Commit Graph

3289 Commits

Author SHA1 Message Date
Matthew Ahrens 4e1d1b4b92 assertion failed in arc_wait_for_eviction()
If the system is very low on memory (specifically,
`arc_free_memory() < arc_sys_free/2`, i.e. less than 1/16th of RAM
free), `arc_evict_state_impl()` will defer wakups.  In this case, the
arc_evict_waiter_t's remain on the list, even though `arc_evict_count`
has been incremented past their `aew_count`.

The problem is that `arc_wait_for_eviction()` assumes that if there are
waiters on the list, the count they are waiting for has not yet been
reached.  However, the deferred wakeups may violate this, causing
`ASSERT(last->aew_count > arc_evict_count)` to fail.

This commit resolves the issue by having new waiters use the greater of
`arc_evict_count` and the last `aew_count`.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: George Amanakis <gamanakis@gmail.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #11285
Closes #11397
2021-01-23 15:47:06 -08:00
Brian Behlendorf dd487640b2
Linux 5.10 compat: restore custom uio_prefaultpages()
As part of commit 1c2358c1 the custom uio_prefaultpages() code
was removed in favor of using the generic kernel provided
iov_iter_fault_in_readable() interface.  Unfortunately, it
turns out that up until the Linux 4.7 kernel the function would
only ever fault in the first iovec of the iov_iter.  The result
being uiomove_iov() may hang waiting for the page.

This commit effectively restores the custom uio_prefaultpages()
pages code for Linux 4.9 and earlier kernels which contain the
troublesome version of iov_iter_fault_in_readable().

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11463 
Closes #11484
2021-01-22 09:58:49 -08:00
Konstantin Khorenko 59570a05d8 VZ 7 kernel compat: introduce ITER-enabled .direct_IO() via IOVECs
Virtuozzo 7 kernels starting 3.10.0-1127.18.2.vz7.163.46
have the following configuration:

  * no HAVE_VFS_RW_ITERATE
  * HAVE_VFS_DIRECT_IO_ITER_RW_OFFSET

=> let's add implementation of zpl_direct_IO() via
zpl_aio_{read,write}() in this case.

https://bugs.openvz.org/browse/OVZ-7243

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Konstantin Khorenko <khorenko@virtuozzo.com>
Closes #11410 
Closes #11411
2021-01-05 10:33:41 -08:00
Toomas Soome 921ec61b77 implicit conversion from 'boolean_t' to 'ds_hold_flags_t'
Build error on illumos with gcc 10 did reveal:

In function 'dmu_objset_refresh_ownership':
../../common/fs/zfs/dmu_objset.c:857:25: error: implicit conversion
from 'boolean_t' to 'ds_hold_flags_t' {aka 'enum ds_hold_flags'}
[-Werror=enum-conversion]
      857 |  dsl_dataset_disown(ds, decrypt, tag);
          |                         ^~~~~~~
cc1: all warnings being treated as errors

libzfs_input_check.c: In function 'zfs_ioc_input_tests':
libzfs_input_check.c:754:28: error: implicit conversion from
'enum dmu_objset_type' to 'enum lzc_dataset_type'
[-Werror=enum-conversion]
  754 |  err = lzc_create(dataset, DMU_OST_ZFS, NULL, NULL, 0);
      |                            ^~~~~~~~~~~
cc1: all warnings being treated as errors

The same issue is present in openzfs, and also the same issue about
ds_hold_flags_t, which currently defines exactly one valid value.

Reviewed-by: Igor Kozhukhov <igor@dilos.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Toomas Soome <tsoome@me.com>
Closes #11406
2021-01-05 10:30:19 -08:00
Brian Behlendorf fcd9966ed9 Linux 5.11 compat: blk_{un}register_region()
As of 5.11 the blk_register_region() and blk_unregister_region()
functions have been retired. This isn't a problem since add_disk()
has implicitly allocated minor numbers for a very long time.

Reviewed-by: Rafael Kitover <rkitover@gmail.com>
Reviewed-by: Coleman Kane <ckane@colemankane.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11387
Closes #11390
2021-01-05 10:26:39 -08:00
Brian Behlendorf a2621753b2 Linux 5.11 compat: revalidate_disk_size()
Both revalidate_disk_size() and revalidate_disk() have been removed.
Functionally this isn't a problem because we only relied on these
functions to call zvol_revalidate_disk() for us and to perform any
additional handling which might be needed for that kernel version.
When neither are available we know there's no additional handling
needed and we can directly call zvol_revalidate_disk().

Reviewed-by: Rafael Kitover <rkitover@gmail.com>
Reviewed-by: Coleman Kane <ckane@colemankane.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11387
Closes #11390
2021-01-05 10:26:32 -08:00
Brian Behlendorf e888f28988 Linux 5.11 compat: bdev_whole()
The bd_contains member was removed from the block_device structure.
Callers needing to determine if a vdev is a whole block device should
use the new bdev_whole() wrapper.  For older kernels we provide our
own bdev_whole() wrapper which relies on bd_contains for compatibility.

Reviewed-by: Rafael Kitover <rkitover@gmail.com>
Reviewed-by: Coleman Kane <ckane@colemankane.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11387
Closes #11390
2021-01-05 10:26:25 -08:00
Brian Behlendorf 305510fd33 Linux 5.11 compat: bio_start_io_acct() / bio_end_io_acct()
The generic IO accounting functions have been removed in favor of the
bio_start_io_acct() and bio_end_io_acct() functions which provide a
better interface.  These new functions were introduced in the 5.8
kernels but it wasn't until the 5.11 kernel that the previous generic
IO accounting interfaces were removed.

This commit updates the blk_generic_*_io_acct() wrappers to provide
and interface similar to the updated kernel interface.  It's slightly
different because for older kernels we need to pass the request queue
as well as the bio.

Reviewed-by: Rafael Kitover <rkitover@gmail.com>
Reviewed-by: Coleman Kane <ckane@colemankane.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11387
Closes #11390
2021-01-05 10:26:18 -08:00
Brian Behlendorf 67cff6e4c1 Linux 5.11 compat: lookup_bdev()
The lookup_bdev() function has been updated to require a dev_t
be passed as the second argument. This is actually pretty nice
since the major number stored in the dev_t was the only part we
were interested in. This allows to us avoid handling the bdev
entirely.  The vdev_lookup_bdev() wrapper was updated to emulate
the behavior of the new lookup_bdev() for all supported kernels.

