Adds a log/journal to dedup. At the end of txg, instead of writing the
entry directly to the ZAP, instead its adding to an in-memory tree and
appended to an on-disk object. The on-disk object is only read at
import, to reload the in-memory tree.
Lookups first go the the log tree before going to the ZAP, so
recently-used entries will remain close by in memory. This vastly
reduces overhead from dedup IO, as it will not have to do so many
read/update/write cycles on ZAP leaf nodes.
A flushing facility is added at end of txg, to push logged entries out
to the ZAP. There's actually two separate "logs" (in-memory tree and
on-disk object), one active (recieving updated entries) and one flushing
(writing out to disk). These are swapped (ie flushing begins) based on
memory used by the in-memory log trees and time since we last flushed
something.
The flushing facility monitors the amount of entries coming in and being
flushed out, and calibrates itself to try to flush enough each txg to
keep up with the ingest rate without competing too much with other IO.
Multiple tuneables are provided to control the flushing facility.
All the histograms and stats are update to accomodate the log as a
separate entry store. zdb gains knowledge of how to count them and dump
them. Documentation included!
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15895
All objects stored in the MOS get copies=3. For a large dedup table,
this requires significant extra IO and disk space, when its not really
necessary - the dedup table itself isn't needed to read or write data,
only to keep data usage down. Losing the dedup table does not render the
pool unusable, it just messes up the accounting somewhat.
This adds a dmu_ddt_copies tuneable. When set to 0, the existing
behaviour is used. When set higher, dedup table blocks (ZAP and log)
will have this many copies rather than the usual 3, while indirect
blocks will have one more again.
This is a tuneable for now mostly for testing. Losing a dedup table can
cause blocks to be leaked, and we currently have no facilities to repair
that.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15895
This yields substantial performance improvements when we only write out
some small % of entries at a time, as it will cause entries that will go
into "nearby" ZAP leaf nodes to be grouped closer together in the AVL, and
so touch fewer blocks. Without this, the distribution is an even spread,
so we touch a lot more ZAP leaf nodes for any given number of entries.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15895
Both the API and the code were kinda mangled and I was really struggling
to follow it. The worst offender was the old ddt_stat_add(); after
fixing it up the rest of the changes are mostly knock-on effects and
targets of opportunity.
Note that the old ddt_stat_add() was safe against overflows - it could
produce crazy numbers, but the compiler wouldn't do anything stupid. The
assertions in ddt_stat_sub() go a lot of the way to protecting against
this; getting in a position where overflows are a problem is definitely
a programming error.
Also expanding ddt_stat_add() and ddt_histogram_empty() produces less
efficient assembly. I'm not bothered about this right now though; these
should not be hot functions, and if they are we'll optimise them later.
If we have to go back to the old form, we'll comment it like crazy.
Finally, I've removed the assertion that the bucket will never be
negative, as it will soon be possible to have entries with zero
refcounts: an entry for a block that is no longer on the pool, but is on
the log waiting to be synced out. It might be better to have a separate
bucket for these, since they're still using real space on disk, but
ultimately these stats are driving UI, and for now I've chosen to keep
them matching how they've looked in the past, as well as match the
operators mental model - pool usage is managed elsewhere.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15895
Traditional dedup keeps a separate ddt_phys_t "type" for each possible
count of DVAs (that is, copies=) parameter. Each of these are tracked
independently of each other, and have their own set of DVAs. This leads
to an (admittedly rare) situation where you can create as many as six
copies of the data, by changing the copies= parameter between copying.
This is both a waste of storage on disk, but also a waste of space in
the stored DDT entries, since there never needs to be more than three
DVAs to handle all possible values of copies=.
This commit adds a new FDT feature, DDT_FLAG_FLAT. When active, only the
first ddt_phys_t is used. Each time a block is written with the dedup
bit set, this single phys is checked to see if it has enough DVAs to
fulfill the request. If it does, the block is filled with the saved DVAs
as normal. If not, an adjusted write is issued to create as many extra
copies as are needed to fulfill the request, which are then saved into
the entry too.
