Compare commits

..

42 Commits

Author SHA1 Message Date
Tony Hutter d99134be83 Tag zfs-2.1.14
META file and changelog updated.

Signed-off-by: Tony Hutter <hutter2@llnl.gov>
2023-11-30 11:09:15 -08:00
наб a339bd791a copy-builtin: add hooks with sed/>>
The order in fs/Makefile doesn't matter,
the order in fs/Kconfig is preserved (ext2 is included as the first
thing in the first if BUILD block, and only once), but I don't think it
matters much either, realistically

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes #13316
2023-11-30 11:09:15 -08:00
Jason King 1ca531971f Zpool can start allocating from metaslab before TRIMs have completed
When doing a manual TRIM on a zpool, the metaslab being TRIMmed is
potentially re-enabled before all queued TRIM zios for that metaslab
have completed. Since TRIM zios have the lowest priority, it is 
possible to get into a situation where allocations occur from the 
just re-enabled metaslab and cut ahead of queued TRIMs to the same 
metaslab.  If the ranges overlap, this will cause corruption.

We were able to trigger this pretty consistently with a small single 
top-level vdev zpool (i.e. small number of metaslabs) with heavy 
parallel write activity while performing a manual TRIM against a 
somewhat 'slow' device (so TRIMs took a bit of time to complete). 
With the patch, we've not been able to recreate it since. It was on 
illumos, but inspection of the OpenZFS trim code looks like the 
relevant pieces are largely unchanged and so it appears it would be 
vulnerable to the same issue.

Reviewed-by: Igor Kozhukhov <igor@dilos.org>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Jason King <jking@racktopsystems.com>
Illumos-issue: https://www.illumos.org/issues/15939
Closes #15395
2023-11-29 12:56:23 -08:00
Rob N 77b0c6f040 dnode_is_dirty: check dnode and its data for dirtiness
Over its history this the dirty dnode test has been changed between
checking for a dnodes being on `os_dirty_dnodes` (`dn_dirty_link`) and
`dn_dirty_record`.

  de198f2d9 Fix lseek(SEEK_DATA/SEEK_HOLE) mmap consistency
  2531ce372 Revert "Report holes when there are only metadata changes"
  ec4f9b8f3 Report holes when there are only metadata changes
  454365bba Fix dirty check in dmu_offset_next()
  66aca2473 SEEK_HOLE should not block on txg_wait_synced()

Also illumos/illumos-gate@c543ec060d illumos/illumos-gate@2bcf0248e9

It turns out both are actually required.

In the case of appending data to a newly created file, the dnode proper
is dirtied (at least to change the blocksize) and dirty records are
added.  Thus, a single logical operation is represented by separate
dirty indicators, and must not be separated.

The incorrect dirty check becomes a problem when the first block of a
file is being appended to while another process is calling lseek to skip
holes. There is a small window where the dnode part is undirtied while
there are still dirty records. In this case, `lseek(fd, 0, SEEK_DATA)`
would not know that the file is dirty, and would go to
`dnode_next_offset()`. Since the object has no data blocks yet, it
returns `ESRCH`, indicating no data found, which results in `ENXIO`
being returned to `lseek()`'s caller.

Since coreutils 9.2, `cp` performs sparse copies by default, that is, it
uses `SEEK_DATA` and `SEEK_HOLE` against the source file and attempts to
replicate the holes in the target. When it hits the bug, its initial
search for data fails, and it goes on to call `fallocate()` to create a
hole over the entire destination file.

This has come up more recently as users upgrade their systems, getting
OpenZFS 2.2 as well as a newer coreutils. However, this problem has been
reproduced against 2.1, as well as on FreeBSD 13 and 14.

This change simply updates the dirty check to check both types of dirty.
If there's anything dirty at all, we immediately go to the "wait for
sync" stage, It doesn't really matter after that; both changes are on
disk, so the dirty fields should be correct.

Sponsored-by: Klara, Inc.
Sponsored-by: Wasabi Technology, Inc.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Rich Ercolani <rincebrain@gmail.com>
Signed-off-by: Rob Norris <rob.norris@klarasystems.com>
Closes #15571
Closes #15526
2023-11-28 09:16:49 -08:00
Tony Hutter eb62221ff0 Tag zfs-2.1.13
META file and changelog updated.

Signed-off-by: Tony Hutter <hutter2@llnl.gov>
2023-09-21 11:07:21 -07:00
наб a3169da877 check-zstd-symbols: also ignore __pfx_ symbols
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=b341b20d648bb7e9a3307c33163e7399f0913e66

Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes #15282
Closes #15284
2023-09-20 13:26:26 -07:00
Richard Yao a93c30a2f2 Cleanup: Replace oldstyle struct hack with C99 flexible array members
The Linux 5.16.14 kernel's coccicheck caught this. The semantic
patch that caught it was:

./scripts/coccinelle/misc/flexible_array.cocci

However, unlike the cases where the GNU zero length array extension had
been used, coccicheck would not suggest patches for the older style
single member arrays. That was good because blindly changing them would
break size calculations in most cases.

Therefore, this required care to make sure that we did not break size
calculations. In the case of `indirect_split_t`, we use
`offsetof(indirect_split_t, is_child[is->is_children])` to calculate
size. This might be subtly wrong according to an old mailing list
thread:

https://inbox.sourceware.org/gcc-prs/20021226123454.27019.qmail@sources.redhat.com/T/

That is because the C99 specification should consider the flexible array
members to start at the end of a structure, but compilers prefer to put
padding at the end. A suggestion was made to allow compilers to allocate
padding after the VLA like compilers already did:

http://std.dkuug.dk/JTC1/SC22/WG14/www/docs/n983.htm

However, upon thinking about it, whether or not we allocate end of
structure padding does not matter, so using offsetof() to calculate the
size of the structure is fine, so long as we do not mix it with sizeof()
on structures with no array members.

In the case that we mix them and padding causes offsetof(struct_t,
vla_member[0]) to differ from sizeof(struct_t), we would be doing unsafe
operations if we underallocate via `offsetof()` and then overcopy via
sizeof().

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Richard Yao <richard.yao@alumni.stonybrook.edu>
Closes #14372
2023-09-20 10:10:41 -07:00
Tony Hutter ae0d644269 Linux 6.5 compat: META (#15265)
Update the META file to reflect compatibility with the 6.5
kernel.

Signed-off-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
2023-09-12 12:52:12 -07:00
Serapheim Dimitropoulos c8a3c74556 checkstyle: fix action failures
Reviewed-by: Don Brady <dev.fs.zfs@gmail.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Closes #15220
2023-09-12 09:22:50 -07:00
Brian Behlendorf a449d8bab0 zed: update zed.d/statechange-slot_off.sh
The statechange-slot_off.sh zedlet which was added in #15200
needed to be installed so it's included by the packages.

Additional testing has also shown that multiple retries are
often needed for the script to operate reliably.

Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #15210
2023-09-12 09:22:50 -07:00
Tony Hutter a4f82db53d zed: Add zedlet to power off slot when drive is faulted
If ZED_POWER_OFF_ENCLOUSRE_SLOT_ON_FAULT is enabled in zed.rc, then
power off the drive's slot in the enclosure if it becomes FAULTED.
This can help silence misbehaving drives.  This assumes your drive
enclosure fully supports slot power control via sysfs.

Reviewed-by: @AllKind
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tony Hutter <hutter2@llnl.gov>
Closes #15200
2023-09-12 09:22:50 -07:00
Coleman Kane 92fc218205 intptr_t definition is canonically signed
Make the version here match that elsewhere in the kernel and system
headers.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15058
2023-09-11 16:39:55 -07:00
Andrea Righi adf428cbef Linux 6.5 compat: spl: properly unregister sysctl entries
When register_sysctl_table() is unavailable we fail to properly
unregister sysctl entries under "kernel/spl".

This leads to errors like the following when spl is unloaded/reloaded,
making impossible to properly reload the spl module:

[  746.995704] sysctl duplicate entry: /kernel/spl/kmem/slab_kvmem_total

Fix by cleaning up all the sub-entries inside "kernel/spl" when the
spl module is unloaded.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
Closes #15239
2023-09-11 16:34:23 -07:00
Andrea Righi cb28c0b770 Linux 6.5 compat: safe cleanup in spl_proc_fini()
If we fail to create a proc entry in spl_proc_init() we may end up
calling unregister_sysctl_table() twice: one in the failure path of
spl_proc_init() and another time during spl_proc_fini().

Avoid the double call to unregister_sysctl_table() and while at it
refactor the code a bit to reduce code duplication.

This was accidentally introduced when the spl code was
updated for Linux 6.5 compatibility.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ameer Hamza <ahamza@ixsystems.com>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
Closes #15234 
Closes #15235
2023-09-11 16:34:17 -07:00
Coleman Kane c74a17a498 Linux 6.5 compat: Use copy_splice_read instead of filemap_splice_read
Using the filemap_splice_read function for the splice_read handler was
leading to occasional data corruption under certain circumstances. Favor
using copy_splice_read instead, which does not demonstrate the same
erroneous behavior under the tested failure cases.

Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15164
2023-09-11 16:34:12 -07:00
Coleman Kane 9b7f7f02e9 Linux 6.5 compat: replace generic_file_splice_read with filemap_splice_read
The generic_file_splice_read function was removed in Linux 6.5 in favor
of filemap_splice_read. Add an autoconf test for filemap_splice_read and
use it if it is found as the handler for .splice_read in the
file_operations struct. Additionally, ITER_PIPE was removed in 6.5. This
change removes the ITER_* macros that OpenZFS doesn't use from being
tested in config/kernel-vfs-iov_iter.m4. The removal of ITER_PIPE was
causing the test to fail, which also affected the code responsible for
setting the .splice_read handler, above. That behavior caused run-time
panics on Linux 6.5.

Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15155
2023-09-11 16:34:01 -07:00
Coleman Kane cb115edfc6 Linux 6.5 compat: register_sysctl_table removed
Additionally, the .child element of ctl_table has been removed in 6.5.
This change adds a new test for the pre-6.5 register_sysctl_table()
function, and uses the old code in that case. If it isn't found, then
the parentage entries in the tables are removed, and the register_sysctl
call is provided the paths of "kernel/spl", "kernel/spl/kmem", and
"kernel/spl/kstat" directly, to populate each subdirectory over three
calls, as is the new API.

Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15138
2023-09-11 16:33:55 -07:00
Brian Atkinson 0ee7a08627 Revert "Linux 6.5 compat: register_sysctl_table removed"
This reverts commit b35374fd64 as there
are error messages when loading the SPL module. Errors seemed to be tied
to duplicate a duplicate entry.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Brian Atkinson <batkinson@lanl.gov>
Closes #15134
2023-09-11 16:33:43 -07:00
Coleman Kane 5ee79af41f Linux 4.20 compat: wrapper function for iov_iter type access
An iov_iter_type() function to access the "type" member of the struct
iov_iter was added at one point. Move the conditional logic to decide
which method to use for accessing it into a macro and simplify the
zpl_uio_init code.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15100
2023-09-11 16:20:50 -07:00
Coleman Kane feb0fa6b38 Linux 6.4 compat: iter_iov() function now used to get old iov member
The iov_iter->iov member is now iov_iter->__iov and must be accessed via
the accessor function iter_iov(). Create a wrapper that is conditionally
compiled to use the access method appropriate for the target kernel
version.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15100
2023-09-11 16:20:42 -07:00
Coleman Kane 7c0618bdb7 Linux 6.5 compat: blkdev changes
Multiple changes to the blkdev API were introduced in Linux 6.5. This
includes passing (void* holder) to blkdev_put, adding a new
blk_holder_ops* arg to blkdev_get_by_path, adding a new blk_mode_t type
that replaces uses of fmode_t, and removing an argument from the release
handler on block_device_operations that we weren't using. The open
function definition has also changed to take gendisk* and blk_mode_t, so
update it accordingly, too.

Implement local wrappers for blkdev_get_by_path() and
vdev_blkdev_put() so that the in-line calls are cleaner, and place the
conditionally-compiled implementation details inside of both of these
local wrappers. Both calls are exclusively used within vdev_disk.c, at
this time.

Add blk_mode_is_open_write() to test FMODE_WRITE / BLK_OPEN_WRITE
The wrapper function is now used for testing using the appropriate
method for the kernel, whether the open mode is writable or not.