Reviewed-by: Rafael Kitover <rkitover@gmail.com>
Reviewed-by: Coleman Kane <ckane@colemankane.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11387
Closes #11390
2021-01-05 10:26:09 -08:00
Andy Fiddaman cee725c9bd Dangling reference from dmu_objset_upgrade
After porting the fix for https://github.com/openzfs/zfs/issues/5295
over to illumos, we started hitting an assertion failure when running
the testsuite:

	assertion failed: rc->rc_count == number, file: .../refcount.c

and the unexpected hold has this stack:

	dsl_dataset_long_hold+0x59 dmu_objset_upgrade+0x73
dmu_objset_id_quota_upgrade+0x15 dmu_objset_own+0x14f

The simplest reproducer for this in illumos is

    zpool create -f -O version=1 testpool c3t0d0; zpool destroy testpool

which is run as part of the zpool_create_tempname test, but I can't get
this to trigger on FreeBSD. This appears to be because of the call to
txg_wait_synced() in dmu_objset_upgrade_stop() (which was missing in
illumos), slows down dmu_objset_disown() enough to avoid the condition.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Andy Fiddaman <andy@omnios.org>
Closes #11368
2020-12-23 14:35:47 -08:00
Michael D Labriola 7a7e101437 Linux 5.10 compat: also zvol_revalidate_disk()
Commit 59b68723 added a configure check for 5.10, which removed
revalidate_disk(), and conditionally replaced it's usage with a call to
the new revalidate_disk_size() function.  However, the old function also
invoked the device's registered callback, in our case
zvol_revalidate_disk().  This commit adds a call to zvol_revalidate_disk()
in zvol_update_volsize() to make sure the code path stays the same.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Michael D Labriola <michael.d.labriola@gmail.com>
Closes #11358
2020-12-23 14:35:47 -08:00
Brian Behlendorf 401ba57ccd Fix maybe uninitialized variable warning
Commit 1c2358c12 restructured this code and introduced a warning
about the variable maybe not being initialized.  This cannot happen
with the updated code but we should initialize the variable anyway
to silence the warning.

    zpl_file.c: In function ‘zpl_iter_write’:
    zpl_file.c:324:9: warning: ‘count’ may be used uninitialized
        in this function [-Wmaybe-uninitialized]

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11373
2020-12-23 14:35:47 -08:00
Brian Behlendorf 188950df9e Remove iov_iter_advance() from iter_read
There's no need to call iov_iter_advance() in zpl_iter_read().
This was preserved from the previous code where it wasn't needed
but also didn't cause any problems.  Now that the iter functions
also handle pipes that's no longer the case.  When fully reading a
pipe buffer iov_iter_advance() may results in the pipe buf release
function being called which will not be registered resulting in
a NULL dereference.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11375 
Closes #11378
2020-12-23 14:35:47 -08:00
Brian Behlendorf 58bc86c5cb Linux 5.10 compat: use iov_iter in uio structure
As of the 5.10 kernel the generic splice compatibility code has been
removed.  All filesystems are now responsible for registering a
->splice_read and ->splice_write callback to support this operation.

The good news is the VFS provided generic_file_splice_read() and
iter_file_splice_write() callbacks can be used provided the ->iter_read
and ->iter_write callback support pipes.  However, this is currently
not the case and only iovecs and bvecs (not pipes) are ever attached
to the uio structure.

This commit changes that by allowing full iov_iter structures to be
attached to uios.  Ever since the 4.9 kernel the iov_iter structure
has supported iovecs, kvecs, bvevs, and pipes so it's desirable to
pass the entire thing when possible.  In conjunction with this the
uio helper functions (i.e uiomove(), uiocopy(), etc) have been
updated to understand the new UIO_ITER type.

Note that using the kernel provided uio_iter interfaces allowed the
existing Linux specific uio handling code to be simplified.  When
there's no longer a need to support kernel's older than 4.9, then
it will be possible to remove the iovec and bvec members from the
uio structure and always use a uio_iter.  Until then we need to
maintain all of the existing types for older kernels.

Some additional refactoring and cleanup was included in this change:

- Added checks to configure to detect available iov_iter interfaces.
  Some are available all the way back to the 3.10 kernel and are used
  when available.  In particular, uio_prefaultpages() now always uses
  iov_iter_fault_in_readable() which is available for all supported
  kernels.

- The unused UIO_USERISPACE type has been removed.  It is no longer
  needed now that the uio_seg enum is platform specific.

- Moved zfs_uio.c from the zcommon.ko module to the Linux specific
  platform code for the zfs.ko module.  This gets it out of libzfs
  where it was never needed and keeps this Linux specific code out
  of the common sources.

- Removed unnecessary O_APPEND handling from zfs_iter_write(), this
  is redundant and O_APPEND is already handled in zfs_write();

NOTE: Cleanly applying this kernel compatibility change required
applying the following commits.  This makes the change larger than
it absolutely needs to be, but the resulting code matches what's in
the branch branch.  This is both more tested and makes it easier to
apply any future backports in this area.

7cf4cd824 Remove incorrect assertion
783be694f Reduce confusion in zfs_write
af5626ac2 Return EFAULT at the end of zfs_write() when set
cc1f85be8 Simplify offset and length limit in zfs_write
9585538d0 Const some unchanging variables in zfs_write
86e74dc16 Remove redundant oid parameter to update_pages
b3d723fb0 Factor uid, gid, and projid out of loop in zfs_write
3d40b6554 Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD

Reviewed-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11351
2020-12-23 14:35:39 -08:00
Brian Behlendorf 7cf4cd8246 Remove incorrect assertion
Commit 85703f6 added a new ASSERT to zfs_write() as part of the
cleanup which isn't correct in the case where multiple processes
are concurrently extending a file.  The `zp->z_size` is updated
atomically while holding a range lock on only a portion of the
file.  Therefore, it's possible for the file size to increase
after a same check is performed earlier in the loop causing this
ASSERT to fail.  The code itself handles this case correctly so
only the invalid ASSERT needs to be removed.

Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11235
2020-12-23 14:35:00 -08:00
Ryan Moeller 783be694f1 Reduce confusion in zfs_write
Is this block when abuf != NULL ever reached? Yes, it is.

Add asserts and comments to prove that when we get here, we have a full
block write at an aligned offset extending past EOF.

Simplify by removing the check that tx_bytes == max_blksz, since we can
assert that it is always true.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11191
2020-12-23 14:35:00 -08:00
Ryan Moeller af5626ac27 Return EFAULT at the end of zfs_write() when set
FreeBSD's VFS expects EFAULT from zfs_write() if we didn't complete
the full write so it can retry the operation.  Add some missing
SET_ERRORs in zfs_write().