Because a single phys is no longer an all-or-nothing, but can be
transitioning from fewer to more DVAs, the write path now has to keep a
copy of the previous "known good" DVA set so we can revert to it in case
an error occurs. zio_ddt_write() has been restructured and heavily
commented to make it much easier to see what's happening.
Backwards compatibility is maintained simply by allocating four
ddt_phys_t when the DDT_FLAG_FLAT flag is not set, and updating the phys
selection macros to check the flag. In the old arrangement, each number
of copies gets a whole phys, so it will always have either zero or all
necessary DVAs filled, with no in-between, so the old behaviour
naturally falls out of the new code.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Co-authored-by: Don Brady <don.brady@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15893
This slims down the in-memory entry to as small as it can be. The
IO-related parts are made into a separate entry, since they're
relatively rarely needed.
The variable allocation for dde_phys is to support the upcoming flat
format.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15893
The idea here is that sometimes you need the contents of an entry with
no intent to modify it, and/or from a place where its difficult to get
hold of its originating ddt_t to know how to interpret it.
A lightweight entry contains everything you might need to "read" an
entry - its key, type and phys contents - but none of the extras for
modifying it or using it in a larger context. It also has the full
complement of phys slots, so it can represent any kind of dedup entry
without having to know the specific configuration of the table it came
from.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15893
The "flat phys" feature will use only a single phys slot for all
entries, which means the old "single", "double" etc naming now makes no
sense, and more importantly, means that choosing the right slot for a
given block pointer will depend on how many slots are in use for a given
DDT.
This removes the old names, and adds accessor macros to decouple
specific phys array indexes from any particular meaning.
(These macros look strange in isolation, mainly in the way they take the
ddt_t* as an arg but don't use it. This is mostly a separate commit to
introduce the concept to the reader before the "flat phys" commit
extends it).
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15893
The upcoming dedup features break the long held assumption that all
blocks on disk with a 'D' dedup bit will always be present in the DDT,
or will have the same set of DVA allocations on disk as in the DDT.
If the DDT is no longer a complete picture of all the dedup blocks that
will be and should be on disk, then it does us no good to walk and prime
it up front, since it won't necessarily match up with every block we'll
see anyway.
Instead, we rework things here to be more like the BRT checks. When we
see a dedup'd block, we look it up in the DDT, consume a refcount, and
for the second-or-later instances, count them as duplicates.
The DDT and BRT are moved ahead of the space accounting. This will
become important for the "flat" feature, which may need to count a
modified version of the block.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Allan Jude <allan@klarasystems.com>
Co-authored-by: Don Brady <don.brady@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15892
This is the supporting infrastructure for the upcoming dedup features.
Traditionally, dedup objects live directly in the MOS root. While their
details vary (checksum, type and class), they are all the same "kind" of
thing - a store of dedup entries.
The new features are more varied than that, and are better thought of as
a set of related stores for the overall state of a dedup table.
This adds a new feature flag, SPA_FEATURE_FAST_DEDUP. Enabling this will
cause new DDTs to be created as a ZAP in the MOS root, named
DDT-<checksum>. The is used as the root object for the normal type/class
store objects, but will also be a place for any storage required by new
features.
This commit adds two new fields to ddt_t, for version and flags. These
are intended to describe the structure and features of the overall dedup
table, and are stored as-is in the DDT root. In this commit, flags are
always zero, but the intent is that they can be used to hang optional
logic or state onto for new dedup features. Version is always 1.
For a "legacy" dedup table, where no DDT root directory exists, the
version will be 0.
ddt_configure() is expected to determine the version and flags features
currently in operation based on whether or not the fast_dedup feature is
enabled, and from what's available on disk. In this way, its possible to
support both old and new tables.