Emphasize fmode_t arg in zvol_release is not used

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15099
2023-09-11 16:20:26 -07:00
Coleman Kane 3ff9e4f8ac Linux 6.5 compat: use disk_check_media_change when it exists
When disk_check_media_change() exists, then define
zfs_check_media_change() to simply call disk_check_media_change() on
the bd_disk member of its argument. Since disk_check_media_change()
is newer than when revalidate_disk was present in bops, we should
be able to safely do this via a macro, instead of recreating a new
implementation of the inline function that forces revalidation.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15101
2023-09-11 15:18:13 -07:00
Coleman Kane 211868b5d0 Linux 6.5 compat: register_sysctl_table removed
Additionally, the .child element of ctl_table has been removed in 6.5.
This change adds a new test for the pre-6.5 register_sysctl_table()
function, and uses the old code in that case. If it isn't found, then
the parentage entries in the tables are removed, and the register_sysctl
call is provided the paths of "kernel/spl", "kernel/spl/kmem", and
"kernel/spl/kstat" directly, to populate each subdirectory over three
calls, as is the new API.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15098
2023-09-11 15:18:05 -07:00
Coleman Kane e45ca514e8 Linux 6.5 compat: disk_check_media_change() was added
The disk_check_media_change() function was added which replaces
bdev_check_media_change.  This change was introduced in 6.5rc1
444aa2c58cb3b6cfe3b7cc7db6c294d73393a894 and the new function takes a
gendisk* as its argument, no longer a block_device*. Thus, bdev->bd_disk
is now used to pass the expected data.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15060
2023-09-11 15:17:57 -07:00
Coleman Kane f52d9b6f42 Linux 6.5 compat: BLK_STS_NEXUS renamed to BLK_STS_RESV_CONFLICT
This change was introduced in Linux commit
7ba150834b840f6f5cdd07ca69a4ccf39df59a66

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Coleman Kane <ckane@colemankane.org>
Closes #15059
2023-09-11 15:17:47 -07:00
Brian Behlendorf 837e426c1f Linux: Never sleep in kmem_cache_alloc(..., KM_NOSLEEP) (#14926)
When a kmem cache is exhausted and needs to be expanded a new
slab is allocated.  KM_SLEEP callers can block and wait for the
allocation, but KM_NOSLEEP callers were incorrectly allowed to
block as well.

Resolve this by attempting an emergency allocation as a best
effort.  This may fail but that's fine since any KM_NOSLEEP
consumer is required to handle an allocation failure.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Adam Moss <c@yotes.com>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Reviewed-by: Richard Yao <richard.yao@alumni.stonybrook.edu>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
2023-09-11 15:17:41 -07:00
Rich Ercolani 426d07d64c quick fix for lingering snapdir unmount problems
Unfortunately, even after e79b6807, I still, much more rarely,
tripped asserts when playing with many ctldir mounts at once.

Since this appears to happen if we dispatched twice too fast, just
ignore it. We don't actually need to do anything if someone already
started doing it for us.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Rich Ercolani <rincebrain@gmail.com>
Closes #14462
2023-08-24 17:16:05 -07:00
Rich Ercolani 692f78045e Workaround issue cleaning up automounted snapshots on Linux
On Linux, sometimes, when ZFS goes to unmount an automounted snap,
it fails a VERIFY check on debug builds, because taskq_cancel_id
returned ENOENT after not finding the taskq it was trying to cancel.

This presumably happens when it already died for some reason; in this
case, we don't really mind it already being dead, since we're just
going to dispatch a new task to unmount it right after.

So we just ignore it if we get back ENOENT trying to cancel here,
retry a couple times if we get back the only other possible condition
(EBUSY), and log to dbgmsg if we got anything but ENOENT or success.

(We also add some locking around taskqid, to avoid one or two cases
of two instances of trying to cancel something at once.)

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Tony Nguyen <tony.nguyen@delphix.com>
Signed-off-by: Rich Ercolani <rincebrain@gmail.com>
Closes #11632
Closes #12670
2023-08-24 17:16:05 -07:00
Ryan Lahfa 245850b66c Linux 6.4 compat: META
Update the META file to reflect compatibility with the 6.4 kernel.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Lahfa <ryan@lahfa.xyz>
Closes #15125
2023-08-02 08:51:43 -07:00
Brian Behlendorf 1abf68b7ae zed: Reduce log noise for large JBODs
For large JBODs the log message "zfs_iter_vdev: no match" can
account for the bulk of the log messages (over 70%).  Since this
message is purely informational and not that useful we remove it.

Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #15086
Closes #15094
2023-07-26 08:46:22 -07:00
Tony Hutter 846a4cc089 zed: Fix zed ASSERT on slot power cycle
We would see zed assert on one of our systems if we powered off a
slot.  Further examination showed zfs_retire_recv() was reporting
a GUID of 0, which in turn would return a NULL nvlist.  Add
in a check for a zero GUID.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tony Hutter <hutter2@llnl.gov>
Closes #15084
2023-07-24 16:43:04 -07:00
Alexander Motin e01e3a4e12 Fix raw receive with different indirect block size.
Unlike regular receive, raw receive require destination to have the
same block structure as the source.  In case of dnode reclaim this
triggers two special cases, requiring special handling:
 - If dn_nlevels == 1, we can change the ibs, but dnode_set_blksz()
should not dirty the data buffer if block size does not change, or
durign receive dbuf_dirty_lightweight() will trigger assertion.
 - If dn_nlevels > 1, we just can't change the ibs, dnode_set_blksz()
would fail and receive_object would trigger assertion, so we should
destroy and recreate the dnode from scratch.

Reviewed-by: Paul Dagnelie <pcd@delphix.com>
Signed-off-by:	Alexander Motin <mav@FreeBSD.org>
Sponsored by:	iXsystems, Inc.
Closes #15039
2023-07-20 08:59:14 -07:00
Alexander Motin 6e79bcd356 Fix remount when setting multiple properties.
The previous code was checking zfs_is_namespace_prop() only for the
last property on the list.  If one was not "namespace", then remount
wasn't called.  To fix that move zfs_is_namespace_prop() inside the
loop and remount if at least one of properties was "namespace".

Reviewed-by: Umer Saleem <usaleem@ixsystems.com>
Reviewed-by: Ameer Hamza <ahamza@ixsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by:	Alexander Motin <mav@FreeBSD.org>
Sponsored by:	iXsystems, Inc.
Closes #15000
2023-07-13 08:48:05 -07:00
George Amanakis ef2156c66e Shorten arcstat_quiescence sleep time
With the latest L2ARC fixes, 2 seconds is too long to wait for
quiescence of arcstats like l2_size. Shorten this interval to avoid
having the persistent L2ARC tests in ZTS prematurely terminated.

Signed-off-by: George Amanakis <gamanakis@gmail.com>
2023-06-26 13:59:36 -07:00
George Amanakis f28cd347c4 Store the L2ARC device ashift in the vdev label
If this is not done, and the pool has an ashift other than the default
(at the moment 9) then the following happens:

1) vdev_alloc() assigns the ashift of the pool to L2ARC device, but
   upon export it is not stored anywhere
2) at the first import, vdev_open() sees an vdev_ashift() of 0 and
   assigns the logical_ashift, which is 9
3) reading the contents of L2ARC, including the header fails
4) L2ARC buffers are not restored in ARC.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #14313 
Closes #14963
2023-06-26 13:59:36 -07:00
George Amanakis c12b5829e1 Fix the L2ARC write size calculating logic (2)
While commit bcd5321 adjusts the write size based on the size of the log
block, this happens after comparing the unadjusted write size to the
evicted (target) size.

In this case l2ad_hand will exceed l2ad_evict and violate an assertion
at the end of l2arc_write_buffers().

Fix this by adding the max log block size to the allocated size of the
buffer to be committed before comparing the result to the target
size.

Also reset the l2arc_trim_ahead ZFS module variable when the adjusted
write size exceeds the size of the L2ARC device.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #14936
Closes #14954
2023-06-26 13:59:36 -07:00
George Amanakis 425f7895dd Fix the L2ARC write size calculating logic
l2arc_write_size() should return the write size after adjusting for trim
and overhead of the L2ARC log blocks. Also take into account the
allocated size of log blocks when deciding when to stop writing buffers
to L2ARC.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #14939
2023-06-26 13:59:36 -07:00
George Amanakis d91778e31f Remove duplicate code in l2arc_evict()
l2arc_evict() performs the adjustment of the size of buffers to be
written on L2ARC unnecessarily. l2arc_write_size() is called right
before l2arc_evict() and performs those adjustments.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #14828
2023-06-26 13:59:36 -07:00
Alexander Motin cb549c7425 Fix memory leak in zil_parse().
482da24e2 missed arc_buf_destroy() calls on log parse errors, possibly
leaking up to 128KB of memory per dataset during ZIL replay.

Signed-off-by:	Alexander Motin <mav@FreeBSD.org>
Sponsored by:	iXsystems, Inc.
2023-06-26 13:58:46 -07:00
наб f9a2d94c95 contrib/bash_completion.d: fix error spew from __zfs_match_snapshot()
Given:
  /sbin/zfs list filling/a-zvol<TAB> -o space,refratio
The rest of the cmdline gets vored by:
  /sbin/zfs list filling/a-zvolcannot open 'filling/a-zvol':
  operation not applicable to datasets of this type

With -x (fragment):
  + COMPREPLY=($(compgen -W "$(__zfs_match_snapshot)" -- "$cur"))
  +++ __zfs_match_snapshot
  +++ local base_dataset=filling/dziadtop-nowe-duchy
  +++ [[ filling/dziadtop-nowe-duchy != filling/dziadtop-nowe-duchy ]]
  +++ [[ filling/dziadtop-nowe-duchy != '' ]]
  +++ __zfs_list_datasets filling/dziadtop-nowe-duchy
  +++ /sbin/zfs list -H -o name -s name -t filesystem
                     -r filling/dziadtop-nowe-duchy
  +++ tail -n +2
  cannot open 'filling/dziadtop-nowe-duchy':
  operation not applicable to datasets of this type
  +++ echo filling/dziadtop-nowe-duchy
  +++ echo filling/dziadtop-nowe-duchy@
  ++ compgen -W 'filling/dziadtop-nowe-duchy

This properly completes with:
  $ /sbin/zfs list filling/a-zvol<TAB> -o space,refratio
  filling/a-zvol   filling/a-zvol@
  $ /sbin/zfs list filling/a-zvol<cursor> -o space,refratio

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Backport-of: 5ece420f03
Closes #12820
2023-06-17 19:52:46 -07:00
наб 6af79c92dd tests: zts-report: issue numbers are numbers
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: John Kennedy <john.kennedy@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes #13259
2023-06-13 11:36:18 -07:00
Jeremy Visser c5bbd80cb9 zfs-dkms rpm: simplify scriptlets, fix uninstall
Two problems led to unexpected behaviour of the scriptlets:

1) Newer DKMS versions change the formatting of "dkms status":

   (old) zfs, 2.1.2, 5.14.10-300.fc35.x86_64, x86_64: installed
   (new) zfs/2.1.2, 5.14.10-300.fc35.x86_64, x86_64: installed

   Which broke a conditional determining whether to uninstall.

2) zfs_config.h not packaged properly, but was attempted to be read
   in the %preun scriptlet:

   CONFIG_H="/var/lib/dkms/zfs/2.1.2/*/*/zfs_config.h"

   Which broke the uninstallation of the module, which left behind a
   dangling symlink, which broke DKMS entirely with this error:

     Error! Could not locate dkms.conf file.
     File: /var/lib/dkms/zfs/2.1.1/source/dkms.conf does not exist.

This change attempts to simplify life by:

*  Avoiding parsing anything (less prone to future breakage)
*  Uses %posttrans instead of %post for module installation, because
   %post happens before %preun, while %posttrans happens afterwards
*  Unconditionally reinstall module on upgrade, which is less
   efficient but the trade-off is that it's more reliable

Alternative approaches could involve fixing the existing parsing bugs
or improving the logic, but this comes at the cost of complexity and
possible future bugs.

Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Jeremy Visser <jeremyvisser@google.com>
Closes #10463
Closes #13182
2023-06-09 10:14:42 -07:00
43 changed files with 720 additions and 251 deletions
+2 -2
View File
@@ -1,10 +1,10 @@
Meta: 1
Name: zfs
Branch: 1.0
Version: 2.1.12
Version: 2.1.14
Release: 1
Release-Tags: relext
License: CDDL
Author: OpenZFS
Linux-Maximum: 6.3
Linux-Maximum: 6.5
Linux-Minimum: 3.10
-2
View File
@@ -597,8 +597,6 @@ zfs_iter_vdev(zpool_handle_t *zhp, nvlist_t *nvl, void *data)
*/
if (nvlist_lookup_string(nvl, dp->dd_prop, &path) != 0 ||
strcmp(dp->dd_compare, path) != 0) {
zed_log_msg(LOG_INFO, " %s: no match (%s != vdev %s)",
__func__, dp->dd_compare, path);
return;
}
if (dp->dd_new_vdev_guid != 0 && dp->dd_new_vdev_guid != guid) {
+5
View File
@@ -415,6 +415,11 @@ zfs_retire_recv(fmd_hdl_t *hdl, fmd_event_t *ep, nvlist_t *nvl,
FM_EREPORT_PAYLOAD_ZFS_VDEV_GUID, &vdev_guid) != 0)
return;
if (vdev_guid == 0) {
fmd_hdl_debug(hdl, "Got a zero GUID");
return;
}
if (spare) {
int nspares = find_and_remove_spares(zhdl, vdev_guid);
fmd_hdl_debug(hdl, "%d spares removed", nspares);
+2
View File
@@ -21,6 +21,7 @@ dist_zedexec_SCRIPTS = \
scrub_finish-notify.sh \
statechange-led.sh \
statechange-notify.sh \
statechange-slot_off.sh \
vdev_clear-led.sh \
vdev_attach-led.sh \
pool_import-led.sh \
@@ -39,6 +40,7 @@ zedconfdefaults = \
scrub_finish-notify.sh \
statechange-led.sh \
statechange-notify.sh \
statechange-slot_off.sh \
vdev_clear-led.sh \
vdev_attach-led.sh \
pool_import-led.sh \
+64
View File
@@ -0,0 +1,64 @@
#!/bin/sh
# shellcheck disable=SC3014,SC2154,SC2086,SC2034
#
# Turn off disk's enclosure slot if it becomes FAULTED.
#
# Bad SCSI disks can often "disappear and reappear" causing all sorts of chaos
# as they flip between FAULTED and ONLINE. If
# ZED_POWER_OFF_ENCLOUSRE_SLOT_ON_FAULT is set in zed.rc, and the disk gets
# FAULTED, then power down the slot via sysfs:
#
# /sys/class/enclosure/<enclosure>/<slot>/power_status
#
# We assume the user will be responsible for turning the slot back on again.
#
# Note that this script requires that your enclosure be supported by the
# Linux SCSI Enclosure services (SES) driver. The script will do nothing
# if you have no enclosure, or if your enclosure isn't supported.
#
# Exit codes:
# 0: slot successfully powered off
# 1: enclosure not available
# 2: ZED_POWER_OFF_ENCLOUSRE_SLOT_ON_FAULT disabled
# 3: vdev was not FAULTED
# 4: The enclosure sysfs path passed from ZFS does not exist
# 5: Enclosure slot didn't actually turn off after we told it to
[ -f "${ZED_ZEDLET_DIR}/zed.rc" ] && . "${ZED_ZEDLET_DIR}/zed.rc"
. "${ZED_ZEDLET_DIR}/zed-functions.sh"
if [ ! -d /sys/class/enclosure ] ; then
# No JBOD enclosure or NVMe slots
exit 1
fi
if [ "${ZED_POWER_OFF_ENCLOUSRE_SLOT_ON_FAULT}" != "1" ] ; then
exit 2
fi
if [ "$ZEVENT_VDEV_STATE_STR" != "FAULTED" ] ; then
exit 3
fi
if [ ! -f "$ZEVENT_VDEV_ENC_SYSFS_PATH/power_status" ] ; then
exit 4
fi
# Turn off the slot and wait for sysfs to report that the slot is off.
# It can take ~400ms on some enclosures and multiple retries may be needed.
for i in $(seq 1 20) ; do
echo "off" | tee "$ZEVENT_VDEV_ENC_SYSFS_PATH/power_status"
for j in $(seq 1 5) ; do
if [ "$(cat $ZEVENT_VDEV_ENC_SYSFS_PATH/power_status)" == "off" ] ; then
break 2
fi
sleep 0.1
done
done
if [ "$(cat $ZEVENT_VDEV_ENC_SYSFS_PATH/power_status)" != "off" ] ; then
exit 5
fi
zed_log_msg "powered down slot $ZEVENT_VDEV_ENC_SYSFS_PATH for $ZEVENT_VDEV_PATH"
+5
View File
@@ -143,3 +143,8 @@ ZED_SYSLOG_SUBCLASS_EXCLUDE="history_event"
# Disabled by default, 1 to enable and 0 to disable.
#ZED_SYSLOG_DISPLAY_GUIDS=1
##
# Power off the drive's slot in the enclosure if it becomes FAULTED. This can
# help silence misbehaving drives. This assumes your drive enclosure fully
# supports slot power control via sysfs.
#ZED_POWER_OFF_ENCLOUSRE_SLOT_ON_FAULT=1
+135 -3
View File
@@ -16,12 +16,63 @@ AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_GET_BY_PATH], [
])
])
dnl #
dnl # 6.5.x API change,
dnl # blkdev_get_by_path() takes 4 args
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_GET_BY_PATH_4ARG], [
ZFS_LINUX_TEST_SRC([blkdev_get_by_path_4arg], [
#include <linux/fs.h>
#include <linux/blkdev.h>
], [
struct block_device *bdev __attribute__ ((unused)) = NULL;
const char *path = "path";
fmode_t mode = 0;
void *holder = NULL;
struct blk_holder_ops h;
bdev = blkdev_get_by_path(path, mode, holder, &h);
])
])
AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_GET_BY_PATH], [
AC_MSG_CHECKING([whether blkdev_get_by_path() exists])
AC_MSG_CHECKING([whether blkdev_get_by_path() exists and takes 3 args])
ZFS_LINUX_TEST_RESULT([blkdev_get_by_path], [
AC_MSG_RESULT(yes)
], [
ZFS_LINUX_TEST_ERROR([blkdev_get_by_path()])
AC_MSG_RESULT(no)
AC_MSG_CHECKING([whether blkdev_get_by_path() exists and takes 4 args])
ZFS_LINUX_TEST_RESULT([blkdev_get_by_path_4arg], [
AC_DEFINE(HAVE_BLKDEV_GET_BY_PATH_4ARG, 1,
[blkdev_get_by_path() exists and takes 4 args])
AC_MSG_RESULT(yes)
], [
ZFS_LINUX_TEST_ERROR([blkdev_get_by_path()])
])
])
])
dnl #
dnl # 6.5.x API change
dnl # blk_mode_t was added as a type to supercede some places where fmode_t
dnl # is used
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_BLK_MODE_T], [
ZFS_LINUX_TEST_SRC([blk_mode_t], [
#include <linux/fs.h>
#include <linux/blkdev.h>
], [
blk_mode_t m __attribute((unused)) = (blk_mode_t)0;
])
])
AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_BLK_MODE_T], [
AC_MSG_CHECKING([whether blk_mode_t is defined])
ZFS_LINUX_TEST_RESULT([blk_mode_t], [
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_BLK_MODE_T, 1, [blk_mode_t is defined])
], [
AC_MSG_RESULT(no)
])
])
@@ -41,12 +92,35 @@ AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_PUT], [
])
])
dnl #
dnl # 6.5.x API change.
dnl # blkdev_put() takes (void* holder) as arg 2
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_PUT_HOLDER], [
ZFS_LINUX_TEST_SRC([blkdev_put_holder], [
#include <linux/fs.h>
#include <linux/blkdev.h>
], [
struct block_device *bdev = NULL;
void *holder = NULL;
blkdev_put(bdev, holder);
])
])
AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_PUT], [
AC_MSG_CHECKING([whether blkdev_put() exists])
ZFS_LINUX_TEST_RESULT([blkdev_put], [
AC_MSG_RESULT(yes)
], [
ZFS_LINUX_TEST_ERROR([blkdev_put()])
AC_MSG_CHECKING([whether blkdev_put() accepts void* as arg 2])
ZFS_LINUX_TEST_RESULT([blkdev_put_holder], [
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_BLKDEV_PUT_HOLDER, 1,
[blkdev_put() accepts void* as arg 2])
], [
ZFS_LINUX_TEST_ERROR([blkdev_put()])
])
])
])
@@ -103,6 +177,33 @@ AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_CHECK_DISK_CHANGE], [
])
])
dnl #
dnl # 6.5.x API change
dnl # disk_check_media_change() was added
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_DISK_CHECK_MEDIA_CHANGE], [
ZFS_LINUX_TEST_SRC([disk_check_media_change], [
#include <linux/fs.h>
#include <linux/blkdev.h>
], [
struct block_device *bdev = NULL;
bool error;
error = disk_check_media_change(bdev->bd_disk);
])
])
AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_DISK_CHECK_MEDIA_CHANGE], [
AC_MSG_CHECKING([whether disk_check_media_change() exists])
ZFS_LINUX_TEST_RESULT([disk_check_media_change], [
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_DISK_CHECK_MEDIA_CHANGE, 1,
[disk_check_media_change() exists])
], [
AC_MSG_RESULT(no)
])
])
dnl #
dnl # bdev_kobj() is introduced from 5.12
dnl #
@@ -443,9 +544,34 @@ AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_GET_ERESTARTSYS], [
])
])
dnl #
dnl # 6.5.x API change
dnl # BLK_STS_NEXUS replaced with BLK_STS_RESV_CONFLICT
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV_BLK_STS_RESV_CONFLICT], [
ZFS_LINUX_TEST_SRC([blk_sts_resv_conflict], [
#include <linux/blkdev.h>
],[
blk_status_t s __attribute__ ((unused)) = BLK_STS_RESV_CONFLICT;
])
])
AC_DEFUN([ZFS_AC_KERNEL_BLKDEV_BLK_STS_RESV_CONFLICT], [
AC_MSG_CHECKING([whether BLK_STS_RESV_CONFLICT is defined])
ZFS_LINUX_TEST_RESULT([blk_sts_resv_conflict], [
AC_DEFINE(HAVE_BLK_STS_RESV_CONFLICT, 1, [BLK_STS_RESV_CONFLICT is defined])
AC_MSG_RESULT(yes)
], [
AC_MSG_RESULT(no)
])
])
])
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV], [
ZFS_AC_KERNEL_SRC_BLKDEV_GET_BY_PATH
ZFS_AC_KERNEL_SRC_BLKDEV_GET_BY_PATH_4ARG
ZFS_AC_KERNEL_SRC_BLKDEV_PUT
ZFS_AC_KERNEL_SRC_BLKDEV_PUT_HOLDER
ZFS_AC_KERNEL_SRC_BLKDEV_REREAD_PART
ZFS_AC_KERNEL_SRC_BLKDEV_INVALIDATE_BDEV
ZFS_AC_KERNEL_SRC_BLKDEV_LOOKUP_BDEV
@@ -458,6 +584,9 @@ AC_DEFUN([ZFS_AC_KERNEL_SRC_BLKDEV], [
ZFS_AC_KERNEL_SRC_BLKDEV_ISSUE_SECURE_ERASE
ZFS_AC_KERNEL_SRC_BLKDEV_BDEV_KOBJ
ZFS_AC_KERNEL_SRC_BLKDEV_PART_TO_DEV
ZFS_AC_KERNEL_SRC_BLKDEV_DISK_CHECK_MEDIA_CHANGE
ZFS_AC_KERNEL_SRC_BLKDEV_BLK_STS_RESV_CONFLICT
ZFS_AC_KERNEL_SRC_BLKDEV_BLK_MODE_T
])
AC_DEFUN([ZFS_AC_KERNEL_BLKDEV], [
@@ -476,4 +605,7 @@ AC_DEFUN([ZFS_AC_KERNEL_BLKDEV], [
ZFS_AC_KERNEL_BLKDEV_ISSUE_SECURE_ERASE
ZFS_AC_KERNEL_BLKDEV_BDEV_KOBJ
ZFS_AC_KERNEL_BLKDEV_PART_TO_DEV
ZFS_AC_KERNEL_BLKDEV_DISK_CHECK_MEDIA_CHANGE