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11193
2020-12-23 14:35:00 -08:00
Ryan Moeller cc1f85be8b Simplify offset and length limit in zfs_write
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11176
2020-12-23 14:34:59 -08:00
Ryan Moeller 9585538d0e Const some unchanging variables in zfs_write
Show that these values will not be changing later.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11176
2020-12-23 14:34:59 -08:00
Ryan Moeller 86e74dc162 Remove redundant oid parameter to update_pages
The oid comes from the znode we are already passing.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11176
2020-12-23 14:34:59 -08:00
Ryan Moeller b3d723fb0e Factor uid, gid, and projid out of loop in zfs_write
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11176
2020-12-23 14:34:59 -08:00
Matthew Macy 3d40b65540 Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD
The zfs_fsync, zfs_read, and zfs_write function are almost identical
between Linux and FreeBSD.  With a little refactoring they can be
moved to the common code which is what is done by this commit.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Closes #11078
2020-12-23 14:34:59 -08:00
Matthew Ahrens a103ae446e special device removal space accounting fixes
The space in special devices is not included in spa_dspace (or
dsl_pool_adjustedsize(), or the zfs `available` property).  Therefore
there is always at least as much free space in the normal class, as
there is allocated in the special class(es).  And therefore, there is
always enough free space to remove a special device.

However, the checks for free space when removing special devices did not
take this into account.  This commit corrects that.

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Don Brady <don.brady@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #11329
2020-12-23 14:34:59 -08:00
Ryan Libby ee49d9e02b lua: avoid gcc -Wreturn-local-addr bug
Avoid a bug with gcc's -Wreturn-local-addr warning with some
obfuscation.  In buggy versions of gcc, if a return value is an
expression that involves the address of a local variable, and even if
that address is legally converted to a non-pointer type, a warning may
be emitted and the value of the address may be replaced with zero.
Howerver, buggy versions don't emit the warning or replace the value
when simply returning a local variable of non-pointer type.

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90737

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Libby <rlibby@FreeBSD.org>
Closes #11337
2020-12-23 14:34:59 -08:00
George Amanakis 900480bd96 Fix reporting of CKSUM errors in indirect vdevs
When removing and subsequently reattaching a vdev, CKSUM errors may
occur as vdev_indirect_read_all() reads from all children of a mirror
in case of a resilver.

Fix this by checking whether a child is missing the data and setting a
flag (ic_error) which is then checked in vdev_indirect_repair() and
suppresses incrementing the checksum counter.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #11277
2020-12-23 14:34:59 -08:00
Ryan Moeller 8847b06bf6 FreeBSD: Implement sysctl for fletcher4 impl
There is a tunable to select the fletcher 4 checksum implementation on
Linux but it was not present in FreeBSD.

Implement the sysctl handler for FreeBSD and use ZFS_MODULE_PARAM_CALL
to provide the tunable on both platforms.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11270
2020-12-23 14:34:59 -08:00
Paul Dagnelie 21adfb031c Fix kernel panic induced by redacted send
In the redaction list traversal code, there is a bug in the binary search
logic when looking for the resume point. Maxbufid can be decremented to -1,
causing us to read the last possible block of the object instead of the one we
wanted. This can cause incorrect resume behavior, or possibly even a hang in
some cases. In addition, when examining non-last blocks, we can treat the
block as being the same size as the last block, causing us to miss entries in
the redaction list when determining where to resume. Finally, we were ignoring
the case where the resume point was found in the buffer being searched, and
resuming from minbufid. All these issues have been corrected, and the code has
been significantly simplified to make future issues less likely.

Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes #11297
2020-12-23 14:34:59 -08:00
Ryan Moeller ae2cfdf8a7 FreeBSD: Fix format of vfs.zfs.arc_no_grow_shift
vfs.zfs.arc_no_grow_shift has an invalid type (15) and this causes
py-sysctl to format it as a bytearray when it should be an integer.

"U" is not a valid format, it should be "I" and the type should match
the variable type, int.  We can return EINVAL if the value is set below
zero.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11318
2020-12-23 14:34:59 -08:00
Brian Behlendorf f217a2b902 Fix possibly uninitialized 'root_inode' variable warning
Resolve an uninitialized variable warning when compiling.

    In function ‘zfs_domount’:
    warning: ‘root_inode’ may be used uninitialized in this
        function [-Wmaybe-uninitialized]
    sb->s_root = d_make_root(root_inode);

Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11306
2020-12-23 14:34:59 -08:00
Ryan Moeller fb3ad5d24e FreeBSD: Do zcommon_init sooner to avoid FPU panic
There has been a panic affecting some system configurations where the
thread FPU context is disturbed during the fletcher 4 benchmarks,
leading to a panic at boot.

module_init() registers zcommon_init to run in the last subsystem
(SI_SUB_LAST).  Running it as soon as interrupts have been configured
(SI_SUB_INT_CONFIG_HOOKS) makes sure we have finished the benchmarks
before we start doing other things.

While it's not clear *how* the FPU context was being disturbed, this
does seem to avoid it.

Add a module_init_early() macro to run zcommon_init() at this earlier
point on FreeBSD.  On Linux this is defined as module_init().

Authored by: Konstantin Belousov <kib@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11302
2020-12-23 14:34:59 -08:00
Brian Behlendorf 038aaec1cd Fix optional "force" arg handing in zfs_ioc_pool_sync()
The fnvlist_lookup_boolean_value() function should not be used
to check the force argument since it's optional.  It may not be
provided or may have been created with the wrong flags.

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11281
Closes #11284
2020-12-23 14:34:59 -08:00
Brian Behlendorf 07ca433973 Reduce fletcher4 and raidz benchmark times
During module load time all of the available fetcher4 and raidz
implementations are benchmarked for a fixed amount of time to
determine the fastest available.  Manual testing has shown that this
time can be significantly reduced with negligible effect on the final
results.

This commit changes the benchmark time to 1ms which can reduce the
module load time by over a second on x86_64.  On an x86_64 system
with sse3, ssse3, and avx2 instructions the benchmark times are:

    Fletcher4    603ms   -> 15ms
    RAIDZ        1,322ms -> 64ms

Reviewed-by: Matthew Macy <mmacy@freebsd.org>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11282
2020-12-23 14:34:59 -08:00
Alexander Motin 1d02bdee6c Fix for "Reduce latency effects of non-interactive I/O"
It was found that setting min_active tunables for non-interactive I/Os
makes them stuck.  It is caused by zfs_vdev_nia_delay, that can never
be reached if we never issue any I/Os due to min_active set to zero.

Fix this by issuing at least one non-interactive I/O at a time when
there are no interactive I/Os.  When there are interactive I/Os, zero
min_active allows to completely block any non-interactive I/O.  It may
min_active starvation in some scenarios, but who we are to deny foot
shooting?

Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes #11261
2020-12-23 13:09:17 -08:00
Alexander Motin 2080c4f27e Reduce latency effects of non-interactive I/O
Investigating influence of scrub (especially sequential) on random read
latency I've noticed that on some HDDs single 4KB read may take up to 4
seconds!  Deeper investigation shown that many HDDs heavily prioritize
sequential reads even when those are submitted with queue depth of 1.

This patch addresses the latency from two sides:
 - by using _min_active queue depths for non-interactive requests while
   the interactive request(s) are active and few requests after;
 - by throttling it further if no interactive requests has completed
   while configured amount of non-interactive did.