This also provides a migration path. A legacy setup can be upgraded to
FDT by creating the DDT root ZAP, moving the existing objects into it,
and setting version and flags appropriately. There's no support for that
here, but it would be straightforward to add later and allows the
possibility that newer features could be applied to existing dedup
tables.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15892
If a zvol is renamed, and it has one or more snapshots, and
snapdev=visible is true for the zvol, then the rename causes a kernel
null pointer dereference error. This has the effect (on Linux, anyway)
of killing the z_zvol taskq kthread, with locks still held; which in
turn causes a variety of zvol-related operations afterward to hang
indefinitely (such as udev workers, among other things).
The problem occurs because of an oversight in #15486
(e36ff84c33). As documented in
dataset_kstats_create, some datasets may not actually have kstats
allocated for them; and at least at the present time, this is true for
snapshots. In practical terms, this means that for snapshots,
dk->dk_kstats will be NULL. The dataset_kstats_rename function
introduced in the patch above does not first check whether dk->dk_kstats
is NULL before proceeding, unlike e.g. the nearby
dataset_kstats_update_* functions.
In the very particular circumstance in which a zvol is renamed, AND that
zvol has one or more snapshots, AND that zvol also has snapdev=visible,
zvol_rename_minors_impl will loop over not just the zvol dataset itself,
but each of the zvol's snapshots as well, so that their device nodes
will be renamed as well. This results in dataset_kstats_create being
called for snapshots, where, as we've established, dk->dk_kstats is
NULL.
Fix this by simply adding a NULL check before doing anything in
dataset_kstats_rename.
This still allows the dataset_name kstat value for the zvol to be
updated (as was the intent of the original patch), and merely blocks
attempts by the code to act upon the zvol's non-kstat-having snapshots.
If at some future time, kstats are added for snapshots, then things
should work as intended in that case as well.
Signed-off-by: Justin Gottula <justin@jgottula.com>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Alan Somers <asomers@gmail.com>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Report the correct error message in libzfs when attaching/replacing a
vdev with a higher ashift.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ameer Hamza <ahamza@ixsystems.com>
Closes#16449
Reviewed-by: Tony Nguyen <tony.nguyen@delphix.com>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes#16431
It's now the caller's responsibility do special handling for holes if
that's something it wants.
This also makes zio_compress_data() and zio_decompress_data() properly
the inverse of each other.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tino Reichardt <milky-zfs@mcmilk.de>
Reviewed-by: Jason Lee <jasonlee@lanl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16326
Previous code pruned 10% of dnodes once 3/4 of metadata appeared
unevictable. On workloads with many millions of dnodes and little
other metadata it creates significant load spikes for many seconds
straight. This change instead gradually increases pruning as
unevictable metadata grow above the 3/4, which may allow it to
stabilize at some level.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Closes#16401
- Skip config lock enter/exit for embedded blocks. They have no
DVAs, so there is nothing to check under the lock.
- Skip CHECKSUM check and properly check PSIZE for embedded blocks.
- Add static branch predictions for unlikely conditions.
- Do not verify DVAs for blocks already in ARC. ARC hit already
"verified" the first (often the only) DVA, and it does not worth to
enter/exit config lock for nothing.
Some profiles show me up to 3% of CPU saving from this change.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Closes#16387
Spare and l2cache vdev labels are not updated during import. Therefore,
if disk paths are updated between pool export and import, the AUX label
still shows the old paths. This patch syncs the AUX label
during import to show the correct path information.
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Umer Saleem <usaleem@ixsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ameer Hamza <ahamza@ixsystems.com>
Closes#15817
Currently, if a minor is in use when we try to remove it, we'll skip it
and never come back to it again. Since the zvol state is hung off the
minor in the kernel, this can get us into weird situations if something
tries to use it after the removal fails. It's even worse at pool export,
as there's now a vestigial zvol state with no pool under it. It's
weirder again if the pool is subsequently reimported, as the zvol code
(reasonably) assumes the zvol state has been properly setup, when it's
actually left over from the previous import of the pool.
This commit attempts to tackle that by setting a flag on the zvol if its
minor can't be removed, and then checking that flag when a request is
made and rejecting it, thus stopping new work coming in.
