in r247265 (ZFS deadman thread). Both new utilities now support the old
kernel and new kernel properly detects old utilities.
For future backwards compatibility, the vfs.zfs.version.ioctl read-only
sysctl has been introduced. With this sysctl zfs utilities will be able
to detect the ioctl interface version of the currently loaded zfs module.
As a side effect, the zfs utilities between r247265 and this revision don't
support the old kernel module. If you are using HEAD newer or equal than
r247265, install the new kernel module (or whole kernel) first.
MFC after: 10 days
Merge the ZFS I/O deadman thread from vendor (illumos).
This feature panics the system on hanging ZFS I/O, helps debugging
and resumes failed service.
The panic behavior can be controlled with the loader-only tunables:
vfs.zfs.deadman_enabled (enable or disable panic on stalled ZFS I/O)
vfs.zfs.deadman_synctime (expiration time for stalled ZFS I/O)
By default, ZFS I/O deadman is enabled by default on amd64 and i386
excluding virtual guest machines.
Illumos ZFS issues:
3246 ZFS I/O deadman thread
References:
https://www.illumos.org/issues/3246
MFC after: 2 weeks
process that has crashed.
sys/cddl/contrib/opensolaris/uts/common/dtrace/fasttrap.c:
In fasttrap_pid_disable(), we cannot PHOLD the proc
structure for a process that no longer exists, but
we still have other, fasttrap specific, state that
must be cleaned up for probes that existed in the
dead process. Instead of returning early if the
process related to our probes isn't found,
conditionalize the locking and carry on with a NULL
proc pointer. The rest of the fasttrap code already
understands that a NULL proc is possible and does
the right things in this case.
Sponsored by: Spectra Logic Corporation
Reviewed by: rpaulo, gnn
MFC after: 1 week
Fixes several problems if working with read-only pools.
Changed code originaly introduced in onnv-gate 13061:bda0decf867b
Contains changes up to illumos-gate 13700:4bc0783f6064
PR: kern/175897
Suggested by: avg
MFC after: 2 weeks
Import vendor bugfixes regarding SA rounding, header size and layout.
This was already partially fixed by avg.
Illumos ZFS issues:
3512 rounding discrepancy in sa_find_sizes()
3513 mismatch between SA header size and layout
References:
https://www.illumos.org/issues/3512https://www.illumos.org/issues/3513
MFC after: 2 weeks
* Illumos zfs issue #3035 [1] LZ4 compression support in ZFS.
LZ4 is a new high-speed BSD-licensed compression algorithm created
by Yann Collet that delivers very high compression and decompression
performance compared to lzjb (>50% faster on compression, >80% faster
on decompression and around 3x faster on compression of incompressible
data), while giving better compression ratio [1].
This version of LZ4 corresponds to upstream's [2] revision 85.
Please note that for obvious reasons this is not backward read
compatible. This means once a pool have LZ4 compressed data, these
data can no longer be read by older ZFS implementations.
Local changes:
- On-stack hash table disabled and using kernel slab allocator
instead, at this time. This requires larger kernel thread stack
for zio workers. This may change in the future should we adjusted
the zio workers' thread stack size.
- likely and unlikely will be undefined if they are already defined,
this is required for i386 XEN build.
- Removed De Bruijn sequence based __builtin_ctz family of builtins
in favor of the latter. Both GCC and clang supports these builtins.
- Changed the way the LZ4 code detects endianness.
- Manual pages modifications to mention the feature based on Illumos
counterpart.
- Boot loader changes to make it support LZ4 decompression.
[1] https://www.illumos.org/issues/3035
[2] http://code.google.com/p/lz4/source/list
Obtained from: Illumos (13921:9d721847e469)
Tested on: FreeBSD/amd64
MFC after: 1 month
A special gfs vnode corresponds to that object.
A regular zfs vnode must not be returned.
This should be upstreamed.
Reported by: pluknet
Submitted by: rmacklem
Tested by: pluknet
MFC after: 10 days
Now they are split into two pairs: page_hold/page_unhold for mappedread
and page_busy/page_unbusy for update_pages.
For mappedread we simply hold a page that is to be used as a source if it
is resident and valid (and not busy). This is sufficient since we are
only doing page -> user buffer copying. There is no page <-> backing
storage I/O involved.
update_pages is now better split to properly handle the putpages case
(page -> arc) and the regular write case (arc -> page).
For the latter we use complete protocol of marking an object with
paging-in-progress and marking a page with io_start (busy count).
Also, in this case we remove the write bit from all page mappings and
clear dirty bits of the pages, the former is needed to ensure that the
latter does the right thing.
Additionally we update a page if it is cached instead of just freeing it
as was done before. This needs to be verified.