While there, I've also modified vdev_queue_class_to_issue() to give
more chances to schedule at least _min_active requests to the lowest
priorities.  It should reduce starvation if several non-interactive
processes are running same time with some interactive and I think should
make possible setting of zfs_vdev_max_active to as low as 1.

I've benchmarked this change with 4KB random reads from ZVOL with 16KB
block size on newly written non-fragmented pool.  On fragmented pool I
also saw improvements, but not so dramatic.  Below are log2 histograms
of the random read latency in milliseconds for different devices:

4 2x mirror vdevs of SATA HDD WDC WD20EFRX-68EUZN0 before:
0, 0, 2,  1,  12,  21,  19,  18, 10, 15, 17, 21
after:
0, 0, 0, 24, 101, 195, 419, 250, 47,  4,  0,  0
, that means maximum latency reduction from 2s to 500ms.

4 2x mirror vdevs of SATA HDD WDC WD80EFZX-68UW8N0 before:
0, 0,  2,  31,  38,  28,  18,  12, 17, 20, 24, 10, 3
after:
0, 0, 55, 247, 455, 470, 412, 181, 36,  0,  0,  0, 0
, i.e. from 4s to 250ms.

1 SAS HDD SEAGATE ST14000NM0048 before:
0,  0,  29,   70, 107,   45,  27, 1, 0, 0, 1, 4, 19
after:
1, 29, 681, 1261, 676, 1633,  67, 1, 0, 0, 0, 0,  0
, i.e. from 4s to 125ms.

1 SAS SSD SEAGATE XS3840TE70014 before (microseconds):
0, 0, 0, 0, 0, 0, 0, 0,  70, 18343, 82548, 618
after:
0, 0, 0, 0, 0, 0, 0, 0, 283, 92351, 34844,  90

I've also measured scrub time during the test and on idle pools.  On
idle fragmented pool I've measured scrub getting few percent faster
due to use of QD3 instead of QD2 before.  On idle non-fragmented pool
I've measured no difference.  On busy non-fragmented pool I've measured
scrub time increase about 1.5-1.7x, while IOPS increase reached 5-9x.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes #11166
2020-12-23 13:09:03 -08:00
Ryan Moeller 7735c9addf FreeBSD: notify userspace when a vdev is removed
This is needed for zfsd to autoreplace vdevs.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11260
2020-12-23 13:08:12 -08:00
Brian Behlendorf 2c36eb763f Revert "Reduce latency effects of non-interactive I/O"
Under certain conditions commit a3a4b8def appears to result in a
hang, or poor performance, when importing a pool.  Until the root
cause can be identified it has been reverted from the release branch.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #11245
2020-11-30 09:43:09 -08:00
Alexander Motin a3a4b8def7 Reduce latency effects of non-interactive I/O
Investigating influence of scrub (especially sequential) on random read
latency I've noticed that on some HDDs single 4KB read may take up to 4
seconds!  Deeper investigation shown that many HDDs heavily prioritize
sequential reads even when those are submitted with queue depth of 1.

This patch addresses the latency from two sides:
 - by using _min_active queue depths for non-interactive requests while
   the interactive request(s) are active and few requests after;
 - by throttling it further if no interactive requests has completed
   while configured amount of non-interactive did.

While there, I've also modified vdev_queue_class_to_issue() to give
more chances to schedule at least _min_active requests to the lowest
priorities.  It should reduce starvation if several non-interactive
processes are running same time with some interactive and I think should
make possible setting of zfs_vdev_max_active to as low as 1.

I've benchmarked this change with 4KB random reads from ZVOL with 16KB
block size on newly written non-fragmented pool.  On fragmented pool I
also saw improvements, but not so dramatic.  Below are log2 histograms
of the random read latency in milliseconds for different devices:

4 2x mirror vdevs of SATA HDD WDC WD20EFRX-68EUZN0 before:
0, 0, 2,  1,  12,  21,  19,  18, 10, 15, 17, 21
after:
0, 0, 0, 24, 101, 195, 419, 250, 47,  4,  0,  0
, that means maximum latency reduction from 2s to 500ms.

4 2x mirror vdevs of SATA HDD WDC WD80EFZX-68UW8N0 before:
0, 0,  2,  31,  38,  28,  18,  12, 17, 20, 24, 10, 3
after:
0, 0, 55, 247, 455, 470, 412, 181, 36,  0,  0,  0, 0
, i.e. from 4s to 250ms.

1 SAS HDD SEAGATE ST14000NM0048 before:
0,  0,  29,   70, 107,   45,  27, 1, 0, 0, 1, 4, 19
after:
1, 29, 681, 1261, 676, 1633,  67, 1, 0, 0, 0, 0,  0
, i.e. from 4s to 125ms.

1 SAS SSD SEAGATE XS3840TE70014 before (microseconds):
0, 0, 0, 0, 0, 0, 0, 0,  70, 18343, 82548, 618
after:
0, 0, 0, 0, 0, 0, 0, 0, 283, 92351, 34844,  90

I've also measured scrub time during the test and on idle pools.  On
idle fragmented pool I've measured scrub getting few percent faster
due to use of QD3 instead of QD2 before.  On idle non-fragmented pool
I've measured no difference.  On busy non-fragmented pool I've measured
scrub time increase about 1.5-1.7x, while IOPS increase reached 5-9x.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes #11166
2020-11-25 08:45:38 -08:00
Matthew Macy 45061cc797 FreeBSD: decouple ZFS_DEBUG from kernel debug settings
Reviewed-by: Martelli Nikola @martellini
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Closes #11213
2020-11-25 08:42:26 -08:00
Brian Behlendorf ccfa35c6f9 Correct missing zil_claim() DTL updates
Commit a1d477c2 accidentally disabled DTL updates for the zil_claim()
case described at the end of vdev_stat_update() by unconditionally
disabling all DTL updates when loading.  This was done to avoid
a deadlock on the vd_dtl_lock when loading the DTLs from disk.

    vdev_dtl_contains <--- Takes vd->vd_dtl_lock
    vdev_mirror_child_missing
    vdev_mirror_io_start
    zio_vdev_io_start
    __zio_execute
    arc_read
    dbuf_issue_final_prefetch
    dbuf_prefetch_impl
    dbuf_prefetch
    dmu_prefetch
    space_map_iterate
    space_map_load_length
    space_map_load
    vdev_dtl_load <--- Takes vd->vd_dtl_lock
    vdev_load
    spa_ld_load_vdev_metadata
    spa_tryimport

The missing DTL updates can be restored by moving the space_map_load()
call outside the vd_dtl_lock.  A private range tree is populated by
reading the space map and then merged in to the DTL_MISSING tree
under the lock.

Furthermore, the SPA_LOAD_NONE check in vdev_dtl_contains() leads to an
additional problem.  Any resilvering which occurs before SPA_LOAD_NONE
is set will incorrectly determine that there's nothing to repair.  This
can result in full redundancy not being restored for some blocks.