ZFS_AC_KERNEL_BLKDEV_BLK_STS_RESV_CONFLICT
ZFS_AC_KERNEL_BLKDEV_BLK_MODE_T
])
+33 -2
View File
@@ -49,12 +49,42 @@ AC_DEFUN([ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS_RELEASE_VOID], [
], [], [])
])
dnl #
dnl # 5.9.x API change
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS_RELEASE_1ARG], [
ZFS_LINUX_TEST_SRC([block_device_operations_release_void_1arg], [
#include <linux/blkdev.h>
void blk_release(struct gendisk *g) {
(void) g;
return;
}
static const struct block_device_operations
bops __attribute__ ((unused)) = {
.open = NULL,
.release = blk_release,
.ioctl = NULL,
.compat_ioctl = NULL,
};
], [], [])
])
AC_DEFUN([ZFS_AC_KERNEL_BLOCK_DEVICE_OPERATIONS_RELEASE_VOID], [
AC_MSG_CHECKING([whether bops->release() is void])
AC_MSG_CHECKING([whether bops->release() is void and takes 2 args])
ZFS_LINUX_TEST_RESULT([block_device_operations_release_void], [
AC_MSG_RESULT(yes)
],[
ZFS_LINUX_TEST_ERROR([bops->release()])
AC_MSG_RESULT(no)
AC_MSG_CHECKING([whether bops->release() is void and takes 1 arg])
ZFS_LINUX_TEST_RESULT([block_device_operations_release_void_1arg], [
AC_MSG_RESULT(yes)
AC_DEFINE([HAVE_BLOCK_DEVICE_OPERATIONS_RELEASE_1ARG], [1],
[Define if release() in block_device_operations takes 1 arg])
],[
ZFS_LINUX_TEST_ERROR([bops->release()])
])
])
])
@@ -92,6 +122,7 @@ AC_DEFUN([ZFS_AC_KERNEL_BLOCK_DEVICE_OPERATIONS_REVALIDATE_DISK], [
AC_DEFUN([ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS], [
ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS_CHECK_EVENTS
ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS_RELEASE_VOID
ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS_RELEASE_1ARG
ZFS_AC_KERNEL_SRC_BLOCK_DEVICE_OPERATIONS_REVALIDATE_DISK
])
+25
View File
@@ -0,0 +1,25 @@
AC_DEFUN([ZFS_AC_KERNEL_SRC_COPY_SPLICE_READ], [
dnl #
dnl # Kernel 6.5 - generic_file_splice_read was removed in favor
dnl # of copy_splice_read for the .splice_read member of the
dnl # file_operations struct.
dnl #
ZFS_LINUX_TEST_SRC([has_copy_splice_read], [
#include <linux/fs.h>
struct file_operations fops __attribute__((unused)) = {
.splice_read = copy_splice_read,
};
],[])
])
AC_DEFUN([ZFS_AC_KERNEL_COPY_SPLICE_READ], [
AC_MSG_CHECKING([whether copy_splice_read() exists])
ZFS_LINUX_TEST_RESULT([has_copy_splice_read], [
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_COPY_SPLICE_READ, 1,
[copy_splice_read exists])
],[
AC_MSG_RESULT(no)
])
])
+27
View File
@@ -0,0 +1,27 @@
dnl #
dnl # Linux 6.5 removes register_sysctl_table
dnl #
AC_DEFUN([ZFS_AC_KERNEL_SRC_REGISTER_SYSCTL_TABLE], [
ZFS_LINUX_TEST_SRC([has_register_sysctl_table], [
#include <linux/sysctl.h>
static struct ctl_table dummy_table[] = {
{}
};
],[
struct ctl_table_header *h
__attribute((unused)) = register_sysctl_table(dummy_table);
])
])
AC_DEFUN([ZFS_AC_KERNEL_REGISTER_SYSCTL_TABLE], [
AC_MSG_CHECKING([whether register_sysctl_table exists])
ZFS_LINUX_TEST_RESULT([has_register_sysctl_table], [
AC_MSG_RESULT([yes])
AC_DEFINE(HAVE_REGISTER_SYSCTL_TABLE, 1,
[register_sysctl_table exists])
],[
AC_MSG_RESULT([no])
])
])
+24 -2
View File
@@ -6,8 +6,7 @@ AC_DEFUN([ZFS_AC_KERNEL_SRC_VFS_IOV_ITER], [
#include <linux/fs.h>
#include <linux/uio.h>
],[
int type __attribute__ ((unused)) =
ITER_IOVEC | ITER_KVEC | ITER_BVEC | ITER_PIPE;
int type __attribute__ ((unused)) = ITER_KVEC;
])
ZFS_LINUX_TEST_SRC([iov_iter_advance], [
@@ -93,6 +92,14 @@ AC_DEFUN([ZFS_AC_KERNEL_SRC_VFS_IOV_ITER], [
struct iov_iter iter = { 0 };
__attribute__((unused)) enum iter_type i = iov_iter_type(&iter);
])
ZFS_LINUX_TEST_SRC([iter_iov], [
#include <linux/fs.h>
#include <linux/uio.h>
],[
struct iov_iter iter = { 0 };
__attribute__((unused)) const struct iovec *iov = iter_iov(&iter);
])
])
AC_DEFUN([ZFS_AC_KERNEL_VFS_IOV_ITER], [
@@ -201,4 +208,19 @@ AC_DEFUN([ZFS_AC_KERNEL_VFS_IOV_ITER], [
AC_DEFINE(HAVE_VFS_IOV_ITER, 1,
[All required iov_iter interfaces are available])
])
dnl #
dnl # Kernel 6.5 introduces the iter_iov() function that returns the
dnl # __iov member of an iov_iter*. The iov member was renamed to this
dnl # __iov member, and is intended to be accessed via the helper
dnl # function now.
dnl #
AC_MSG_CHECKING([whether iter_iov() is available])
ZFS_LINUX_TEST_RESULT([iter_iov], [
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_ITER_IOV, 1,
[iter_iov() is available])
],[
AC_MSG_RESULT(no)
])
])
+4
View File
@@ -148,6 +148,8 @@ AC_DEFUN([ZFS_AC_KERNEL_TEST_SRC], [
ZFS_AC_KERNEL_SRC_FILEMAP
ZFS_AC_KERNEL_SRC_WRITEPAGE_T
ZFS_AC_KERNEL_SRC_RECLAIMED
ZFS_AC_KERNEL_SRC_REGISTER_SYSCTL_TABLE
ZFS_AC_KERNEL_SRC_COPY_SPLICE_READ
case "$host_cpu" in
powerpc*)
ZFS_AC_KERNEL_SRC_CPU_HAS_FEATURE
@@ -275,6 +277,8 @@ AC_DEFUN([ZFS_AC_KERNEL_TEST_RESULT], [
ZFS_AC_KERNEL_FILEMAP
ZFS_AC_KERNEL_WRITEPAGE_T
ZFS_AC_KERNEL_RECLAIMED
ZFS_AC_KERNEL_REGISTER_SYSCTL_TABLE
ZFS_AC_KERNEL_COPY_SPLICE_READ
case "$host_cpu" in
powerpc*)
ZFS_AC_KERNEL_CPU_HAS_FEATURE
+1 -1
View File
@@ -69,7 +69,7 @@ __zfs_match_snapshot()
else
if [ "$cur" != "" ] && __zfs_list_datasets "$cur" &> /dev/null
then
$__ZFS_CMD list -H -o name -s name -t filesystem -r "$cur" | tail -n +2
$__ZFS_CMD list -H -o name -s name -t filesystem,volume -r "$cur" | tail -n +2
# We output the base dataset name even though we might be
# completing a command that can only take a snapshot, because it
# prevents bash from considering the completion finished when it
+2 -26
View File
@@ -43,32 +43,8 @@ config ZFS
If unsure, say N.
EOF
add_after()
{
FILE="$1"
MARKER="$2"
NEW="$3"
while IFS='' read -r LINE
do
printf "%s\n" "$LINE"
if [ -n "$MARKER" ] && [ "$LINE" = "$MARKER" ]
then
printf "%s\n" "$NEW"
MARKER=''
if IFS='' read -r LINE
then
[ "$LINE" != "$NEW" ] && printf "%s\n" "$LINE"
fi
fi
done < "$FILE" > "$FILE.new"
mv "$FILE.new" "$FILE"
}
add_after "$KERNEL_DIR/fs/Kconfig" 'if BLOCK' 'source "fs/zfs/Kconfig"'
add_after "$KERNEL_DIR/fs/Makefile" 'endif' 'obj-$(CONFIG_ZFS) += zfs/'
sed -i '/source "fs\/ext2\/Kconfig\"/i\source "fs/zfs/Kconfig"' "$KERNEL_DIR/fs/Kconfig"
echo 'obj-$(CONFIG_ZFS) += zfs/' >> "$KERNEL_DIR/fs/Makefile"
echo "$0: done. now you can build the kernel with ZFS support." >&2
echo "$0: make sure you enable ZFS support (CONFIG_ZFS) before building." >&2
@@ -170,7 +170,11 @@ bi_status_to_errno(blk_status_t status)
return (ENOLINK);
case BLK_STS_TARGET:
return (EREMOTEIO);
#ifdef HAVE_BLK_STS_RESV_CONFLICT
case BLK_STS_RESV_CONFLICT:
#else
case BLK_STS_NEXUS:
#endif
return (EBADE);
case BLK_STS_MEDIUM:
return (ENODATA);
@@ -204,7 +208,11 @@ errno_to_bi_status(int error)
case EREMOTEIO:
return (BLK_STS_TARGET);
case EBADE:
#ifdef HAVE_BLK_STS_RESV_CONFLICT
return (BLK_STS_RESV_CONFLICT);
#else
return (BLK_STS_NEXUS);
#endif
case ENODATA:
return (BLK_STS_MEDIUM);
case EILSEQ:
@@ -326,6 +334,9 @@ zfs_check_media_change(struct block_device *bdev)
return (0);
}
#define vdev_bdev_reread_part(bdev) zfs_check_media_change(bdev)
#elif defined(HAVE_DISK_CHECK_MEDIA_CHANGE)
#define vdev_bdev_reread_part(bdev) disk_check_media_change(bdev->bd_disk)
#define zfs_check_media_change(bdev) disk_check_media_change(bdev->bd_disk)
#else
/*
* This is encountered if check_disk_change() and bdev_check_media_change()
@@ -376,6 +387,12 @@ vdev_lookup_bdev(const char *path, dev_t *dev)
#endif
}
#if defined(HAVE_BLK_MODE_T)
#define blk_mode_is_open_write(flag) ((flag) & BLK_OPEN_WRITE)
#else
#define blk_mode_is_open_write(flag) ((flag) & FMODE_WRITE)
#endif
/*
* Kernels without bio_set_op_attrs use bi_rw for the bio flags.
*/
+1 -1
View File
@@ -38,7 +38,7 @@ typedef unsigned long ulong_t;
typedef unsigned long long u_longlong_t;
typedef long long longlong_t;
typedef unsigned long intptr_t;
typedef long intptr_t;
typedef unsigned long long rlim64_t;
typedef struct task_struct kthread_t;
+12
View File
@@ -146,4 +146,16 @@ zfs_uio_iov_iter_init(zfs_uio_t *uio, struct iov_iter *iter, offset_t offset,
}
#endif
#if defined(HAVE_ITER_IOV)
#define zfs_uio_iter_iov(iter) iter_iov((iter))
#else
#define zfs_uio_iter_iov(iter) (iter)->iov
#endif
#if defined(HAVE_IOV_ITER_TYPE)
#define zfs_uio_iov_iter_type(iter) iov_iter_type((iter))
#else
#define zfs_uio_iov_iter_type(iter) (iter)->type
#endif
#endif /* SPL_UIO_H */
+4 -3
View File
@@ -1762,7 +1762,8 @@ zfs_prop_set_list(zfs_handle_t *zhp, nvlist_t *props)
nvlist_t *nvl;
int nvl_len = 0;
int added_resv = 0;
zfs_prop_t prop = 0;
zfs_prop_t prop;
boolean_t nsprop = B_FALSE;
nvpair_t *elem;
(void) snprintf(errbuf, sizeof (errbuf),
@@ -1809,6 +1810,7 @@ zfs_prop_set_list(zfs_handle_t *zhp, nvlist_t *props)
elem = nvlist_next_nvpair(nvl, elem)) {
prop = zfs_name_to_prop(nvpair_name(elem));
nsprop |= zfs_is_namespace_prop(prop);
assert(cl_idx < nvl_len);
/*
@@ -1909,8 +1911,7 @@ zfs_prop_set_list(zfs_handle_t *zhp, nvlist_t *props)
* if one of the options handled by the generic
* Linux namespace layer has been modified.
*/
if (zfs_is_namespace_prop(prop) &&
zfs_is_mounted(zhp, NULL))
if (nsprop && zfs_is_mounted(zhp, NULL))
ret = zfs_mount(zhp, MNTOPT_REMOUNT, 0);
}
}
+2 -2
View File
@@ -13,10 +13,10 @@
#define sizeCclosure(n) (cast(int, sizeof(CClosure)) + \
cast(int, sizeof(TValue)*((n)-1)))
cast(int, sizeof(TValue)*((n))))
#define sizeLclosure(n) (cast(int, sizeof(LClosure)) + \
cast(int, sizeof(TValue *)*((n)-1)))
cast(int, sizeof(TValue *)*((n))))
LUAI_FUNC Proto *luaF_newproto (lua_State *L);
+2 -2
View File
@@ -514,14 +514,14 @@ typedef struct UpVal {
typedef struct CClosure {
ClosureHeader;
lua_CFunction f;
TValue upvalue[1]; /* list of upvalues */
TValue upvalue[]; /* list of upvalues */
} CClosure;
typedef struct LClosure {
ClosureHeader;
struct Proto *p;
UpVal *upvals[1]; /* list of upvalues */
UpVal *upvals[]; /* list of upvalues */
} LClosure;
+2 -2
View File
@@ -30,7 +30,7 @@ typedef struct zfs_dbgmsg {
list_node_t zdm_node;
time_t zdm_timestamp;
int zdm_size;
char zdm_msg[1]; /* variable length allocation */
char zdm_msg[];
} zfs_dbgmsg_t;
list_t zfs_dbgmsgs;
@@ -159,7 +159,7 @@ __zfs_dbgmsg(char *buf)
DTRACE_PROBE1(zfs__dbgmsg, char *, buf);
size = sizeof (zfs_dbgmsg_t) + strlen(buf);
size = sizeof (zfs_dbgmsg_t) + strlen(buf) + 1;
zdm = kmem_zalloc(size, KM_SLEEP);
zdm->zdm_size = size;
zdm->zdm_timestamp = gethrestime_sec();
+11 -1
View File
@@ -1020,9 +1020,19 @@ spl_cache_grow(spl_kmem_cache_t *skc, int flags, void **obj)
ASSERT0(flags & ~KM_PUBLIC_MASK);
ASSERT(skc->skc_magic == SKC_MAGIC);
ASSERT((skc->skc_flags & KMC_SLAB) == 0);
might_sleep();
*obj = NULL;
/*
* Since we can't sleep attempt an emergency allocation to satisfy
* the request. The only alterative is to fail the allocation but
* it's preferable try. The use of KM_NOSLEEP is expected to be rare.
*/
if (flags & KM_NOSLEEP)
return (spl_emergency_alloc(skc, flags, obj));
might_sleep();
/*
* Before allocating a new slab wait for any reaping to complete and
* then return so the local magazine can be rechecked for new objects.
+56 -21
View File
@@ -46,6 +46,10 @@ static unsigned long table_min = 0;
static unsigned long table_max = ~0;
static struct ctl_table_header *spl_header = NULL;
#ifndef HAVE_REGISTER_SYSCTL_TABLE
static struct ctl_table_header *spl_kmem = NULL;
static struct ctl_table_header *spl_kstat = NULL;
#endif
static struct proc_dir_entry *proc_spl = NULL;
static struct proc_dir_entry *proc_spl_kmem = NULL;
static struct proc_dir_entry *proc_spl_kmem_slab = NULL;
@@ -624,6 +628,7 @@ static struct ctl_table spl_table[] = {
.mode = 0644,
.proc_handler = &proc_dohostid,
},
#ifdef HAVE_REGISTER_SYSCTL_TABLE
{
.procname = "kmem",
.mode = 0555,
@@ -634,9 +639,11 @@ static struct ctl_table spl_table[] = {
.mode = 0555,
.child = spl_kstat_table,
},
#endif
{},
};
#ifdef HAVE_REGISTER_SYSCTL_TABLE
static struct ctl_table spl_dir[] = {
{
.procname = "spl",
@@ -648,21 +655,64 @@ static struct ctl_table spl_dir[] = {
static struct ctl_table spl_root[] = {
{
.procname = "kernel",
.mode = 0555,
.child = spl_dir,
.procname = "kernel",
.mode = 0555,
.child = spl_dir,
},
{}
};
#endif
static void spl_proc_cleanup(void)
{
remove_proc_entry("kstat", proc_spl);
remove_proc_entry("slab", proc_spl_kmem);
remove_proc_entry("kmem", proc_spl);
remove_proc_entry("taskq-all", proc_spl);
remove_proc_entry("taskq", proc_spl);
remove_proc_entry("spl", NULL);
#ifndef HAVE_REGISTER_SYSCTL_TABLE
if (spl_kstat) {
unregister_sysctl_table(spl_kstat);
spl_kstat = NULL;
}
if (spl_kmem) {
unregister_sysctl_table(spl_kmem);
spl_kmem = NULL;
}
#endif
if (spl_header) {
unregister_sysctl_table(spl_header);
spl_header = NULL;
}
}
int
spl_proc_init(void)
{
int rc = 0;
#ifdef HAVE_REGISTER_SYSCTL_TABLE
spl_header = register_sysctl_table(spl_root);
if (spl_header == NULL)
return (-EUNATCH);
#else
spl_header = register_sysctl("kernel/spl", spl_table);
if (spl_header == NULL)
return (-EUNATCH);
spl_kmem = register_sysctl("kernel/spl/kmem", spl_kmem_table);
if (spl_kmem == NULL) {
rc = -EUNATCH;
goto out;
}
spl_kstat = register_sysctl("kernel/spl/kstat", spl_kstat_table);
if (spl_kstat == NULL) {
rc = -EUNATCH;
goto out;
}
#endif
proc_spl = proc_mkdir("spl", NULL);
if (proc_spl == NULL) {
@@ -703,15 +753,8 @@ spl_proc_init(void)
goto out;
}
out:
if (rc) {
remove_proc_entry("kstat", proc_spl);
remove_proc_entry("slab", proc_spl_kmem);
remove_proc_entry("kmem", proc_spl);
remove_proc_entry("taskq-all", proc_spl);
remove_proc_entry("taskq", proc_spl);
remove_proc_entry("spl", NULL);
unregister_sysctl_table(spl_header);
}
if (rc)
spl_proc_cleanup();
return (rc);
}
@@ -719,13 +762,5 @@ out:
void
spl_proc_fini(void)
{
remove_proc_entry("kstat", proc_spl);
remove_proc_entry("slab", proc_spl_kmem);
remove_proc_entry("kmem", proc_spl);
remove_proc_entry("taskq-all", proc_spl);
remove_proc_entry("taskq", proc_spl);
remove_proc_entry("spl", NULL);
ASSERT(spl_header != NULL);
unregister_sysctl_table(spl_header);
spl_proc_cleanup();
}
+57 -8
View File
@@ -74,9 +74,22 @@ typedef struct dio_request {
struct bio *dr_bio[0]; /* Attached bio's */
} dio_request_t;
#ifdef HAVE_BLK_MODE_T
static blk_mode_t
#else
static fmode_t
#endif
vdev_bdev_mode(spa_mode_t spa_mode)
{
#ifdef HAVE_BLK_MODE_T
blk_mode_t mode = 0;
if (spa_mode & SPA_MODE_READ)
mode |= BLK_OPEN_READ;
if (spa_mode & SPA_MODE_WRITE)
mode |= BLK_OPEN_WRITE;
#else
fmode_t mode = 0;
if (spa_mode & SPA_MODE_READ)
@@ -84,6 +97,7 @@ vdev_bdev_mode(spa_mode_t spa_mode)
if (spa_mode & SPA_MODE_WRITE)
mode |= FMODE_WRITE;
#endif
return (mode);
}
@@ -191,12 +205,47 @@ vdev_disk_kobj_evt_post(vdev_t *v)
}
}
#if !defined(HAVE_BLKDEV_GET_BY_PATH_4ARG)
/*
* Define a dummy struct blk_holder_ops for kernel versions
* prior to 6.5.
*/
struct blk_holder_ops {};
#endif
static struct block_device *
vdev_blkdev_get_by_path(const char *path, spa_mode_t mode, void *holder,
const struct blk_holder_ops *hops)
{
#ifdef HAVE_BLKDEV_GET_BY_PATH_4ARG
return (blkdev_get_by_path(path,
vdev_bdev_mode(mode) | BLK_OPEN_EXCL, holder, hops));
#else
return (blkdev_get_by_path(path,
vdev_bdev_mode(mode) | FMODE_EXCL, holder));
#endif
}
static void
vdev_blkdev_put(struct block_device *bdev, spa_mode_t mode, void *holder)
{
#ifdef HAVE_BLKDEV_PUT_HOLDER
return (blkdev_put(bdev, holder));
#else
return (blkdev_put(bdev, vdev_bdev_mode(mode) | FMODE_EXCL));
#endif
}
static int
vdev_disk_open(vdev_t *v, uint64_t *psize, uint64_t *max_psize,
uint64_t *logical_ashift, uint64_t *physical_ashift)
{
struct block_device *bdev;
#ifdef HAVE_BLK_MODE_T
blk_mode_t mode = vdev_bdev_mode(spa_mode(v->vdev_spa));
#else
fmode_t mode = vdev_bdev_mode(spa_mode(v->vdev_spa));
#endif
hrtime_t timeout = MSEC2NSEC(zfs_vdev_open_timeout_ms);
vdev_disk_t *vd;
@@ -246,15 +295,15 @@ vdev_disk_open(vdev_t *v, uint64_t *psize, uint64_t *max_psize,
reread_part = B_TRUE;
}
blkdev_put(bdev, mode | FMODE_EXCL);
vdev_blkdev_put(bdev, mode, zfs_vdev_holder);
}
if (reread_part) {
bdev = blkdev_get_by_path(disk_name, mode | FMODE_EXCL,
zfs_vdev_holder);
bdev = vdev_blkdev_get_by_path(disk_name, mode,
zfs_vdev_holder, NULL);
if (!IS_ERR(bdev)) {
int error = vdev_bdev_reread_part(bdev);
blkdev_put(bdev, mode | FMODE_EXCL);
vdev_blkdev_put(bdev, mode, zfs_vdev_holder);
if (error == 0) {
timeout = MSEC2NSEC(
zfs_vdev_open_timeout_ms * 2);
@@ -299,8 +348,8 @@ vdev_disk_open(vdev_t *v, uint64_t *psize, uint64_t *max_psize,
hrtime_t start = gethrtime();
bdev = ERR_PTR(-ENXIO);
while (IS_ERR(bdev) && ((gethrtime() - start) < timeout)) {
bdev = blkdev_get_by_path(v->vdev_path, mode | FMODE_EXCL,
zfs_vdev_holder);
bdev = vdev_blkdev_get_by_path(v->vdev_path, mode,
zfs_vdev_holder, NULL);
if (unlikely(PTR_ERR(bdev) == -ENOENT)) {
/*
* There is no point of waiting since device is removed
@@ -376,8 +425,8 @@ vdev_disk_close(vdev_t *v)
return;
if (vd->vd_bdev != NULL) {
blkdev_put(vd->vd_bdev,
vdev_bdev_mode(spa_mode(v->vdev_spa)) | FMODE_EXCL);
vdev_blkdev_put(vd->vd_bdev, spa_mode(v->vdev_spa),
zfs_vdev_holder);
}
rw_destroy(&vd->vd_lock);
+33 -3
View File
@@ -118,6 +118,7 @@ typedef struct {
spa_t *se_spa; /* pool spa */
uint64_t se_objsetid; /* snapshot objset id */
struct dentry *se_root_dentry; /* snapshot root dentry */
krwlock_t se_taskqid_lock; /* scheduled unmount taskqid lock */
taskqid_t se_taskqid; /* scheduled unmount taskqid */
avl_node_t se_node_name; /* zfs_snapshots_by_name link */
avl_node_t se_node_objsetid; /* zfs_snapshots_by_objsetid link */
@@ -144,6 +145,7 @@ zfsctl_snapshot_alloc(const char *full_name, const char *full_path, spa_t *spa,
se->se_objsetid = objsetid;
se->se_root_dentry = root_dentry;
se->se_taskqid = TASKQID_INVALID;
rw_init(&se->se_taskqid_lock, NULL, RW_DEFAULT, NULL);
zfs_refcount_create(&se->se_refcount);
@@ -160,6 +162,7 @@ zfsctl_snapshot_free(zfs_snapentry_t *se)
zfs_refcount_destroy(&se->se_refcount);
kmem_strfree(se->se_name);
kmem_strfree(se->se_path);
rw_destroy(se->se_taskqid_lock);
kmem_free(se, sizeof (zfs_snapentry_t));
}
@@ -335,7 +338,9 @@ snapentry_expire(void *data)
return;
}
rw_enter(&se->se_taskqid_lock, RW_WRITER);
se->se_taskqid = TASKQID_INVALID;
rw_exit(&se->se_taskqid_lock);
(void) zfsctl_snapshot_unmount(se->se_name, MNT_EXPIRE);
zfsctl_snapshot_rele(se);
@@ -359,8 +364,18 @@ snapentry_expire(void *data)
static void
zfsctl_snapshot_unmount_cancel(zfs_snapentry_t *se)
{
if (taskq_cancel_id(system_delay_taskq, se->se_taskqid) == 0) {
se->se_taskqid = TASKQID_INVALID;
int err = 0;
rw_enter(&se->se_taskqid_lock, RW_WRITER);
err = taskq_cancel_id(system_delay_taskq, se->se_taskqid);
/*
* if we get ENOENT, the taskq couldn't be found to be
* canceled, so we can just mark it as invalid because
* it's already gone. If we got EBUSY, then we already
* blocked until it was gone _anyway_, so we don't care.
*/
se->se_taskqid = TASKQID_INVALID;
rw_exit(&se->se_taskqid_lock);
if (err == 0) {
zfsctl_snapshot_rele(se);
}
}
@@ -371,14 +386,29 @@ zfsctl_snapshot_unmount_cancel(zfs_snapentry_t *se)
static void
zfsctl_snapshot_unmount_delay_impl(zfs_snapentry_t *se, int delay)
{
ASSERT3S(se->se_taskqid, ==, TASKQID_INVALID);
if (delay <= 0)
return;
zfsctl_snapshot_hold(se);
rw_enter(&se->se_taskqid_lock, RW_WRITER);
/*
* If this condition happens, we managed to:
* - dispatch once
* - want to dispatch _again_ before it returned
*
* So let's just return - if that task fails at unmounting,
* we'll eventually dispatch again, and if it succeeds,
* no problem.
*/
if (se->se_taskqid != TASKQID_INVALID) {
rw_exit(&se->se_taskqid_lock);
zfsctl_snapshot_rele(se);
return;
}
se->se_taskqid = taskq_dispatch_delay(system_delay_taskq,
snapentry_expire, se, TQ_SLEEP, ddi_get_lbolt() + delay * HZ);
rw_exit(&se->se_taskqid_lock);
}
/*
+2 -2
View File
@@ -30,7 +30,7 @@ typedef struct zfs_dbgmsg {
procfs_list_node_t zdm_node;
uint64_t zdm_timestamp;
int zdm_size;
char zdm_msg[1]; /* variable length allocation */
char zdm_msg[]; /* variable length allocation */
} zfs_dbgmsg_t;
procfs_list_t zfs_dbgmsgs;
@@ -134,7 +134,7 @@ __set_error(const char *file, const char *func, int line, int err)
void
__zfs_dbgmsg(char *buf)
{
int size = sizeof (zfs_dbgmsg_t) + strlen(buf);
int size = sizeof (zfs_dbgmsg_t) + strlen(buf) + 1;
zfs_dbgmsg_t *zdm = kmem_zalloc(size, KM_SLEEP);
zdm->zdm_size = size;
zdm->zdm_timestamp = gethrestime_sec();
+1 -1
View File
@@ -198,7 +198,7 @@ zfs_open(struct inode *ip, int mode, int flag, cred_t *cr)
ZFS_VERIFY_ZP(zp);
/* Honor ZFS_APPENDONLY file attribute */
if ((mode & FMODE_WRITE) && (zp->z_pflags & ZFS_APPENDONLY) &&
if (blk_mode_is_open_write(mode) && (zp->z_pflags & ZFS_APPENDONLY) &&
((flag & O_APPEND) == 0)) {
ZFS_EXIT(zfsvfs);
return (SET_ERROR(EPERM));
+1 -1
View File
@@ -40,7 +40,7 @@
static int
zpl_common_open(struct inode *ip, struct file *filp)
{
if (filp->f_mode & FMODE_WRITE)
if (blk_mode_is_open_write(filp->f_mode))
return (-EACCES);
return (generic_file_open(ip, filp));
+7 -8
View File
@@ -294,15 +294,10 @@ zpl_uio_init(zfs_uio_t *uio, struct kiocb *kiocb, struct iov_iter *to,
#if defined(HAVE_VFS_IOV_ITER)
zfs_uio_iov_iter_init(uio, to, pos, count, skip);
#else
#ifdef HAVE_IOV_ITER_TYPE
zfs_uio_iovec_init(uio, to->iov, to->nr_segs, pos,
iov_iter_type(to) & ITER_KVEC ? UIO_SYSSPACE : UIO_USERSPACE,
zfs_uio_iovec_init(uio, zfs_uio_iter_iov(to), to->nr_segs, pos,
zfs_uio_iov_iter_type(to) & ITER_KVEC ?
UIO_SYSSPACE : UIO_USERSPACE,
count, skip);
#else
zfs_uio_iovec_init(uio, to->iov, to->nr_segs, pos,
to->type & ITER_KVEC ? UIO_SYSSPACE : UIO_USERSPACE,
count, skip);
#endif
#endif
}
@@ -1147,7 +1142,11 @@ const struct file_operations zpl_file_operations = {
.read_iter = zpl_iter_read,
.write_iter = zpl_iter_write,
#ifdef HAVE_VFS_IOV_ITER
#ifdef HAVE_COPY_SPLICE_READ
.splice_read = copy_splice_read,
#else
.splice_read = generic_file_splice_read,
#endif
.splice_write = iter_file_splice_write,
#endif
#else
+24 -4
View File
@@ -492,7 +492,11 @@ out:
}
static int
#ifdef HAVE_BLK_MODE_T
zvol_open(struct gendisk *disk, blk_mode_t flag)
#else
zvol_open(struct block_device *bdev, fmode_t flag)
#endif
{
zvol_state_t *zv;
int error = 0;
@@ -507,10 +511,14 @@ retry:
/*
* Obtain a copy of private_data under the zvol_state_lock to make
* sure that either the result of zvol free code path setting
* bdev->bd_disk->private_data to NULL is observed, or zvol_free()
* disk->private_data to NULL is observed, or zvol_os_free()
* is not called on this zv because of the positive zv_open_count.
*/
#ifdef HAVE_BLK_MODE_T
zv = disk->private_data;
#else
zv = bdev->bd_disk->private_data;
#endif
if (zv == NULL) {
rw_exit(&zvol_state_lock);
return (SET_ERROR(-ENXIO));
@@ -590,14 +598,15 @@ retry:
}
}
error = -zvol_first_open(zv, !(flag & FMODE_WRITE));
error = -zvol_first_open(zv, !(blk_mode_is_open_write(flag)));
if (drop_namespace)
mutex_exit(&spa_namespace_lock);
}
if (error == 0) {
if ((flag & FMODE_WRITE) && (zv->zv_flags & ZVOL_RDONLY)) {
if ((blk_mode_is_open_write(flag)) &&
(zv->zv_flags & ZVOL_RDONLY)) {
if (zv->zv_open_count == 0)
zvol_last_close(zv);
@@ -612,14 +621,25 @@ retry:
rw_exit(&zv->zv_suspend_lock);
if (error == 0)
#ifdef HAVE_BLK_MODE_T
disk_check_media_change(disk);
#else
zfs_check_media_change(bdev);
#endif
return (error);
}
static void
zvol_release(struct gendisk *disk, fmode_t mode)
#ifdef HAVE_BLOCK_DEVICE_OPERATIONS_RELEASE_1ARG
zvol_release(struct gendisk *disk)
#else
zvol_release(struct gendisk *disk, fmode_t unused)
#endif
{
#if !defined(HAVE_BLOCK_DEVICE_OPERATIONS_RELEASE_1ARG)
(void) unused;
#endif
zvol_state_t *zv;
boolean_t drop_suspend = B_TRUE;
+45 -28
View File
@@ -946,7 +946,7 @@ static void l2arc_hdr_restore(const l2arc_log_ent_phys_t *le,
l2arc_dev_t *dev);
/* L2ARC persistence write I/O routines. */
static void l2arc_log_blk_commit(l2arc_dev_t *dev, zio_t *pio,
static uint64_t l2arc_log_blk_commit(l2arc_dev_t *dev, zio_t *pio,
l2arc_write_callback_t *cb);
/* L2ARC persistence auxiliary routines. */
@@ -8415,7 +8415,7 @@ l2arc_write_eligible(uint64_t spa_guid, arc_buf_hdr_t *hdr)
static uint64_t
l2arc_write_size(l2arc_dev_t *dev)
{
uint64_t size, dev_size, tsize;
uint64_t size;
/*
* Make sure our globals have meaningful values in case the user
@@ -8432,18 +8432,23 @@ l2arc_write_size(l2arc_dev_t *dev)
if (arc_warm == B_FALSE)
size += l2arc_write_boost;
/* We need to add in the worst case scenario of log block overhead. */
size += l2arc_log_blk_overhead(size, dev);
if (dev->l2ad_vdev->vdev_has_trim && l2arc_trim_ahead > 0) {
/*
* Trim ahead of the write size 64MB or (l2arc_trim_ahead/100)
* times the writesize, whichever is greater.
*/
size += MAX(64 * 1024 * 1024,
(size * l2arc_trim_ahead) / 100);
}
/*
* Make sure the write size does not exceed the size of the cache
* device. This is important in l2arc_evict(), otherwise infinite
* iteration can occur.
*/
dev_size = dev->l2ad_end - dev->l2ad_start;
tsize = size + l2arc_log_blk_overhead(size, dev);
if (dev->l2ad_vdev->vdev_has_trim && l2arc_trim_ahead > 0)
tsize += MAX(64 * 1024 * 1024,
(tsize * l2arc_trim_ahead) / 100);
if (tsize >= dev_size) {
if (size > dev->l2ad_end - dev->l2ad_start) {
cmn_err(CE_NOTE, "l2arc_write_max or l2arc_write_boost "
"plus the overhead of log blocks (persistent L2ARC, "
"%llu bytes) exceeds the size of the cache device "
@@ -8452,8 +8457,19 @@ l2arc_write_size(l2arc_dev_t *dev)
dev->l2ad_vdev->vdev_guid, L2ARC_WRITE_SIZE);
size = l2arc_write_max = l2arc_write_boost = L2ARC_WRITE_SIZE;
if (l2arc_trim_ahead > 1) {
cmn_err(CE_NOTE, "l2arc_trim_ahead set to 1");
l2arc_trim_ahead = 1;
}
if (arc_warm == B_FALSE)
size += l2arc_write_boost;
size += l2arc_log_blk_overhead(size, dev);
if (dev->l2ad_vdev->vdev_has_trim && l2arc_trim_ahead > 0) {
size += MAX(64 * 1024 * 1024,
(size * l2arc_trim_ahead) / 100);
}
}
return (size);
@@ -9074,22 +9090,9 @@ l2arc_evict(l2arc_dev_t *dev, uint64_t distance, boolean_t all)
buflist = &dev->l2ad_buflist;
/*
* We need to add in the worst case scenario of log block overhead.
*/
distance += l2arc_log_blk_overhead(distance, dev);
if (vd->vdev_has_trim && l2arc_trim_ahead > 0) {
/*
* Trim ahead of the write size 64MB or (l2arc_trim_ahead/100)
* times the write size, whichever is greater.
*/
distance += MAX(64 * 1024 * 1024,
(distance * l2arc_trim_ahead) / 100);
}
top:
rerun = B_FALSE;
if (dev->l2ad_hand >= (dev->l2ad_end - distance)) {
if (dev->l2ad_hand + distance > dev->l2ad_end) {
/*
* When there is no space to accommodate upcoming writes,
* evict to the end. Then bump the write and evict hands
@@ -9283,7 +9286,7 @@ out:
*/
ASSERT3U(dev->l2ad_hand + distance, <, dev->l2ad_end);
if (!dev->l2ad_first)
ASSERT3U(dev->l2ad_hand, <, dev->l2ad_evict);
ASSERT3U(dev->l2ad_hand, <=, dev->l2ad_evict);
}
}
@@ -9549,7 +9552,13 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz)
uint64_t asize = vdev_psize_to_asize(dev->l2ad_vdev,
psize);
if ((write_asize + asize) > target_sz) {
/*
* If the allocated size of this buffer plus the max
* size for the pending log block exceeds the evicted
* target size, terminate writing buffers for this run.
*/
if (write_asize + asize +
sizeof (l2arc_log_blk_phys_t) > target_sz) {
full = B_TRUE;
mutex_exit(hash_lock);
break;
@@ -9669,8 +9678,14 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz)
* arcstat_l2_{size,asize} kstats are updated
* internally.
*/
if (l2arc_log_blk_insert(dev, hdr))
l2arc_log_blk_commit(dev, pio, cb);
if (l2arc_log_blk_insert(dev, hdr)) {
/*
* l2ad_hand will be adjusted in
* l2arc_log_blk_commit().
*/
write_asize +=
l2arc_log_blk_commit(dev, pio, cb);
}
zio_nowait(wzio);
}
@@ -10820,7 +10835,7 @@ l2arc_dev_hdr_update(l2arc_dev_t *dev)
* This function allocates some memory to temporarily hold the serialized
* buffer to be written. This is then released in l2arc_write_done.
*/
static void
static uint64_t
l2arc_log_blk_commit(l2arc_dev_t *dev, zio_t *pio, l2arc_write_callback_t *cb)
{
l2arc_log_blk_phys_t *lb = &dev->l2ad_log_blk;
@@ -10933,6 +10948,8 @@ l2arc_log_blk_commit(l2arc_dev_t *dev, zio_t *pio, l2arc_write_callback_t *cb)
dev->l2ad_log_ent_idx = 0;
dev->l2ad_log_blk_payload_asize = 0;
dev->l2ad_log_blk_payload_start = 0;
return (asize);
}
/*
+12 -10
View File
@@ -1533,17 +1533,19 @@ receive_handle_existing_object(const struct receive_writer_arg *rwa,
}
/*
* The dmu does not currently support decreasing nlevels
* or changing the number of dnode slots on an object. For
* non-raw sends, this does not matter and the new object
* can just use the previous one's nlevels. For raw sends,
* however, the structure of the received dnode (including
* nlevels and dnode slots) must match that of the send
* side. Therefore, instead of using dmu_object_reclaim(),
* we must free the object completely and call
* dmu_object_claim_dnsize() instead.
* The dmu does not currently support decreasing nlevels or changing
* indirect block size if there is already one, same as changing the
* number of of dnode slots on an object. For non-raw sends this
* does not matter and the new object can just use the previous one's
* parameters. For raw sends, however, the structure of the received
* dnode (including indirects and dnode slots) must match that of the
* send side. Therefore, instead of using dmu_object_reclaim(), we
* must free the object completely and call dmu_object_claim_dnsize()
* instead.
*/
if ((rwa->raw && drro->drr_nlevels < doi->doi_indirection) ||
if ((rwa->raw && ((doi->doi_indirection > 1 &&
indblksz != doi->doi_metadata_block_size) ||
drro->drr_nlevels < doi->doi_indirection)) ||
dn_slots != doi->doi_dnodesize >> DNODE_SHIFT) {
err = dmu_free_long_object(rwa->os, drro->drr_object);
if (err != 0)
+26 -17
View File
@@ -1773,7 +1773,14 @@ dnode_try_claim(objset_t *os, uint64_t object, int slots)
}
/*
* Checks if the dnode contains any uncommitted dirty records.
* Checks if the dnode itself is dirty, or is carrying any uncommitted records.
* It is important to check both conditions, as some operations (eg appending
* to a file) can dirty both as a single logical unit, but they are not synced
* out atomically, so checking one and not the other can result in an object
* appearing to be clean mid-way through a commit.
*
* Do not change this lightly! If you get it wrong, dmu_offset_next() can
* detect a hole where there is really data, leading to silent corruption.
*/
boolean_t
dnode_is_dirty(dnode_t *dn)
@@ -1781,7 +1788,8 @@ dnode_is_dirty(dnode_t *dn)
mutex_enter(&dn->dn_mtx);
for (int i = 0; i < TXG_SIZE; i++) {
if (multilist_link_active(&dn->dn_dirty_link[i])) {
if (multilist_link_active(&dn->dn_dirty_link[i]) ||
!list_is_empty(&dn->dn_dirty_records[i])) {
mutex_exit(&dn->dn_mtx);
return (B_TRUE);
}
@@ -1891,7 +1899,7 @@ dnode_set_blksz(dnode_t *dn, uint64_t size, int ibs, dmu_tx_t *tx)
if (ibs == dn->dn_indblkshift)
ibs = 0;
if (size >> SPA_MINBLOCKSHIFT == dn->dn_datablkszsec && ibs == 0)
if (size == dn->dn_datablksz && ibs == 0)
return (0);
rw_enter(&dn->dn_struct_rwlock, RW_WRITER);
@@ -1914,24 +1922,25 @@ dnode_set_blksz(dnode_t *dn, uint64_t size, int ibs, dmu_tx_t *tx)
if (ibs && dn->dn_nlevels != 1)
goto fail;
/* resize the old block */
err = dbuf_hold_impl(dn, 0, 0, TRUE, FALSE, FTAG, &db);
if (err == 0) {
dbuf_new_size(db, size, tx);
} else if (err != ENOENT) {
goto fail;
}
dnode_setdblksz(dn, size);
dnode_setdirty(dn, tx);
dn->dn_next_blksz[tx->tx_txg&TXG_MASK] = size;
if (size != dn->dn_datablksz) {
/* resize the old block */
err = dbuf_hold_impl(dn, 0, 0, TRUE, FALSE, FTAG, &db);
if (err == 0) {
dbuf_new_size(db, size, tx);
} else if (err != ENOENT) {
goto fail;
}
dnode_setdblksz(dn, size);
dn->dn_next_blksz[tx->tx_txg & TXG_MASK] = size;
if (db)
dbuf_rele(db, FTAG);
}
if (ibs) {
dn->dn_indblkshift = ibs;
dn->dn_next_indblkshift[tx->tx_txg&TXG_MASK] = ibs;
dn->dn_next_indblkshift[tx->tx_txg & TXG_MASK] = ibs;
}
/* release after we have fixed the blocksize in the dnode */
if (db)
dbuf_rele(db, FTAG);
rw_exit(&dn->dn_struct_rwlock);
return (0);
+1 -1
View File
@@ -270,7 +270,7 @@ typedef struct indirect_split {
*/
indirect_child_t *is_good_child;
indirect_child_t is_child[1]; /* variable-length */
indirect_child_t is_child[];
} indirect_split_t;
/*
+3
View File
@@ -468,6 +468,9 @@ vdev_config_generate(spa_t *spa, vdev_t *vd, boolean_t getstats,
if (vd->vdev_isspare)
fnvlist_add_uint64(nv, ZPOOL_CONFIG_IS_SPARE, 1);
if (flags & VDEV_CONFIG_L2CACHE)
fnvlist_add_uint64(nv, ZPOOL_CONFIG_ASHIFT, vd->vdev_ashift);
if (!(flags & (VDEV_CONFIG_SPARE | VDEV_CONFIG_L2CACHE)) &&
vd == vd->vdev_top) {
fnvlist_add_uint64(nv, ZPOOL_CONFIG_METASLAB_ARRAY,
+19 -9
View File
@@ -23,6 +23,7 @@
* Copyright (c) 2016 by Delphix. All rights reserved.
* Copyright (c) 2019 by Lawrence Livermore National Security, LLC.
* Copyright (c) 2021 Hewlett Packard Enterprise Development LP
* Copyright 2023 RackTop Systems, Inc.
*/
#include <sys/spa.h>
@@ -572,6 +573,7 @@ vdev_trim_ranges(trim_args_t *ta)
uint64_t extent_bytes_max = ta->trim_extent_bytes_max;
uint64_t extent_bytes_min = ta->trim_extent_bytes_min;
spa_t *spa = vd->vdev_spa;
int error = 0;
ta->trim_start_time = gethrtime();
ta->trim_bytes_done = 0;
@@ -591,19 +593,32 @@ vdev_trim_ranges(trim_args_t *ta)
uint64_t writes_required = ((size - 1) / extent_bytes_max) + 1;
for (uint64_t w = 0; w < writes_required; w++) {
int error;
error = vdev_trim_range(ta, VDEV_LABEL_START_SIZE +
rs_get_start(rs, ta->trim_tree) +
(w *extent_bytes_max), MIN(size -
(w * extent_bytes_max), extent_bytes_max));
if (error != 0) {
return (error);
goto done;
}
}
}
return (0);
done:
/*
* Make sure all TRIMs for this metaslab have completed before
* returning. TRIM zios have lower priority over regular or syncing
* zios, so all TRIM zios for this metaslab must complete before the
* metaslab is re-enabled. Otherwise it's possible write zios to
* this metaslab could cut ahead of still queued TRIM zios for this
* metaslab causing corruption if the ranges overlap.
*/
mutex_enter(&vd->vdev_trim_io_lock);
while (vd->vdev_trim_inflight[0] > 0) {
cv_wait(&vd->vdev_trim_io_cv, &vd->vdev_trim_io_lock);
}
mutex_exit(&vd->vdev_trim_io_lock);
return (error);
}
static void
@@ -922,11 +937,6 @@ vdev_trim_thread(void *arg)
}
spa_config_exit(spa, SCL_CONFIG, FTAG);
mutex_enter(&vd->vdev_trim_io_lock);
while (vd->vdev_trim_inflight[0] > 0) {
cv_wait(&vd->vdev_trim_io_cv, &vd->vdev_trim_io_lock);
}
mutex_exit(&vd->vdev_trim_io_lock);
range_tree_destroy(ta.trim_tree);
+6 -2
View File
@@ -418,12 +418,16 @@ zil_parse(zilog_t *zilog, zil_parse_blk_func_t *parse_blk_func,
lr_t *lr = (lr_t *)lrp;
reclen = lr->lrc_reclen;
ASSERT3U(reclen, >=, sizeof (lr_t));
if (lr->lrc_seq > claim_lr_seq)
if (lr->lrc_seq > claim_lr_seq) {
arc_buf_destroy(abuf, &abuf);
goto done;
}
error = parse_lr_func(zilog, lr, arg, txg);
if (error != 0)
if (error != 0) {
arc_buf_destroy(abuf, &abuf);
goto done;
}
ASSERT3U(max_lr_seq, <, lr->lrc_seq);
max_lr_seq = lr->lrc_seq;
lr_count++;
+1 -1
View File
@@ -42,4 +42,4 @@ gensymbols:
@OBJDUMP@ -t lib/zstd.o | awk '$$2 == "g" && !/ zfs_/ {print "#define\t" $$6 " zfs_" $$6}' | sort >> include/zstd_compat_wrapper.h
checksymbols:
@OBJDUMP@ -t lib/zstd.o | awk '/file format/ {print} $$2 == "g" && !/ zfs_/ {++ret; print} END {exit ret}'
@OBJDUMP@ -t lib/zstd.o | awk '/file format/ {print} $$2 == "g" && (!/ zfs_/ && !/ __pfx_zfs_/) {++ret; print} END {exit ret}'
+4 -41
View File
@@ -68,46 +68,9 @@ fi
%defattr(-,root,root)
/usr/src/%{module}-%{version}
%post
for POSTINST in /usr/lib/dkms/common.postinst; do
if [ -f $POSTINST ]; then
$POSTINST %{module} %{version}
exit $?
fi
echo "WARNING: $POSTINST does not exist."
done
echo -e "ERROR: DKMS version is too old and %{module} was not"
echo -e "built with legacy DKMS support."
echo -e "You must either rebuild %{module} with legacy postinst"
echo -e "support or upgrade DKMS to a more current version."
exit 1
%preun
# Are we doing an upgrade?
if [ "$1" = "1" -o "$1" = "upgrade" ] ; then
# Yes we are. Are we upgrading to a new ZFS version?
NEWEST_VER=$(dkms status zfs | tr -d , | sort -r -V | awk '/installed/{print $2; exit}')
if [ "$NEWEST_VER" != "%{version}" ] ; then
# Yes, it's a new ZFS version. We'll uninstall the old module
# later on in this script.
true
else
# No, it's probably an upgrade of the same ZFS version
# to a new distro (zfs-dkms-0.7.12.fc28->zfs-dkms-0.7.12.fc29).
# Don't remove our modules, since the rebuild for the new
# distro will automatically delete the old modules.
exit 0
fi
fi
dkms remove -m %{module} -v %{version} --all
%posttrans
/usr/lib/dkms/common.postinst %{module} %{version}
# If we're here then we're doing an uninstall (not upgrade).
CONFIG_H="/var/lib/dkms/%{module}/%{version}/*/*/%{module}_config.h"
SPEC_META_ALIAS="@PACKAGE@-@VERSION@-@RELEASE@"
DKMS_META_ALIAS=`cat $CONFIG_H 2>/dev/null |
awk -F'"' '/META_ALIAS\s+"/ { print $2; exit 0 }'`
if [ "$SPEC_META_ALIAS" = "$DKMS_META_ALIAS" ]; then
echo -e
echo -e "Uninstall of %{module} module ($SPEC_META_ALIAS) beginning:"
dkms remove -m %{module} -v %{version} --all %{!?not_rpm:--rpm_safe_upgrade}
fi
exit 0
+32 -32
View File
@@ -164,8 +164,8 @@ summary = {
# reasons listed above can be used.
#
known = {
'casenorm/mixed_none_lookup_ci': ['FAIL', '7633'],
'casenorm/mixed_formd_lookup_ci': ['FAIL', '7633'],
'casenorm/mixed_none_lookup_ci': ['FAIL', 7633],
'casenorm/mixed_formd_lookup_ci': ['FAIL', 7633],
'cli_root/zfs_unshare/zfs_unshare_002_pos': ['SKIP', na_reason],
'cli_root/zfs_unshare/zfs_unshare_006_pos': ['SKIP', na_reason],
'cli_root/zpool_import/import_rewind_device_replaced':
@@ -175,7 +175,7 @@ known = {
'privilege/setup': ['SKIP', na_reason],
'refreserv/refreserv_004_pos': ['FAIL', known_reason],
'rootpool/setup': ['SKIP', na_reason],
'rsend/rsend_008_pos': ['SKIP', '6066'],
'rsend/rsend_008_pos': ['SKIP', 6066],
'vdev_zaps/vdev_zaps_007_pos': ['FAIL', known_reason],
}
@@ -193,10 +193,10 @@ if sys.platform.startswith('freebsd'):
})
elif sys.platform.startswith('linux'):
known.update({
'casenorm/mixed_formd_lookup': ['FAIL', '7633'],
'casenorm/mixed_formd_delete': ['FAIL', '7633'],
'casenorm/sensitive_formd_lookup': ['FAIL', '7633'],
'casenorm/sensitive_formd_delete': ['FAIL', '7633'],
'casenorm/mixed_formd_lookup': ['FAIL', 7633],
'casenorm/mixed_formd_delete': ['FAIL', 7633],
'casenorm/sensitive_formd_lookup': ['FAIL', 7633],
'casenorm/sensitive_formd_delete': ['FAIL', 7633],
'removal/removal_with_zdb': ['SKIP', known_reason],
})
@@ -220,49 +220,49 @@ maybe = {
'cli_root/zfs_destroy/zfs_destroy_dev_removal_condense':
['FAIL', known_reason],
'cli_root/zfs_get/zfs_get_004_pos': ['FAIL', known_reason],
'cli_root/zfs_get/zfs_get_009_pos': ['SKIP', '5479'],
'cli_root/zfs_get/zfs_get_009_pos': ['SKIP', 5479],
'cli_root/zfs_rollback/zfs_rollback_001_pos': ['FAIL', known_reason],
'cli_root/zfs_rollback/zfs_rollback_002_pos': ['FAIL', known_reason],
'cli_root/zfs_share/setup': ['SKIP', share_reason],
'cli_root/zfs_snapshot/zfs_snapshot_002_neg': ['FAIL', known_reason],
'cli_root/zfs_unshare/setup': ['SKIP', share_reason],
'cli_root/zpool_add/zpool_add_004_pos': ['FAIL', known_reason],
'cli_root/zpool_destroy/zpool_destroy_001_pos': ['SKIP', '6145'],
'cli_root/zpool_import/zpool_import_missing_003_pos': ['SKIP', '6839'],
'cli_root/zpool_destroy/zpool_destroy_001_pos': ['SKIP', 6145],
'cli_root/zpool_import/zpool_import_missing_003_pos': ['SKIP', 6839],
'cli_root/zpool_initialize/zpool_initialize_import_export':
['FAIL', '11948'],
['FAIL', 11948],
'cli_root/zpool_labelclear/zpool_labelclear_removed':
['FAIL', known_reason],
'cli_root/zpool_trim/setup': ['SKIP', trim_reason],
'cli_root/zpool_upgrade/zpool_upgrade_004_pos': ['FAIL', '6141'],
'cli_root/zpool_upgrade/zpool_upgrade_004_pos': ['FAIL', 6141],
'delegate/setup': ['SKIP', exec_reason],
'fallocate/fallocate_punch-hole': ['SKIP', fspacectl_reason],
'history/history_004_pos': ['FAIL', '7026'],
'history/history_005_neg': ['FAIL', '6680'],
'history/history_006_neg': ['FAIL', '5657'],
'history/history_004_pos': ['FAIL', 7026],
'history/history_005_neg': ['FAIL', 6680],
'history/history_006_neg': ['FAIL', 5657],
'history/history_008_pos': ['FAIL', known_reason],
'history/history_010_pos': ['SKIP', exec_reason],
'io/mmap': ['SKIP', fio_reason],
'largest_pool/largest_pool_001_pos': ['FAIL', known_reason],
'mmp/mmp_on_uberblocks': ['FAIL', known_reason],
'pyzfs/pyzfs_unittest': ['SKIP', python_deps_reason],
'pool_checkpoint/checkpoint_discard_busy': ['FAIL', '11946'],
'pool_checkpoint/checkpoint_discard_busy': ['FAIL', 11946],
'pam/setup': ['SKIP', "pamtester might be not available"],
'projectquota/setup': ['SKIP', exec_reason],
'removal/removal_condense_export': ['FAIL', known_reason],
'reservation/reservation_008_pos': ['FAIL', '7741'],
'reservation/reservation_018_pos': ['FAIL', '5642'],
'reservation/reservation_008_pos': ['FAIL', 7741],
'reservation/reservation_018_pos': ['FAIL', 5642],
'snapshot/clone_001_pos': ['FAIL', known_reason],
'snapshot/snapshot_009_pos': ['FAIL', '7961'],
'snapshot/snapshot_010_pos': ['FAIL', '7961'],
'snapused/snapused_004_pos': ['FAIL', '5513'],
'snapshot/snapshot_009_pos': ['FAIL', 7961],
'snapshot/snapshot_010_pos': ['FAIL', 7961],
'snapused/snapused_004_pos': ['FAIL', 5513],
'tmpfile/setup': ['SKIP', tmpfile_reason],
'trim/setup': ['SKIP', trim_reason],
'upgrade/upgrade_projectquota_001_pos': ['SKIP', project_id_reason],
'user_namespace/setup': ['SKIP', user_ns_reason],
'userquota/setup': ['SKIP', exec_reason],
'vdev_zaps/vdev_zaps_004_pos': ['FAIL', known_reason],
'zvol/zvol_ENOSPC/zvol_ENOSPC_001_pos': ['FAIL', '5848'],
'zvol/zvol_ENOSPC/zvol_ENOSPC_001_pos': ['FAIL', 5848],
}
if sys.platform.startswith('freebsd'):
@@ -274,11 +274,11 @@ if sys.platform.startswith('freebsd'):
['FAIL', known_reason],
'cli_root/zpool_import/zpool_import_012_pos': ['FAIL', known_reason],
'delegate/zfs_allow_003_pos': ['FAIL', known_reason],
'inheritance/inherit_001_pos': ['FAIL', '11829'],
'inheritance/inherit_001_pos': ['FAIL', 11829],
'resilver/resilver_restart_001': ['FAIL', known_reason],
'pool_checkpoint/checkpoint_big_rewind': ['FAIL', '12622'],
'pool_checkpoint/checkpoint_indirect': ['FAIL', '12623'],
'snapshot/snapshot_002_pos': ['FAIL', '14831'],
'pool_checkpoint/checkpoint_big_rewind': ['FAIL', 12622],
'pool_checkpoint/checkpoint_indirect': ['FAIL', 12623],
'snapshot/snapshot_002_pos': ['FAIL', 14831],
})
elif sys.platform.startswith('linux'):
maybe.update({
@@ -296,7 +296,7 @@ elif sys.platform.startswith('linux'):
'mmp/mmp_active_import': ['FAIL', known_reason],
'mmp/mmp_exported_import': ['FAIL', known_reason],
'mmp/mmp_inactive_import': ['FAIL', known_reason],
'zvol/zvol_misc/zvol_misc_snapdev': ['FAIL', '12621'],
'zvol/zvol_misc/zvol_misc_snapdev': ['FAIL', 12621],
'zvol/zvol_misc/zvol_misc_volmode': ['FAIL', known_reason],
})
@@ -327,7 +327,7 @@ if os.environ.get('CI') == 'true':
})
maybe.update({
'events/events_002_pos': ['FAIL', '11546'],
'events/events_002_pos': ['FAIL', 11546],
})
elif sys.platform.startswith('linux'):
maybe.update({
@@ -451,13 +451,13 @@ if __name__ == "__main__":
if test in known:
if known[test][1] == na_reason:
continue
elif known[test][1].isdigit():
expect = issue_url + known[test][1]
elif isinstance(known[test][1], int):
expect = f"{issue_url}{known[test][1]}"
else:
expect = known[test][1]
elif test in maybe:
if maybe[test][1].isdigit():
expect = issue_url + maybe[test][1]
if isinstance(maybe[test][1], int):
expect = f"{issue_url}{maybe[test][1]}"
else:
expect = maybe[test][1]
elif setup in known and known[setup][0] == "SKIP" and setup != test:
+1 -1
View File
@@ -4268,7 +4268,7 @@ function arcstat_quiescence # stat echo
while $do_once || [ $stat1 -ne $stat2 ] || [ $stat2 -eq 0 ]; do
typeset stat1=$(get_arcstat $stat)
sleep 2
sleep 0.5
typeset stat2=$(get_arcstat $stat)
do_once=false
done
@@ -27,15 +27,14 @@
#
# STRATEGY:
# 1. Create pool with a cache device.
# 2. Export and re-import pool without writing any data.
# 3. Create a random file in that pool and random read for 10 sec.
# 4. Export pool.
# 5. Read the amount of log blocks written from the header of the
# 2. Create a random file in that pool and random read for 10 sec.
# 3. Export pool.
# 4. Read the amount of log blocks written from the header of the
# L2ARC device.
# 6. Import pool.
# 7. Read the amount of log blocks rebuilt in arcstats and compare to
# 5. Import pool.
# 6. Read the amount of log blocks rebuilt in arcstats and compare to
# (5).
# 8. Check if the labels of the L2ARC device are intact.
# 7. Check if the labels of the L2ARC device are intact.
#
# * We can predict the minimum bytes of L2ARC restored if we subtract
# from the effective size of the cache device the bytes l2arc_evict()
@@ -75,10 +74,8 @@ export FILE_SIZE=$(( floor($fill_mb / $NUMJOBS) ))M
log_must truncate -s ${cache_sz}M $VDEV_CACHE
log_must zpool create -f $TESTPOOL $VDEV cache $VDEV_CACHE
log_must zpool export $TESTPOOL
log_must zpool import -d $VDIR $TESTPOOL
log_must zpool create -f -o ashift=12 $TESTPOOL $VDEV
log_must zpool add $TESTPOOL cache $VDEV_CACHE
log_must fio $FIO_SCRIPTS/mkfiles.fio
log_must fio $FIO_SCRIPTS/random_reads.fio
@@ -65,7 +65,7 @@ typeset VDEV_MIN_MB=$((MINVDEVSIZE * 0.30 / 1024 / 1024))
log_must zpool create -f $TESTPOOL $TRIM_VDEV1 cache $TRIM_VDEV2
verify_vdevs "-le" "$VDEV_MIN_MB" $TRIM_VDEV2
typeset fill_mb=$(( floor(2 * MINVDEVSIZE) ))
typeset fill_mb=$(( floor(3 * MINVDEVSIZE) ))
export DIRECTORY=/$TESTPOOL
export NUMJOBS=1
export FILE_SIZE=${fill_mb}