linux.git/drivers/md/dm-writecache.c, branch v5.8-rc2

dm writecache: improve performance on DDR persistent memory (Optane)

2020-05-15T14:29:36+00:00

When testing the dm-writecache target on a real DDR persistent memory
(Intel Optane), it turned out that explicit cache flushing using the
clflushopt instruction performs better than non-temporal stores for
block sizes 1k, 2k and 4k.

The dm-writecache target is singlethreaded (all the copying is done
while holding the writecache lock), so it benefits from clwb, see:
http://lore.kernel.org/r/alpine.LRH.2.02.2004160411460.7833@file01.intranet.prod.int.rdu2.redhat.com

Add a new function memcpy_flushcache_optimized() that tests if
clflushopt is present - and if it is, we use it instead of
memcpy_flushcache.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: remove superfluous test in persistent_memory_claim

2020-05-15T14:29:36+00:00

Remove superfluous test if dax_dev is NULL - dax_direct_access already
does this test.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: fix data corruption when reloading the target

2020-04-16T20:04:13+00:00

The dm-writecache reads metadata in the target constructor. However, when
we reload the target, there could be another active instance running on
the same device. This is the sequence of operations when doing a reload:

1. construct new target
2. suspend old target
3. resume new target
4. destroy old target

Metadata that were written by the old target between steps 1 and 2 would
not be visible by the new target.

Fix the data corruption by loading the metadata in the resume handler.

Also, validate block_size is at least as large as both the devices'
logical block size and only read 1 block from the metadata during
target constructor -- no need to read entirety of metadata now that it
is done during resume.

Fixes: 48debafe4f2f ("dm: add writecache target")
Cc: stable@vger.kernel.org # v4.18+
Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: add cond_resched to avoid CPU hangs

2020-03-27T18:36:50+00:00

Initializing a dm-writecache device can take a long time when the
persistent memory device is large.  Add cond_resched() to a few loops
to avoid warnings that the CPU is stuck.

Cc: stable@vger.kernel.org # v4.18+
Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: optimize superblock write

2020-03-24T15:55:09+00:00

If we write a superblock in writecache_flush, we don't need to set bit and
scan the bitmap for it - we can just write the superblock directly. Also,
we can set the flag REQ_FUA on the write bio, so that we don't need to
submit a flush bio afterwards.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: implement gradual cleanup

2020-03-24T15:55:08+00:00

If a block is stored in the cache for too long, it will now be
written to the underlying device and cleaned up.

Add a new option "max_age" that specifies the maximum age of a block
in milliseconds.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: implement the "cleaner" policy

2020-03-24T15:50:30+00:00

The "flush" or "flush_on_suspend" messages flush the whole cache. However,
these flushing methods can take some time and the process is left in
an interruptible state during the flush.

Implement a "cleaner" option that offers an alternate flushing method.
When this option is activated (either by a message or in the constructor
arguments), the cache will not promote new writes (however, writes to
already cached blocks are promoted, to avoid data corruption due to
misordered writes) and it will gradually writeback any cached data. The
userspace can then monitor the cleaning process with "dmsetup status".
When the number of cached bloks drops to zero, the userspace can unload
the dm-writecache target and replace it with dm-linear or other targets.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm writecache: do direct write if the cache is full

2020-03-24T15:26:18+00:00

If the cache device is full, we do a direct write to the origin device.
Note that we must not do it if the written block is already in the cache.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm: bump version of core and various targets

2020-03-03T16:10:21+00:00

Changes made during the 5.6 cycle warrant bumping the version number
for DM core and the targets modified by this commit.

It should be noted that dm-thin, dm-crypt and dm-raid already had
their target version bumped during the 5.6 merge window.

Signed-off-by; Mike Snitzer

dm writecache: verify watermark during resume

2020-02-27T21:44:24+00:00

Verify the watermark upon resume - so that if the target is reloaded
with lower watermark, it will start the cleanup process immediately.

Fixes: 48debafe4f2f ("dm: add writecache target")
Cc: stable@vger.kernel.org # 4.18+
Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer