tio() takes the number of bvecs to allocate for the clone-bio.
However, with v3.14's immutable biovec changes DM now uses
__bio_clone_fast() and no longer needs to allocate bvecs.
In practice, the 'nr_iovecs' passed to alloc_tio() is always effectively
0. __clone_and_map_simple_bio() looked like it was passing non-zero
nr_iovecs, but its value was always within the range of inline bvecs and
no allocation actually happened. If allocation happened, the BUG_ON() in
__bio_clone_fast() would've triggered.
Remove the nr_iovecs parameter from alloc_tio() to prevent possible
future bio_alloc_bioset() mis-use of a new bioset interface that will no
longer allow bvecs to be allocated.
Also fix extra whitespace before the __bio_clone_fast() call in
__clone_and_map_simple_bio().
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Junichi Nomura [Fri, 3 Oct 2014 21:27:12 +0000 (17:27 -0400)]
block: add bioset_create_nobvec()
Users of bio_clone_fast() do not want bios with their own bvecs.
Allocating a bvec mempool as part of the bioset intended for such users
is a waste of memory.
bioset_create_nobvec() creates a bioset that doesn't have the bvec
mempool.
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Junichi Nomura [Fri, 3 Oct 2014 21:27:11 +0000 (17:27 -0400)]
block: use bio_clone_fast() in blk_rq_prep_clone()
Request cloning clones bios in the request to track the completion
of each bio.
For that purpose, we can use bio_clone_fast() instead of bio_clone()
to avoid unnecessary allocation and copy of bvecs.
This patch reduces memory footprint of request-based device-mapper
(about 1-4KB for each request) and is a preparation for further
reduction of memory usage by removing unused bvec mempool.
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Hannes Reinecke [Wed, 1 Oct 2014 12:32:31 +0000 (14:32 +0200)]
block: misplaced rq_complete tracepoint
The rq_complete tracepoint was never issued for empty requests,
causing the resulting blktrace information to never show any
completion for those request.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:08 +0000 (19:20 -0400)]
sd: Honor block layer integrity handling flags
A set of flags introduced in the block layer enable better control over
how protection information is handled. These flags are useful for both
error injection and data recovery purposes. Checking can be enabled and
disabled for controller and disk, and the guard tag format is now a
per-I/O property.
Update sd_protect_op to communicate the relevant information to the
low-level device driver via a set of flags in scsi_cmnd.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Rasmus Villemoes [Tue, 16 Sep 2014 20:51:16 +0000 (22:51 +0200)]
block: Replace strnicmp with strncasecmp
The kernel used to contain two functions for length-delimited,
case-insensitive string comparison, strnicmp with correct semantics
and a slightly buggy strncasecmp. The latter is the POSIX name, so
strnicmp was renamed to strncasecmp, and strnicmp made into a wrapper
for the new strncasecmp to avoid breaking existing users.
To allow the compat wrapper strnicmp to be removed at some point in
the future, and to avoid the extra indirection cost, do
s/strnicmp/strncasecmp/g.
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:07 +0000 (19:20 -0400)]
block: Add T10 Protection Information functions
The T10 Protection Information format is also used by some devices that
do not go through the SCSI layer (virtual block devices, NVMe). Relocate
the relevant functions to a block layer library that can be used without
involving SCSI.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:06 +0000 (19:20 -0400)]
block: Don't merge requests if integrity flags differ
We'd occasionally merge requests with conflicting integrity flags.
Introduce a merge helper which checks that the requests have compatible
integrity payloads.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:05 +0000 (19:20 -0400)]
block: Integrity checksum flag
Make the choice of checksum a per-I/O property by introducing a flag
that can be inspected by the SCSI layer. There are several reasons for
this:
1. It allows us to switch choice of checksum without unloading and
reloading the HBA driver.
2. During error recovery we need to be able to tell the HBA that
checksums read from disk should not be verified and converted to IP
checksums.