Reviewed-by: Matt Ahrens <matt@delphix.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11218
2020-11-22 10:01:43 -08:00
Matthew Macy 043ef5c25e Fix problems in zvol_set_volmode_impl
- Don't leave fstrans set when passed a snapshot
- Don't remove minor if volmode already matches new value
- (FreeBSD) Wait for GEOM ops to complete before trying
  remove (at create time GEOM will be "tasting" in parallel)
- (FreeBSD) Don't leak zvol_state_lock on open if zv == NULL
- (FreeBSD) Don't try to unlock zv->zv_state lock if zv == NULL

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Closes #11199
2020-11-17 12:20:09 -08:00
loli10K f6f3089cf6 Fix 'zfs userspace' for received datasets in encrypted root
For encrypted receives, where user accounting is initially disabled on
creation, both 'zfs userspace' and 'zfs groupspace' fails with
EOPNOTSUPP: this is because dmu_objset_id_quota_upgrade_cb() forgets to
set OBJSET_FLAG_USERACCOUNTING_COMPLETE on the objset flags after a
successful dmu_objset_space_upgrade().

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: loli10K <ezomori.nozomu@gmail.com>
Closes #9501
Closes #9596
2020-11-17 12:19:51 -08:00
George Amanakis a09aeb9fc4 Fix ASSERT logic in l2arc_evict()
In case of cache device removal it is possible that at the end of
l2arc_evict() we have l2ad_hand = l2ad_evict. This can lead to the
following panic in case of a debug build:

VERIFY3(dev->l2ad_hand < dev->l2ad_evict) failed (321920512 < 321920512)
Call Trace:
 dump_stack+0x66/0x90
 spl_panic+0xef/0x117 [spl]
 l2arc_remove_vdev+0x11d/0x290 [zfs]
 spa_load_l2cache+0x275/0x5b0 [zfs]
 spa_vdev_remove+0x4a5/0x6e0 [zfs]
 zfs_ioc_vdev_remove+0x59/0xa0 [zfs]
 zfsdev_ioctl_common+0x5b3/0x630 [zfs]
 zfsdev_ioctl+0x53/0xe0 [zfs]
 do_vfs_ioctl+0x42e/0x6b0
 ksys_ioctl+0x5e/0x90
 do_syscall_64+0x5b/0x1a0
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

In case of cache device removal it also possible that l2ad_hand +
distance > l2ad_end since we do not iterate l2arc_evict() and l2ad_hand
is not reset. This has no functional consequence however as the cache
device is about to be removed.

Fix this by omitting the ASSERT in case of device removal.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #11205
2020-11-17 12:19:46 -08:00
Brian Behlendorf d02fc15ba1 Linux: Fix ZFS_ENTER/ZFS_EXIT/ZFS_VERFY_ZP usage
The ZFS_ENTER/ZFS_EXIT/ZFS_VERFY_ZP macros should not be used
in the Linux zpl_*.c source files.  They return a positive error
value which is correct for the common code, but not for the Linux
specific kernel code which expects a negative return value.  The
ZPL_ENTER/ZPL_EXIT/ZPL_VERFY_ZP macros should be used instead.

Furthermore, the ZPL_EXIT macro has been updated to not call the
zfs_exit_fs() function.  This prevents a possible deadlock which
can occur when a snapshot is automatically unmounted because the
zpl_show_devname() must never wait on in progress automatic
snapshot unmounts.

Reviewed-by: Adam Moss <c@yotes.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11169 
Closes #11201
2020-11-14 10:51:27 -08:00
Matthew Ahrens 435dc4baab Assertion failure when logging large output of channel program
The output of ZFS channel programs is logged on-disk in the zpool
history, and printed by `zpool history -i`.  Channel programs can use
10MB of memory by default, and up to 100MB by using the `zfs program -m`
flag.  Therefore their output can be up to some fraction of 100MB.

In addition to being somewhat wasteful of the limited space reserved for
the pool history (which for large pools is 1GB), in extreme cases this
can result in a failure of `ASSERT(length <= DMU_MAX_ACCESS);` in
`dmu_buf_hold_array_by_dnode()`.

This commit limits the output size that will be logged to 1MB.  Larger
outputs will not be logged, instead a entry will be logged indicating
the size of the omitted output.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #11194
2020-11-14 10:51:21 -08:00
Matthew Ahrens 4a87c280dc Channel program may spuriously fail with "memory limit exhausted"
ZFS channel programs (invoked by `zfs program`) are executed in a LUA
sandbox with a limit on the amount of memory they can consume.  The
limit is 10MB by default, and can be raised to 100MB with the `-m` flag.
If the memory limit is exceeded, the LUA program exits and the command
fails with a message like `Channel program execution failed: Memory
limit exhausted.`

The LUA sandbox allocates memory with `vmem_alloc(KM_NOSLEEP)`, which
will fail if the requested memory is not immediately available.  In this
case, the program fails with the same message, `Memory limit exhausted`.
However, in this case the specified memory limit has not been reached,
and the memory may only be temporarily unavailable.

This commit changes the LUA memory allocator `zcp_lua_alloc()` to use
`vmem_alloc(KM_SLEEP)`, so that we won't spuriously fail when memory is
temporarily low.  Instead, we rely on the system to be able to free up
memory (e.g. by evicting from the ARC), and we assume that even at the
highest memory limit of 100MB, the channel program will not truly
exhaust the system's memory.

External-issue: DLPX-71924
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #11190
2020-11-12 09:02:00 -08:00
Brian Behlendorf d237d9a918 Linux: Fix mount/unmount when dataset name has a space
The custom zpl_show_devname() helper should translate spaces in
to the octal escape sequence \040.  The getmntent(2) function
is aware of this convention and properly translates the escape
character back to a space when reading the fsname.

Without this change the `zfs mount` and `zfs unmount` commands
incorrectly detect when a dataset with a name containing spaces
is mounted.

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11182
Closes #11187
2020-11-12 09:01:55 -08:00
Tony Perkins 2132ae465d Start snapdir_iterate traversals to begin wtih the value of zero.
The microzap hash can sometimes be zero for single digit snapnames.
The zap cursor can then have a serialized value of two (for . and ..),
and skip the first entry in the avl tree for the .zfs/snapshot directory
listing, and therefore does not return all snapshots.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Cedric Berger <cedric@precidata.com>
Signed-off-by: Tony Perkins <tperkins@datto.com>
Closes #11039
2020-11-12 09:01:27 -08:00
Mateusz Guzik 87f01fc158 G/C data_alloc_arena
It is a leftover from illumos always set to NULL and introducing a
spurious difference between zio_buf and zio_data_buf.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11188
2020-11-11 18:46:22 -08:00
Mateusz Guzik 995b80fa3a G/C struct znode -> z_moved
The field is yet another leftover from unsupported zfs_znode_move.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11186
2020-11-11 11:40:15 -08:00
Ryan Moeller cb4d3fb737 FreeBSD: Simplify zvol_geom_open and zvol_cdev_open
We can consolidate the unlocking procedure into one place by starting
with drop_suspend set to B_FALSE and moving the open count check up.