The flag also causes a condvar to be signaled when the last client
finishes. For the case where a single minor is being removed (eg
changing volmode), it will wait for this signal before proceeding.
Meanwhile, when removing all minors, a background task is created for
each minor that couldn't be removed on the spot, and those tasks then
wake and clean up.
Since any new tasks are queued on to the pool's spa_zvol_taskq,
spa_export_common() will continue to wait at export until all minors are
removed.
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#14872Closes#16364
This includes the last 12.x release (now EOL) and 13.0 development
versions (<1300139).
Sponsored-by: https://despairlabs.com/sponsor/
Signed-off-by: Rob Norris <robn@despairlabs.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tino Reichardt <milky-zfs@mcmilk.de>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
dmu_buf_will_clone() calls arc_buf_destroy() if there is an associated
ARC buffer with the dbuf. However, this can only be done conditionally.
If the previous dirty record's dr_data is pointed at db_dbf then
destroying it can lead to NULL pointer deference when syncing out the
previous dirty record.
This updates dmu_buf_fill_clone() to only call arc_buf_destroy() if the
previous dirty records dr_data is not pointing to db_buf. The block
clone wil still set the dbuf's db_buf and db_data to NULL, but this will
not cause any issues as any previous dirty record dr_data will still be
pointing at the ARC buffer.
Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Brian Atkinson <batkinson@lanl.gov>
Closes#16337
Fix multiple build errors on FreeBSD.
The main reason is, that the variable 'dxattr_obj' is used
uninitialized within the start of the 'out label'.
Signed-off-by: Tino Reichardt <milky-zfs@mcmilk.de>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
sa_add_projid() gets called via zfs_setattr() for setting project id
on old file/dir, which were created before upgrading to project quota
feature. This function does lookup for all possible SA and update them
all together along with project ID at needed fixed offset. But its
missing lookup and update of SA_ZPL_DXATTR, effectively it losses
SA_ZPL_DXATTR.
Closes#16287
Signed-off-by: Jitendra Patidar <jitendra.patidar@nutanix.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Rob Norris <rob.norris@klarasystems.com>
Make sure log record don't stray beyond valid memory region.
There is a lack of verification of the space occupied by fixed members
of lr_t in the zil_parse.
We can create a crafted image to trigger an out of bounds read by
following these steps:
1) Do some file operations and reboot to simulate abnormal exit
without umount
2) zil_chain.zc_nused: 0x1000
3) First lr_t
lr_t.lrc_txtype: 0x0
lr_t.lrc_reclen: 0x1000-0xb8-0x1
lr_t.lrc_txg: 0x0
lr_t.lrc_seq: 0x1
4) Update checksum in zil_chain.zc_eck
Fix:
Add some checks to make sure the remaining bytes are large enough to
hold an log record.
Signed-off-by: XDTG <click1799@163.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Linux provides SLAB_RECLAIM_ACCOUNT and __GFP_RECLAIMABLE flags to
mark memory allocations that can be freed via shinker calls. It
should allow kernel to tune and group such allocations for lower
memory fragmentation and better reclamation under pressure.
This patch marks as reclaimable most of ARC memory, directly
evictable via ZFS shrinker, plus also dnode/znode/sa memory,
indirectly evictable via kernel's superblock shrinker.
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Allan Jude <allan@klarasystems.com>
spa_preferred_class() selects a storage class based on (among other
things) the DMU object type. This only works for old-style object types
that match only one specific kind of thing. For DMU_OTN_ types we need
another way to signal the storage class.
This commit allows the object type to be overridden in the IO policy for
the purposes of choosing a storage class. It then adds the ability to
set the storage type on a dnode hold, such that all writes generated
under that hold will get it.
This method has two shortcomings:
- it would be better if we could "name" a set of storage class
preferences rather than it being implied by the object type.
- it would be better if this info were stored in the dnode on disk.
In the absence of those things, this seems like the smallest possible
change.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15894
Rather than picking out specific values out of the properties, just pass
the entire zio in, to make it easier in the future to use more of that
info to decide on the storage class.