A minor detail: ZFS-backed pages should always be either fully valid
or fully invalid. Assert this and use simpler API that does not deal
with sub-page blocks.
Reviewed by: kib
MFC after: 26 days
- there is no such flag in Solaris and derivatives
- the flag was added in an unrelated change
- the flag is not used
The proper way to allocate zeroed out memory is to use kmem_zalloc.
MFC after: 3 days
This should allow the kernel linker to easily detect a situation
when the module is present both in a kernel and in a preloaded file
(zfs.ko).
Reviewed by: jhb
MFC after: 5 days
It seems that old ZFS versions (v15) completely omit "vdev_children"
property when there is a single child.
Reported by: jase
Tested by: jase
MFC after: 1 week
the underlying zap_count() to return no errors. However, it is possible
that the pool reaches to such a state where zap_count would return error,
leading to panics when a pool is imported.
This commit changes the ddt_zap_count to return error returned from
zap_count and handle the error appropriately. With this change, it's now
possible to let zpool rollback damaged transaction groups and import the
pool.
Obtained from: ZFS on Linux github (e8fd45a0f975c6b8ae8cd644714fc21f14fac2bf)
MFC after: 1 month
- Use spinlock_enter()/spinlock_exit() to prevent a thread holding a
debug lock from being preempted to prevent other threads waiting
on that lock from starvation.
- Handle the possibility of CPU migration in between the fetch of curcpu
and the call to spinlock_enter() by saving curcpu in a local variable.
- Use memory barriers to prevent reordering of loads and stores of the
data protected by the lock outside of the critical section
- Eliminate false sharing of the locks by moving them into the structures
that they protect and aligning them to a cacheline boundary.
- Record the owning thread in the lock to make debugging future problems
easier.
Reviewed by: rpaulo (initial version)
MFC after: 2 weeks
initialisation to be enabled (1) / disabled (0) defaults to enabled.
This is useful for devices which have a slow trim speed and are either
new or have otherwise already been wiped e.g. secure erase.
PR: kern/173116
Submitted by: Steven Hartland
Approved by: pjd (mentor)
making range consolidation much more effective particularly for small
deletes.
This reduces memory used by the free map as well as reducing the number
of bio requests down to geom required to process all deletes.
In tests this achieved a factor of 10 reduction of trim ranges / geom
call downs.
While I'm here correct the description of zio_vdev_io_start.
PR: kern/173254
Submitted by: Steven Hartland
Approved by: pjd (mentor)
Also, make it explicit that V_XATTRDIR is not properly supported in gfs
code yet.
The bad code was plain incorrect: (a) it spoiled handling of v_usecount
reaching zero and (b) it leaked v_holdcnt.
The ugly code employs potentially unsafe locking tricks.
Ideally we should separate vnode lifecycle and gfs node lifecycle.
A gfs node should have its own reference count where its child nodes
should be accounted.
PR: kern/151111
Reviewed by: kib
MFC after: 13 days
Introduce a new dataset aclmode setting "restricted" to protect ACL's
being destroyed or corrupted by a drive-by chmod.
illumos-gate 13889:a67716f16746
3254 add support in zfs for aclmode=restricted
References:
https://www.illumos.org/issues/3254
MFC after: 2 weeks
Import the zio nop-write improvement from Illumos. To reduce I/O,
nop-write omits overwriting data if the checksum (cryptographically
secure) of new data matches the checksum of existing data.
It also saves space if snapshots are in use.
It currently works only on datasets with enabled compression, disabled
deduplication and sha256 checksums.
IllumOS 13887:196932ec9e6a and 13888:7204b3392a58
3236 zio nop-write
References:
https://www.illumos.org/issues/3236
MFC after: 2 weeks
* There is no need for the delayed destruction of znodes via taskqueue,
now that we do not need to fear recursion from getnewvnode into
zfs_inactive and zfs_freebsd_reclaim, thus making znode/vnode state
machine a bit simpler.
* More complete porting of zfs_inactive from Solaris VFS model to FreeBSD
vop_inactive and vop_reclaim model. All destructive actions are done
in zfs_freebsd_reclaim.
This allows to simplify zfs_zget logic.
* Allow zfs_zget to return a doomed vnode if the current thread already
has an exclusive lock on the vnode.
* Clean up Solaris-isms like bailing out of reclaim/inactive on certain
values of v_usecount (aka v_count) or directly messing with this counter.
* Do not clear z_vnode while znode is still accessible.
z_vnode should be cleared only after zfs_znode_dmu_fini.
Otherwise zfs_zget may get an effectively half-deconstructed znode.
This allows to simplify zfs_zget logic further.
The above changes fix at least two known/reported problems:
o An indefinite wait in the following code path:
vgone -> VOP_RECLAIM -> zfs_freebsd_reclaim -> vnode_destroy_vobject ->
put_pages -> zfs_write -> zil_commit -> zfs_zget
This happened because vgone marks a vnode as VI_DOOMED before calling
VOP_RECLAIM, but zfs_zget would not return a doomed vnode under any
circumstances.
The fix in this change is not complete as it won't fix a deadlock between
two threads doing VOP_RECLAIM where one thread is in zil_commit trying to
zfs_zget a znode/vnode being reclaimed by the other thread, which would be
blocked trying to enter zil_commit. This type of deadlock has not been
reported as of now.
o An indefinite wait in the unmount path caused by a znode "falling through
the cracks" in inactive+reclaim. This would happen if the znode is unlinked
while its vnode is still active.
To Do: pass locking flags parameter to zfs_zget, so that the zfs-vfs
glue code doesn't have to re-lock a vnode but could ask for proper locking
from the very start. This would also allow for the higher level code to
obtain a doomed vnode when it is expected/requested. Or to avoid blocking
when it is not allowed (see zil_commit example above).
ffs_vgetf seems like a good source of inspiration.
Tested by: Willem Jan Withagen <wjw@digiware.nl>
MFC after: 6 weeks
... otherwise zfs_getpages would mostly be called with one page at a time.
It is expected that ZFS VOP_BMAP is only called from vnode_pager_haspage.
Since ZFS files can have variable block sizes and also because we don't
really know if any given blocks are consecutive, we can not really report
any additional blocks behind or ahead of a given block. Since physical
block numbers do not make sense for ZFS, we do not do any real translation
and thus pass back blk = lblk. The net effect is that vnode_pager_haspage
knows that the block exists and that the pages backed by the block can be
accessed. vnode_pager_haspage may be wrong about the exact count of the
pages backed by the block, because of a variable block size, which
vnode_pager_haspage doesn't really know - it only knows max block size in
a filesystem. So pages from multiple blocks can be passed to zfs_getpages,
but that is expected and correctly handled.
vnode_pager should not call zfs_bmap for any other reason, because ZFS
implements VOP_PUTPAGES and thus vnode_pager_generic_getpages is not used.
vfs_cluster code vfs_bio code should not be called for ZFS, because ZFS does
not use buffer cache layer.
Also, ZFS does not use vn_bmap_seekhole, it has its prviate mechanism for
working with holes.
The above list should cover all the current calls to VOP_BMAP.
Reviewed by: kib
MFC after: 6 weeks
Illumos 13886:e3261d03efbf
3349 zpool upgrade -V bumps the on disk version number, but leaves
the in core version
References:
https://www.illumos.org/issues/3349
MFC after: 1 week
... because the latter makes some decision based on the version.
This is especially important for raidz vdevs.
This is similar to what spa_load does.
This is not an issue for upstream because they do not seem to support
using raidz as a root pool.
Reported by: Andrei Lavreniyuk <andy.lavr@gmail.com>
Tested by: Andrei Lavreniyuk <andy.lavr@gmail.com>
MFC after: 6 days
The call is a NOP, because pool version in spa_ubsync.ub_version is not
initialized and thus appears to be zero.
If the version is properly set then the call leads to a NULL pointer
dereference because the spa object is still under-constructed.
The same change was independently made in the upstream as a part of
a larger change (4445fffbbb1ea25fd0e9ea68b9380dd7a6709025).
MFC after: 6 days
if we fail to generate a proper root pool configuration based on disk
probing. Currently we can not properly generate the configuration for
multi-vdev pools. Make that explicit.
Reported by: madpilot, Bartosz Stec <bartosz.stec@it4pro.pl>
Tested by: madpilot, Bartosz Stec <bartosz.stec@it4pro.pl>
MFC after: 4 days
designator to select a process which is waited for. The system call
optionally returns siginfo_t which would be otherwise provided to
SIGCHLD handler, as well as extended structure accounting for child
and cumulative grandchild resource usage.
Allow to get the current rusage information for non-exited processes
as well, similar to Solaris.
The explicit WEXITED flag is required to wait for exited processes,
allowing for more fine-grained control of the events the waiter is
interested in.
Fix the handling of siginfo for WNOWAIT option for all wait*(2)
family, by not removing the queued signal state.
PR: standards/170346
Submitted by: "Jukka A. Ukkonen" <jau@iki.fi>
MFC after: 1 month
... before trying to destroy the zvol snapshots themselves.