While here, a little code cleanup. Match the out labels between
zvol_geom_open and zvol_cdev_open, and add a missing period in some
comments.

Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11175
2020-11-11 11:07:40 -08:00
Ryan Moeller 4a2e9811e9 FreeBSD: Avoid spurious EINTR in zvol_cdev_open
zvol_first_open can fail with EINTR if spa_namespace_lock is not held
and cannot be taken without waiting.

Apply the same logic that was done for zvol_geom_open to take
spa_namespace_lock if not already held on first open in zvol_cdev_open.

Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11175
2020-11-11 11:07:40 -08:00
Alexander Motin 050dfc5045 Fix dmu_tx_dirty_throttle after arc_c reduction
After initial arc_c was reduced to arc_c_min it became possible that
on datasets with primarycache=metadata or none dirty data make up most
of ARC capacity and easily more than configured 50% of initial arc_c,
that causes forced txg commits by arc_tempreserve_space() and periodic
very long write delays.

This patch makes arc_tempreserve_space() to use arc_c only after ARC
warmed up once and arc_c really means something, but use arc_c_max
before that.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes #11178
2020-11-11 11:03:43 -08:00
Matthew Macy b49118220c Fix dnode refcount tracking
Fix a couple of places where the wrong tag is passed
to dnode_{hold, rele}

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Closes #11184
2020-11-11 11:03:24 -08:00
Mariusz Zaborski 957b4e9fbd FreeBSD: Prevent a NULL reference in zvol_cdev_open
Check if the ZVOL has been written before calling zil_async_to_sync.
The ZIL will be opened on the first write, not earlier.

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Mariusz Zaborski <oshogbo@vexillium.org>
Closes #11152
2020-11-11 11:00:31 -08:00
khng300 ef648fec0e FreeBSD: Prevent NULL pointer dereference of resid
spa_config_load() passes NULL into resid when doing zfs_file_read().
This would trip over when vfs.zfs.autoimport_disable=0.

Sponsored by: The FreeBSD Foundation
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Ka Ho Ng <khng@freebsdfoundation.org>
Closes #11149
2020-11-11 11:00:19 -08:00
Ryan Moeller 62d549d757 FreeBSD: zvol_os: Use SET_ERROR more judiciously
SET_ERROR is useful to trace errors, so use it where the errors occur
rather than factored out to the end of a function.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11146
2020-11-03 09:51:49 -08:00
Coleman Kane a30fed54f4 Linux 5.10 compat: revalidate_disk_size() added
A new function was added named revalidate_disk_size() and the old
revalidate_disk() appears to have been deprecated. As the only ZFS
code that calls this function is zvol_update_volsize, swapping the
old function call out for the new one should be all that is required.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #11085
2020-11-03 09:51:31 -08:00
Coleman Kane e767b1cacc Linux 5.10 compat: check_disk_change() removed
Kernel 5.10 removed check_disk_change() in favor of callers using
the faster bdev_check_media_change() instead, and explicitly forcing
bdev revalidation when they desire that behavior. To preserve prior
behavior, I have wrapped this into a zfs_check_media_change() macro
that calls an inline function for the new API that mimics the old
behavior when check_disk_change() doesn't exist, and just calls
check_disk_change() if it exists.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #11085
2020-11-03 09:51:26 -08:00
Coleman Kane d2090becab Linux 5.10 compat: percpu_ref added data member
Kernel commit 2b0d3d3e4fcfb brought in some changes to the struct
percpu_ref structure that moves most of its fields into a member
struct named "data" of type struct percpu_ref_data. This includes
the "count" member which is updated by vdev_blkg_tryget(), so update
this function to chase the API change, and detect it via configure.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #11085
2020-11-03 09:51:20 -08:00
Sebastian Gottschall f8460e7e62 Optimize locking checks in mempool allocator
Avoid checking the whole array of objects each time by removing the self
organized memory reaping. this can be managed by the global memory reap
callback which is called every 60 seconds. this will reduce the use if
locking operations significant.

Reviewed-by: Kjeld Schouten <kjeld@schouten-lebbing.nl>
Reviewed-by: Mateusz Guzik <mjguzik@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Sebastian Gottschall <s.gottschall@dd-wrt.com>
Closes #11126
2020-11-03 09:51:10 -08:00
Ryan Moeller aaeffd09bf zvol_os: Fix handling of zvol private data
zvol private data is supposed to be nulled by zvol_clear_private before
zvol_free is called as an indicator that the zvol is going away.

Implement zvol_clear_private for volmode=dev.

Assert that zvol_clear_private has been called before zvol_free.

Check that zvol_clear_private has not been called when updating
volsize.  If it has, fail with ENXIO.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11117
2020-10-30 16:06:54 -07:00
Ryan Moeller 0c270bb6c4 zvol_os: Don't leak doi in cdev error path
Make sure to free doi in zvol_create_minor impl when make_dev_s fails.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11117
2020-10-30 16:06:49 -07:00
Ryan Moeller c2c643256c zvol_os: Properly ignore error in volmode lookup
We fall back to a default volmode and continue when looking up a zvol's
volmode property fails.  After this we should set the error to 0 to
ensure we take the success paths in the out section.

While here, make sure we only log that the zvol was created on success.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11117
2020-10-30 16:06:42 -07:00
Ryan Moeller 896d0f0906 zvol_os: Code cleanup in zvol_create_minor_impl
Nonfunctional changes for readability and consistency.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11117
2020-10-30 16:06:36 -07:00
Ryan Moeller ef525e0841 zvol_os: Keep better track of open count in close
zvol_geom_close gets a count of the number of close operations to do.

Make sure we're always using this count to check if this will be the
last close operation performed on the zvol.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11117
2020-10-30 16:06:30 -07:00
Ryan Moeller 00a27515f0 zvol_os: Tidy up asserts
Using more specific assert variants gives better messages on failure.

No functional change.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11117
2020-10-30 16:06:24 -07:00
Mateusz Guzik 52f1ef3b2d zstd: track allocator statistics
Note that this only tracks sizes as requested by the caller.
Actual allocated space will almost always be bigger (e.g., rounded up to
the next power of 2 or page size). Additionally the allocated buffer may
be holding other areas hostage. Nonetheless, this is a starting point
for tracking memory usage in zstd.

Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Ryan Moeller <ryan@ixsystems.com>
Reviewed-by: Kjeld Schouten <kjeld@schouten-lebbing.nl>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11129
2020-10-30 16:06:15 -07:00
Attila Fülöp c6b0680d9b ICP: gcm: Allocate hash subkey table separately
While evaluating other assembler implementations it turns out that
the precomputed hash subkey tables vary in size, from 8*16 bytes
(avx2/avx512) up to 48*16 bytes (avx512-vaes), depending on the
implementation.