I would have rathered just pass io_prop in, but having spa.h include
zio.h gets a bit tricky.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Closes#15894
Neither FreeBSD nor Linux currently implement kmem_cache_set_move(),
which means dnode_move() is never called. In such situation use of
dnode handles with respective locking to access dnode from dbuf is
a waste of time for no benefit.
This patch implements optional simplified code for such platforms,
saving at least 3 dnode lock/dereference/unlock per dbuf life cycle.
Originally I hoped to drop the handles completely to save memory,
but they are still used in dnodes allocation code, so left for now.
Before this change in CPU profiles of some workloads I saw 4-20% of
CPU time spent in zrl_add_impl()/zrl_remove(), which are gone now.
Reviewed-by: Rob Wing <rob.wing@klarasystems.com
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Closes#16374
- Use the macros in few places it was missed.
- Reduce scope of DB_DNODE_ENTER/EXIT() and inline some DB_DNODE()
uses to make it more obvious what exactly is protected there and
make unprotected accesses by mistake more difficult.
- Make use of zrl_owner().
Reviewed-by: Rob Wing <rob.wing@klarasystems.com
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Closes#16374
This change adds a new `zpool prefetch -t ddt $pool` command which
causes a pool's DDT to be loaded into the ARC. The primary goal is to
remove the need to "warm" a pool's cache before deduplication stops
slowing write performance. It may also provide a way to reload portions
of a DDT if they have been flushed due to inactivity.
Sponsored-by: iXsystems, Inc.
Sponsored-by: Catalogics, Inc.
Sponsored-by: Klara, Inc.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Will Andrews <will.andrews@klarasystems.com>
Signed-off-by: Fred Weigel <fred.weigel@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Signed-off-by: Don Brady <don.brady@klarasystems.com>
Co-authored-by: Will Andrews <will.andrews@klarasystems.com>
Co-authored-by: Don Brady <don.brady@klarasystems.com>
Closes#15890
There's no good way to tell when a ZIL commit fails and falls back to a
transaction sync, other than perhaps a throughput drop. This adds
counters so we can see when it happens and why.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
- When receiving memory pressure signal from OS be more strict
trying to free some memory. Otherwise kernel may come again and
request much more. Return as result how much arc_c was actually
reduced due to this request, that may be less than requested.
- On Linux when receiving direct reclaim from some file system
(that may be ZFS) instead of ignoring request completely, just
shrink the ARC, but do not wait for eviction. Waiting there may
cause deadlock. Ignoring it as before may put extra pressure on
other caches and/or swap, and cause OOM if nothing help. While
not waiting may result in more ARC evicted later, and may be too
late if OOM killer activate right now, but I hope it to be better
than doing nothing at all.
- On Linux set arc_no_grow before waiting for reclaim, not after,
or it may grow back while we are waiting.
- On Linux add new parameter zfs_arc_shrinker_seeks to balance
ARC eviction cost, relative to page cache and other subsystems.
- Slightly update Linux arc_set_sys_free() math for new kernels.
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Reviewed-by: Rob Norris <rob.norris@klarasystems.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
This adds two new pool properties:
- dedup_table_size, the total size of all DDTs on the pool; and
- dedup_table_quota, the maximum possible size of all DDTs in the pool
When set, quota will be enforced by checking when a new entry is about
to be created. If the pool is over its dedup quota, the entry won't be
created, and the corresponding write will be converted to a regular
non-dedup write. Note that existing entries can be updated (ie their
refcounts changed), as that reuses the space rather than requiring more.
dedup_table_quota can be set to 'auto', which will set it based on the
size of the devices backing the "dedup" allocation device. This makes it
possible to limit the DDTs to the size of a dedup vdev only, such that
when the device fills, no new blocks are deduplicated.
Sponsored-by: iXsystems, Inc.