3. For error injection purposes we need to be able to write a bad guard
tag to storage. Since the storage device only supports T10 CRC we
need to be able to disable IP checksum conversion on the HBA.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:04 +0000 (19:20 -0400)]
block: Relocate bio integrity flags
Move flags affecting the integrity code out of the bio bi_flags and into
the block integrity payload.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:03 +0000 (19:20 -0400)]
block: Add a disk flag to block integrity profile
So far we have relied on the app tag size to determine whether a disk
has been formatted with T10 protection information or not. However, not
all target devices provide application tag storage.
Add a flag to the block integrity profile that indicates whether the
disk has been formatted with protection information.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@dev.mellanox.co.il>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:02 +0000 (19:20 -0400)]
block: Add prefix to block integrity profile flags
Add a BLK_ prefix to the integrity profile flags. Also rename the flags
to be more consistent with the generate/verify terminology in the rest
of the integrity code.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:01 +0000 (19:20 -0400)]
block: Clean up the code used to generate and verify integrity metadata
Instead of the "operate" parameter we pass in a seed value and a pointer
to a function that can be used to process the integrity metadata. The
generation function is changed to have a return value to fit into this
scheme.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:20:00 +0000 (19:20 -0400)]
block: Make protection interval calculation generic
Now that the protection interval has been detached from the sector size
we need to be able to handle sizes that are different from 4K and
512. Make the interval calculation generic.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:19:59 +0000 (19:19 -0400)]
block: Deprecate the use of the term sector in the context of block integrity
The protection interval is not necessarily tied to the logical block
size of a block device. Stop using the terms "sector" and "sectors".
Going forward we will use the term "seed" to describe the initial
reference tag value for a given I/O. "Interval" will be used to describe
the portion of the data buffer that a given piece of protection
information is associated with.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:19:58 +0000 (19:19 -0400)]
block: Remove bip_buf
bip_buf is not really needed so we can remove it.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:19:57 +0000 (19:19 -0400)]
block: Remove integrity tagging functions
None of the filesystems appear interested in using the integrity tagging
feature. Potentially because very few storage devices actually permit
using the application tag space.
Remove the tagging functions.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:19:56 +0000 (19:19 -0400)]
block: Replace bi_integrity with bi_special
For commands like REQ_COPY we need a way to pass extra information along
with each bio. Like integrity metadata this information must be
available at the bottom of the stack so bi_private does not suffice.
Rename the existing bi_integrity field to bi_special and make it a union
so we can have different bio extensions for each class of command.
We previously used bi_integrity != NULL as a way to identify whether a
bio had integrity metadata or not. Introduce a REQ_INTEGRITY to be the
indicator now that bi_special can contain different things.
In addition, bio_integrity(bio) will now return a pointer to the
integrity payload (when applicable).
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Martin K. Petersen [Fri, 26 Sep 2014 23:19:55 +0000 (19:19 -0400)]
block: Get rid of bdev_integrity_enabled()
bdev_integrity_enabled() is only used by bio_integrity_enabled().
Combine these two functions.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:47 +0000 (23:23 +0800)]
blk-mq: support per-distpatch_queue flush machinery
This patch supports to run one single flush machinery for
each blk-mq dispatch queue, so that:
- current init_request and exit_request callbacks can
cover flush request too, then the buggy copying way of
initializing flush request's pdu can be fixed
- flushing performance gets improved in case of multi hw-queue
In fio sync write test over virtio-blk(4 hw queues, ioengine=sync,
iodepth=64, numjobs=4, bs=4K), it is observed that througput gets
increased a lot over my test environment:
- throughput: +70% in case of virtio-blk over null_blk
- throughput: +30% in case of virtio-blk over SSD image
The multi virtqueue feature isn't merged to QEMU yet, and patches for
the feature can be found in below tree:
git://kernel.ubuntu.com/ming/qemu.git v2.1.0-mq.4
And simply passing 'num_queues=4 vectors=5' should be enough to
enable multi queue(quad queue) feature for QEMU virtio-blk.