To be able to handle the size differences later, allocate
`gcm_Htable` dynamically rather then having a fixed size array, and
adapt consumers.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Attila Fülöp <attila@fueloep.org>
Closes #11102
2020-10-30 16:06:09 -07:00
Attila Fülöp 2c37e1416b Add some missing cfi frame info in aesni-gcm-x86_64.S
While preparing #9749 some .cfi_{start,end}proc directives
were missed. Add the missing ones.

See upstream https://github.com/openssl/openssl/commit/275a048f

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Attila Fülöp <attila@fueloep.org>
Closes #11101
2020-10-30 16:06:00 -07:00
Mateusz Guzik 6e4845aee3 FreeBSD: catch up with 1300124 version bump
- use cache_vop_mkdir
- cache_rename -> cache_vop_rename

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11136
2020-10-30 16:05:18 -07:00
Ryan Moeller 48cf7d674a FreeBSD: Fix 12.2-STABLE after AT_BENEATH MFC
AT_BENEATH was merged to stable/12, where kern_unlinkat takes a
non-const path.  DECONST the path passed to kern_unlinkat in the
case where AT_BENEATH is defined.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11139
2020-10-30 16:05:10 -07:00
Alexander Motin ca54e52122 Yield periodically when rebuilding L2ARC
L2ARC devices of several terabytes filled with 4KB blocks may take 15
minutes to rebuild.  Due to the way L2ARC log reading is implemented
it is quite likely that for all that time rebuild thread will never
sleep.  At least on FreeBSD kernel threads have absolute priority and
can not be preempted by threads with lower priorities.  If some thread
is also bound to that specific CPU it may not get any CPU time for all
the 15 minutes.

Reviewed-by: Cedric Berger <cedric@precidata.com>
Reviewed-by: Ryan Moeller <freqlabs@FreeBSD.org>
Reviewed-by: George Amanakis <gamanakis@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes #11116
2020-10-30 16:04:53 -07:00
Ryan Moeller c3ae9321bf Update references to nonexistent man pages in code
Refer to the correct section or alternative for FreeBSD and Linux.

Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11132
2020-10-30 16:04:41 -07:00
Alexander Motin 2dd2e49cc7 FreeBSD: Remove BIO_ORDERED flag from BIO_FLUSH
ZFS always waits for the write completion before flushing the cache.
That is why it does not require explicit ordering fences around it,
which are pretty difficult to implement for NVMe, since one has no
internal concept of strict request ordering.

This was already removed from FreeBSD once, but got resurrected
by mistake during OpenZFS merge.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes #11130
2020-10-30 16:04:32 -07:00
Mateusz Guzik 7e76d21bc8 Linux: g/c leftover fence in zfs_znode_alloc
The port removed provisions for zfs_znode_move but the cleanup missed
this bit. To quote the original:

[snip]
    list_insert_tail(&zfsvfs->z_all_znodes, zp);
    membar_producer();
    /*
     * Everything else must be valid before assigning z_zfsvfs makes the
     * znode eligible for zfs_znode_move().
     */
    zp->z_zfsvfs = zfsvfs;
[/snip]

In the current code it is immediately followed by unlock which issues
the same fence, thus plays no role in correctness.

Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11115
2020-10-30 16:04:05 -07:00
Mateusz Guzik e579a4ed0f FreeBSD: g/c unused zfs_znode_move support
The allocator does not provide the functionality to begin with.

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11114
2020-10-30 16:03:58 -07:00
Brian Behlendorf 6867d00403 Use known license string for zlua
The Linux kernel MODULE_LICENSE macro only recognizes a handful of
license strings and "MIT" is not one of the them.  Update the macro
to use "Dual MIT/GPL" which is recognized and what the kernel expects
MIT licensed modules to use.

Reviewed-by: George Melikov <mail@gmelikov.ru>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11112
Closes #11113
2020-10-30 16:03:37 -07:00
Ryan Moeller 0dc6fb730f FreeBSD: Skip RAW kstat sysctls by default
These kstats are often expensive to compute so we want to avoid them
unless specifically requested.

The following kstats are affected by this change:

kstat.zfs.${pool}.multihost
kstat.zfs.${pool}.misc.state
kstat.zfs.${pool}.txgs
kstat.zfs.misc.fletcher_4_bench
kstat.zfs.misc.vdev_raidz_bench
kstat.zfs.misc.dbufs
kstat.zfs.misc.dbgmsg

In FreeBSD 13, sysctl(8) has been updated to still list the
names/description/type of skipped sysctls so they are still
discoverable.

Reviewed-by: Allan Jude <allan@klarasystems.com>
Reviewed-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11099
2020-10-30 16:03:22 -07:00
Mateusz Guzik f5bffd3748 FreeBSD: catch up with 1300123 version bump
- removed thread argument from VOP_INACTIVE
- removed cred argument from VOP_VPTOCNP

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11104
2020-10-30 16:03:13 -07:00
Ryan Moeller 73511e3dde Add missing zfs_arc_evict_batch_limit tunable
It's even documented already.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11094
2020-10-30 16:02:24 -07:00
Kyle Evans 4df31aa98c Makefile.bsd: remove directory that no longer exists
This was removed in a reorganization of directories preparing for the
merge of FreeBSD support, 006e9a4088 by mmacy. While llvm is perfectly
happy with the nonexistent -I directory, the gcc6 and gcc9 we can elect
to use as cross-toolchains both trip over it.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Kyle Evans <kevans@FreeBSD.org>
Closes #11077
2020-10-30 15:57:46 -07:00
Matthew Macy aeeada355c FreeBSD: delete unreferenced file
zfs_onexit_os.c was not deleted when it was removed from the build

Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Closes #11079
2020-10-30 15:57:15 -07:00
Mateusz Guzik bd565f3e24 FreeBSD: add missing fplookup_vexec handler to special vop vectors
Otherwise lookup can fail with EOPNOTSUPP or panic.

Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11066
2020-10-16 13:05:43 -07:00
Mateusz Guzik 3c4e580e9a FreeBSD: g/c unused vop vector zfsctl_ops_shares_dir
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11066
2020-10-16 13:05:39 -07:00
Don Brady b3f4436d37 Ignore special vdev ashift for spa ashift min/max
The removal of a vdev in the normal class would fail if there was a
special or deup vdev that had a different ashift than the vdevs in
the normal class.

Moved the initialization of spa_min_ashift / spa_max_ashift from
vdev_open so that it occurs after the vdev allocation bias was
initialized (i.e. after vdev_load).

Caveat -- In order to remove a special/dedup vdev it must have the
same ashift as the normal pool vdevs.  This could perhaps be lifted
in the future (i.e. for the case where there is ample space in any
surviving special class vdevs)

Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Don Brady <don.brady@delphix.com>
Closes #9363
Closes #9364
Closes #11053
2020-10-16 13:05:34 -07:00
Christian Schwarz 05f8be3b49 Fix crash caused by invalid snapshot names in redactnvl
This is a follow up fix for commit 0fdd6106bb.  The VERIFY is
only true when we haven't hit an error code path.  See added
test case for a reproducer.

Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Christian Schwarz <me@cschwarz.com>
Closes #11048
2020-10-16 13:05:28 -07:00
Paul Dagnelie d8091c9294 Fix incorrect deletion order in range_tree_add_impl gap case
After a side-effectful call like add or remove, references to range
segs stored in btrees can no longer be used safely.  We move the
remove call to just before the reinsertion call so that the seg
remains valid for as long as we need it.

Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes #11044
Closes #11056
2020-10-16 13:05:23 -07:00
Mateusz Guzik 05613fa7a3 FreeBSD: fix panic due to tqid overflow
The 32-bit counter eventually wraps to 0 which is a sentinel for invalid
id.

Make it 64-bit on LP64 platforms and 0-check otherwise.

Note: Linux counterpart uses id stored per queue instead of a global.
I did not check going that way is feasible with the goal being the
minimal fix doing the job.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11059
2020-10-16 13:05:18 -07:00
Ryan Moeller 725c9e22ca Cross-platform acltype
The acltype property is currently hidden on FreeBSD and does not
reflect the NFSv4 style ZFS ACLs used on the platform.  This makes it
difficult to observe that a pool imported from FreeBSD on Linux has a
different type of ACL that is being ignored, and vice versa.

Add an nfsv4 acltype and expose the property on FreeBSD.

Make the default acltype nfsv4 on FreeBSD.

Setting acltype to an unhanded style is treated the same as setting
it to off.  The ACLs will not be removed, but they will be ignored.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #10520
2020-10-16 13:05:00 -07:00
Warner Losh fbfc7e843a FreeBSD: make adjustments for the standalone environment
In FreeBSD, there are three compile environments that are supported:
user land, the kernel and the bootloader / standalone. Adjust the
headers to compile in the standalone environment. Limit kernel-only
items from view when _STANDALONE is defined.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Warner Losh <imp@FreeBSD.org>
Closes #10998
2020-10-16 13:04:41 -07:00
Warner Losh faa62966b1 aarch64: Use proper guards for NEON instructions
The zstd code assumes that if you are on aarch64, you have NEON
instructions. This is not necessarily true. In a boot loader, where
you might not have the VFP properly initialized, these instructions
may not be available. It's also an error to include arm_neon.h when
the NEON insturctions aren't enabled. Change the guards for using the
NEON instructions from __aarch64__ to __ARM_NEON which is the standard
symbol for knowing if they are available.

__ARM_NEON is the proper symbol, defined in ARM C Language Extensions
Release 2.1 (https://developer.arm.com/documentation/ihi0053/d/). Some
sources suggest __ARM_NEON__, but that's the obsolete spelling from
prior versions of the standard.

Updated based on zstd pull request https://github.com/facebook/zstd/pull/2356

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Warner Losh <imp@bsdimp.com>
Closes #11055
2020-10-16 13:03:13 -07:00
Mateusz Guzik 7f0b3fa042 FreeBSD: use cache_rename if available
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #11045
2020-10-16 13:03:00 -07:00
Ryan Moeller c71847b77b Expose zfetch_max_idistance tunable
FreeBSD had this value tunable before the switch to the new OpenZFS.
The tunable name has changed, breaking legacy compat.

Restore legacy compat for this tunable, properly expose the tunable
with the new name on all platforms, and document it in
zfs-module-parameters(5).

While here, clean up the documentation for zfetch_max_distance a bit.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #11038
2020-10-16 13:02:39 -07:00
Christian Schwarz 5c6d3c21b1 zil_parse: make callback parameters const
Code cleanup, a follow up commit to 4d55ea81.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Co-authored-by: Ryan Moeller <ryan@freqlabs.com>
Signed-off-by: Christian Schwarz <me@cschwarz.com>
Closes #11020
2020-10-16 13:01:53 -07:00
Ryan Moeller 5e7198b873 Linux: Initialize zp in zfs_setattr_dir
The value of zp is used without having been initialized under some
conditions.  Initialize the pointer to NULL.

Add a regression test case using chown in acl/posix.  However, this is
not enough because the setup sets xattr=sa, which means zfs_setattr_dir
will not be called.  Create a second group of acl tests in acl/posix-sa
duplicating the acl/posix tests with symlinks, and remove xattr=sa from
the original acl/posix tests.  This provides more coverage for the
default xattr=on code.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes #10043
Closes #11025
2020-10-16 13:01:29 -07:00
Brian Behlendorf 46c71074ca Replace ZFS on Linux references with OpenZFS
This change updates the documentation to refer to the project
as OpenZFS instead ZFS on Linux.  Web links have been updated
to refer to https://github.com/openzfs/zfs.  The extraneous
zfsonlinux.org web links in the ZED and SPL sources have been
dropped.

Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #11007
2020-10-16 13:01:24 -07:00
Jacob Adams 35ba2ca5b7 Fix Linux modules uninstall
A missing semicolon between kmoddir variable declaration and the
uninstall for loop caused modules_uninstall-Linux to fail with:

    Syntax error: "do" unexpected

Reviewed-by: Kjeld Schouten <kjeld@schouten-lebbing.nl>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Jacob Adams <jacob@tookmund.com>
Closes #11032
2020-10-16 13:01:14 -07:00
Chuck Tuffli 0df5b5737c Fix ubsan: shift exponent is too large
When running libzpool with the Undefined Behavior Sanitizer (ubsan)
enabled, a zpool create causes a run-time error:

    module/zfs/vdev_label.c:600:14: runtime error: shift exponent 64 is
    too large for 64-bit type 'long long unsigned int'`

in vdev_config_generate()

Fix is to convert vdev_removal_max_span to its base-2 logarithm, using
highbit64(), and then compare the "shifts".

Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Chuck Tuffli <ctuffli@gmail.com>
Closes #9744
Closes #11024
2020-10-16 13:00:44 -07:00
Ryan Moeller 25e44a17ff Make dbufstat work on FreeBSD
With procfs_list kstats implemented for FreeBSD, dbufs are now exposed
as kstat.zfs.misc.dbufs.

On FreeBSD, dbufstats can use the sysctl instead of procfs when no
input file has been given.

Enable the dbufstats tests on FreeBSD.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <freqlabs@FreeBSD.org>
Closes #11008
2020-10-16 13:00:28 -07:00
Ryan Moeller 718d20ed93 FreeBSD: Sort and dedup includes in kmod_core
Code cleanup. Sort includes, remove duplicates, and drop
some extra blank lines in kmod_core.c.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <freqlabs@FreeBSD.org>
Closes #11000
2020-10-16 13:00:23 -07:00