Sponsored-By: Klara Inc.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Signed-off-by: Don Brady <don.brady@klarasystems.com>
Co-authored-by: Don Brady <don.brady@klarasystems.com>
Co-authored-by: Rob Wing <rob.wing@klarasystems.com>
Co-authored-by: Sean Eric Fagan <sean.fagan@klarasystems.com>
Closes#15889
Sponsored-by: Klara, Inc.
Sponsored-By: Wasabi Technology, Inc.
Signed-off-by: Don Brady <don.brady@klarasystems.com>
Co-authored-by: Don Brady <don.brady@klarasystems.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
For files smaller than recordsize, it's most likely that they don't have
L1 blocks. However, current calculation will always return at least 1 L1
block.
In this change, we check dnode level to figure out if it has L1 blocks
or not, and return 0 if it doesn't. This will reduce the chance of
unnecessary throttling when deleting a large number of small files.
Signed-off-by: Chunwei Chen <david.chen@nutanix.com>
Co-authored-by: Chunwei Chen <david.chen@nutanix.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
After c3f2f1aa2, vdev_fault_wanted is set on a vdev after a probe fails.
An end-of-txg async task is charged with actually faulting the vdev.
In a single-disk pool, the probe failure will degrade the last disk, and
then suspend the pool. However, vdev_fault_wanted is not cleared. After
the pool returns, the transaction finishes and the async task runs and
faults the vdev, which suspends the pool again.
The fix is simple: when reopening a vdev, clear the async fault flag. If
the vdev is still failed, the startup probe will quickly notice and
degrade/suspend it again. If not, all is well!
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Co-authored-by: Don Brady <don.brady@klarasystems.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Reviewed-by: Jorgen Lundman <lundman@lundman.net>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Don Brady <don.brady@klarasystems.com>
In the commit of the head_errlog feature we introduced a bug in
dsl_dataset_promote_sync(): we may dereference origin_head and hds, both
dereferencing ddpa after calling promote_sync() on ddpa.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Chunwei Chen <david.chen@nutanix.com>
Reviewed-by: Rob Norris <robn@despairlabs.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#16272Closes#16273
We always call it twice with JUSTLOOKING and then FORREAL.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Pawel Jakub Dawidek <pawel@dawidek.net>
Closes#16225
Optionally turn off disk's enclosure slot if an I/O is hung
triggering the deadman.
It's possible for outstanding I/O to a misbehaving SCSI disk to
neither promptly complete or return an error. This can occur due
to retry and recovery actions taken by the SCSI layer, driver, or
disk. When it occurs the pool will be unresponsive even though
there may be sufficient redundancy configured to proceeded without
this single disk.
When a hung I/O is detected by the kmods it will be posted as a
deadman event. By default an I/O is considered to be hung after
5 minutes. This value can be changed with the zfs_deadman_ziotime_ms
module parameter. If ZED_POWER_OFF_ENCLOSURE_SLOT_ON_DEADMAN is set
the disk's enclosure slot will be powered off causing the outstanding
I/O to fail. The ZED will then handle this like a normal disk failure.
By default ZED_POWER_OFF_ENCLOSURE_SLOT_ON_DEADMAN is not set.
As part of this change `zfs_deadman_events_per_second` is added
to control the ratelimitting of deadman events independantly of
delay events. In practice, a single deadman event is sufficient
and more aren't particularly useful.
Alphabetize the zfs_deadman_* entries in zfs.4.
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#16226
- Add old eviction for special and dedup metaslab classes. Those
vdevs may be potentially big and fragmented with large metaslabs,
while their asynchronous write pattern is not really different
from normal class. It seems an omission to not evict old metaslabs
from them.
- If we have metaslab preload enabled, which means we are not too
low on memory, do not evict active metaslabs even if they are not
used for some time. Eviction of active metaslabs means we won't
be able to write anything until we load them, that may take some
time, that is straight opposite to metaslab preload goals. For
small systems the memory saving should be less important after
recent reduction in number of allocators and so open metaslabs.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Closes#16214
In case of error dmu_buf_fill_done() returns the buffer back into
DB_UNCACHED state. Since during transition from DB_UNCACHED into
DB_FILL state dbuf_noread() allocates an ARC buffer, we must free
it here, otherwise it will be leaked.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Jorgen Lundman <lundman@lundman.net>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored by: iXsystems, Inc.