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:46 +0000 (23:23 +0800)]
block: introduce 'blk_mq_ctx' parameter to blk_get_flush_queue
This patch adds 'blk_mq_ctx' parameter to blk_get_flush_queue(),
so that this function can find the corresponding blk_flush_queue
bound with current mq context since the flush queue will become
per hw-queue.
For legacy queue, the parameter can be simply 'NULL'.
For multiqueue case, the parameter should be set as the context
from which the related request is originated. With this context
info, the hw queue and related flush queue can be found easily.
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:45 +0000 (23:23 +0800)]
block: flush: avoid to figure out flush queue unnecessarily
Just figuring out flush queue at the entry of kicking off flush
machinery and request's completion handler, then pass it through.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:44 +0000 (23:23 +0800)]
block: remove blk_init_flush() and its pair
Now mission of the two helpers is over, and just call
blk_alloc_flush_queue() and blk_free_flush_queue() directly.
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:43 +0000 (23:23 +0800)]
block: introduce blk_flush_queue to drive flush machinery
This patch introduces 'struct blk_flush_queue' and puts all
flush machinery related fields into this structure, so that
- flush implementation details aren't exposed to driver
- it is easy to convert to per dispatch-queue flush machinery
This patch is basically a mechanical replacement.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:42 +0000 (23:23 +0800)]
block: avoid to use q->flush_rq directly
This patch trys to use local variable to access flush request,
so that we can convert to per-queue flush machinery a bit easier.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:41 +0000 (23:23 +0800)]
block: move flush initialization to blk_flush_init
These fields are always used with the flush request, so
initialize them together.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:40 +0000 (23:23 +0800)]
block: introduce blk_init_flush and its pair
These two temporary functions are introduced for holding flush
initialization and de-initialization, so that we can
introduce 'flush queue' easier in the following patch. And
once 'flush queue' and its allocation/free functions are ready,
they will be removed for sake of code readability.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:39 +0000 (23:23 +0800)]
blk-mq: allocate flush_rq in blk_mq_init_flush()
It is reasonable to allocate flush req in blk_mq_init_flush().
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Thu, 25 Sep 2014 15:23:38 +0000 (23:23 +0800)]
blk-mq: handle failure path for initializing hctx
Failure of initializing one hctx isn't handled, so this patch
introduces blk_mq_init_hctx() and its pair to handle it explicitly.
Also this patch makes code cleaner.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Mon, 22 Sep 2014 13:59:31 +0000 (15:59 +0200)]
scsi: move blk_mq_start_request call earlier
Some ATA drivers need the dma drain size workaround, and thus need to
call blk_mq_start_request before the S/G mapping.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Mon, 22 Sep 2014 16:21:48 +0000 (10:21 -0600)]
block: fix blk_abort_request on blk-mq
Signed-off-by: Christoph Hellwig <hch@lst.de>
Moved blk_mq_rq_timed_out() definition to the private blk-mq.h header.
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Fri, 19 Sep 2014 13:53:46 +0000 (21:53 +0800)]
blk-timeout: fix blk_add_timer
Commit
8cb34819cdd5d(blk-mq: unshared timeout handler) introduces
blk-mq's own timeout handler, and removes following line:
blk_queue_rq_timed_out(q, blk_mq_rq_timed_out);
which then causes blk_add_timer() to bypass adding the timer,
since blk-mq no longer has q->rq_timed_out_fn defined.
This patch fixes the problem by bypassing the check for blk-mq,
so that both request deadlines are still set and the rolling
timer updated.
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 17 Sep 2014 14:27:03 +0000 (08:27 -0600)]
blk-mq: limit memory consumption if a crash dump is active
It's not uncommon for crash dump kernels to be limited to 128MB or
something low in that area. This is normally not a problem for
devices as we don't use that much memory, but for some shared SCSI
setups with huge queue depths, it can potentially fill most of
memory with tons of request allocations. blk-mq does scale back
when it fails to allocate memory, but it scales back just enough
so that blk-mq succeeds. This could still leave the system with
not enough memory to make any real progress.
Check if we are in a kdump environment and limit the hardware
queues and tag depth.
Signed-off-by: Jens Axboe <axboe@fb.com>
Ming Lei [Wed, 17 Sep 2014 09:47:58 +0000 (17:47 +0800)]
blk-mq: remove unnecessary blk_clear_rq_complete()
This patch removes two unnecessary blk_clear_rq_complete(),
the REQ_ATOM_COMPLETE flag is cleared inside blk_mq_start_request(),
so:
- The blk_clear_rq_complete() in blk_flush_restore_request()
needn't because the request will be freed later, and clearing
it here may open a small race window with timeout.
- The blk_clear_rq_complete() in blk_mq_requeue_request() isn't
necessary too, even though REQ_ATOM_STARTED is cleared in
__blk_mq_requeue_request(), in theory it still may cause a small
race window with timeout since the two clear_bit() may be
reordered.
Signed-off-by: Ming Lei <ming.lei@canoical.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Sat, 13 Sep 2014 23:40:13 +0000 (16:40 -0700)]
blk-mq: pass a reserved argument to the timeout handler
Allow blk-mq to pass an argument to the timeout handler to indicate
if we're timing out a reserved or regular command. For many drivers
those need to be handled different.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Sat, 13 Sep 2014 23:40:12 +0000 (16:40 -0700)]
blk-mq: unshared timeout handler
Duplicate the (small) timeout handler in blk-mq so that we can pass
arguments more easily to the driver timeout handler. This enables
the next patch.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Sat, 13 Sep 2014 23:40:11 +0000 (16:40 -0700)]
blk-mq: fix and simplify tag iteration for the timeout handler
Don't do a kmalloc from timer to handle timeouts, chances are we could be
under heavy load or similar and thus just miss out on the timeouts.
Fortunately it is very easy to just iterate over all in use tags, and doing
this properly actually cleans up the blk_mq_busy_iter API as well, and
prepares us for the next patch by passing a reserved argument to the
iterator.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Sat, 13 Sep 2014 23:40:10 +0000 (16:40 -0700)]
blk-mq: rename blk_mq_end_io to blk_mq_end_request
Now that we've changed the driver API on the submission side use the
opportunity to fix up the name on the completion side to fit into the
general scheme.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Sat, 13 Sep 2014 23:40:09 +0000 (16:40 -0700)]
blk-mq: call blk_mq_start_request from ->queue_rq
When we call blk_mq_start_request from the core blk-mq code before calling into
->queue_rq there is a racy window where the timeout handler can hit before we've
fully set up the driver specific part of the command.
Move the call to blk_mq_start_request into the driver so the driver can start
the request only once it is fully set up.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Christoph Hellwig [Sat, 13 Sep 2014 23:40:08 +0000 (16:40 -0700)]
blk-mq: remove REQ_END
Pass an explicit parameter for the last request in a batch to ->queue_rq
instead of using a request flag. Besides being a cleaner and non-stateful
interface this is also required for the next patch, which fixes the blk-mq
I/O submission code to not start a time too early.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 22 Sep 2014 17:57:32 +0000 (11:57 -0600)]
Merge branch 'for-linus' into for-3.18/core
Moving patches from for-linus to 3.18 instead, pull in this changes
that will go to Linus today.
Jens Axboe [Fri, 19 Sep 2014 19:10:29 +0000 (13:10 -0600)]
blk-mq: use blk_mq_start_hw_queues() when running requeue work
When requests are retried due to hw or sw resource shortages,
we often stop the associated hardware queue. So ensure that we
restart the queues when running the requeue work, otherwise the
queue run will be a no-op.
Signed-off-by: Jens Axboe <axboe@fb.com>