Closes#15665Closes#15802Closes#16216
At the end of l2arc_evict() fix an assertion in the case that l2ad_hand
+ distance == l2ad_end.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#16202Closes#16207
If a shrink or truncate had recently freed a portion of the ZAP, the
dbuf could still be sitting on the dbuf cache waiting for eviction. If
it is then allocated for a new leaf before it can be evicted, the
zap_leaf_t is still attached as userdata, tripping the VERIFY.
Instead, just check for the userdata, and if we find it, reuse it.
Sponsored-by: Klara, Inc.
Sponsored-by: iXsystems, Inc.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16157.
Closes#16204
On 32-bit platforms long long is generally 64-bits. Sufficiently modern
versions of gcc (13 in my testing) complains when casting a pointer to
an integer of a different width so cast to uintptr_t first to avoid the
warning.
Fixes: c183d164aa Parallel pool import
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Don Brady <don.brady@klarasystems.com>
Signed-off-by: Brooks Davis <brooks.davis@sri.com>
Closes#16203
Even though block cloning is much faster than regular copying,
it is not instantaneous - the file might be large and the recordsize
small. It would be nice to be able to interrupt it with a signal
(e.g., SIGINFO on FreeBSD to see the progress).
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Pawel Jakub Dawidek <pawel@dawidek.net>
Closes#16208
In 92dc4ad83 I updated the dbuf_cache accounting to track the size of
userdata associated with dbufs. This adds the size of the dbuf+userdata
together in a single call to zfs_refcount_add_many(), but sometime
removes them in separate calls to zfs_refcount_remove_many(), if dbuf
and userdata are evicted separately.
What I didn't realise is that when refcount tracking is on,
zfs_refcount_add_many() and zfs_refcount_remove_many() are expected to
be paired, with their second & third args (count & holder) the same on
both sides. Splitting the remove part into two calls means the counts
don't match up, tripping a panic.
This commit fixes that, by always adding and removing the dbuf and
userdata counts separately.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reported-by: Mark Johnston <markj@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16191
This renames it to spa_taskq_dispatch(), and reduces and simplifies its
arguments based on these observations from its two call sites:
- arg is always the zio, so it can be typed that way, and we don't need
to provide it twice;
- ent is always &zio->io_tqent, and zio is always provided, so we can
use it directly;
- the only flag used is TQ_FRONT, which can just be a bool;
- zio != NULL was part of the "use allocator" test, but it never would
have got that far, because that arg was only set to NULL in the
reexecute path, which is forced to type CLAIM, so the condition would
fail at t == WRITE anyway.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16151
It is the only user of spa_taskq_dispatch_select(), so might as well
just carry it directly.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16151
It has no callers anymore.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16151
When stack space is tight, the stream is written to its target on a
separate taskq thread to make sure there's enough stack space to
complete it.
This has always used an IO taskq, but that doesn't really make sense for
it, and moving it onto a regular taskq lets us get rid of
spa_taskq_dispatch_sync(), which is not used anywhere else.
Stream writes may block for a long time depending on what the target is,
and we have no way of discovering this, so we can't risk using the
system taskq, as there may be many tens of sends in progress. Instead,
we create a dedicated taskq thread for each send writer to run on, and
clean it up when it's done.
Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes#16151
If we wait until after we check for no spa references to drop the
namespace lock, then we know that spa consumers will need to call
spa_lookup() and end up waiting on the spa_namespace_cv until we
finish. This narrows the external checks to spa_lookup and we no
longer need to worry about the spa_vdev_enter case.
Sponsored-By: Klara Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: George Wilson <gwilson@delphix.com>
Signed-off-by: Don Brady <don.brady@klarasystems.com>
Closes#16153