PR: kern/173442
Reported by: Petri Helenius <petri@helenius.fi>,
mm
Obtained from: Brian Behlendorf <behlendorf1@llnl.gov>,
Illumos Bug #3170
Tested by: Petri Helenius <petri@helenius.fi>
MFC after: 10 days
Illumos r13840:97fd5cdf328a:
3145 single-copy arc
3212 ztest: race condition between vdev_online() and spa_vdev_remove()
Illumos r13849:3468a95b27cd:
3258 ztest's use of file descriptors is unstable
There is one known issue: Some probes will display an error message along the
lines of: "Invalid address (0)"
I tested this with both a simple dtrace probe and dtruss on a few different
binaries on 32-bit. I only compiled 64-bit, did not run it, but I don't expect
problems without the modules loaded. Volunteers are welcome.
MFC after: 1 month
Otherwise we could fail with an incorrect error if e.g. parent
object id is removed too or we can even return a wrong vnode if
parent object has been already re-used.
Discussed with: pjd
Also see: http://article.gmane.org/gmane.os.freebsd.devel.file-systems/13863
MFC after: 26 days
vn_lock should do the right thing with respect to given vnode lock
flags. If a caller doesn't mind a doomed vnode, then zfs should deliver.
Reviewed by: kib
MFC after: 19 days
It turned out to be not that useful, because its default value may lead
to a problem when a root pool is present in zpool.cache, but its
on-disk status is 'exported'. This may happen if the pool was imported
in a different environment with -f flag and then exported.
MFC after: 12 days
POOL_STATE_SPARE and POOL_STATE_L2CACHE were not handled correctly
and thus the cache and spare disks would not be correctly probed.
Reported by: Michael Schmiedgen <schmiedgen@gmx.net>,
Matthew D. Fuller <fullermd@over-yonder.net>
Tested by: Michael Schmiedgen <schmiedgen@gmx.net>,
flo
MFC after: 5 days
In particular, do not lock Giant conditionally when calling into the
filesystem module, remove the VFS_LOCK_GIANT() and related
macros. Stop handling buffers belonging to non-mpsafe filesystems.
The VFS_VERSION is bumped to indicate the interface change which does
not result in the interface signatures changes.
Conducted and reviewed by: attilio
Tested by: pho
... otherwise the current thread might be holding ARC locks and thus run
into a deadlock. This happens, for example, when a thread does memory
allocation in the ARC code and runs into KVA shortage.
Also, it really makes the most sense to wait in pageproc, so that the
results of ARC reclamation are seen before the page cache is acted upon.
In other cases where vm_lowmem is invoked, e.g. on KVA space shortage,
the callers perform multiple attempts (up to 8) and wait for rather
long intervals between them (up to 4 seconds), so ARC reclaim results
should become visible even without explicit waiting on the ARC thread.
Note that this is not a critical issue for typical ZFS usages where KVA
space should already be large enough. On amd64 systems setting KVA size
to twice the physical memory size is known to mitigate KVA fragmentation
issues in practice.
Side note: perhaps vm_lowmem 'how' parameter should be used to
differentiate between causes of the event.
Reported by: Nikolay Denev <ndenev@gmail.com>
MFC after: 19 days
getnewvnode_reserve helps to avoid "recursing" back into zfs code
via getnewvnode when that latter needs to reclaim some vnodes.
zfs code may hold a number of locks around getnewvnode and doesn't
expect any recursion to happen on those locks, because that never
happens in solaris.
I believe that this change also eleiminates a need for the delayed
znode destruction via the taskqueue.
Many thanks to kib for devising getnewvnode_reserve.
Reported by: flo
Tested by: bapt, kwm, swills
MFC after: 2 weeks
X-MFC after: r241556
... instead of deferring the action until first open.
Unlike upstream this has no benefit on FreeBSD.
We know that as soon as the provider is created it is going to be tasted
and thus opened. Initial mediasize of zero causes tasting failure
and subsequent retasting because of the size change.
MFC after: 14 days
This should allow to mount a dataset as a root filesystem even if
it belongs to a pool that is not described in zpool.cache.
This adds some overhead to the boot process though.
If the root filesystem's pool is found in zpool.cache, the by default
its cached configuration will be used for import.
vfs.zfs.rootpool.prefer_cached_config could be set to zero to force
the config to be retasted.
Discussed with: gibbs, pjd, des
MFC after: 25 days
doesn't exist on a dataset we are starting from. For example if we
have the following configuration:
tank
tank/foo
tank/foo@snap
tank/bar
tank/bar@snap
We can execute:
# zfs destroy -t tank@snap
eventhough tank@snap doesn't exit.
Unfortunately it is not possible to do the same with recursive rename:
# zfs rename -r tank@snap tank@pans
cannot open 'tank@snap': dataset does not exist
...until now. This change allows to recursively rename snapshots even if
snapshot doesn't exist on the starting dataset.
Sponsored by: rsync.net
MFC after: 2 weeks
The code builds a map of regions that were freed. On every write the
code consults the map and eventually removes ranges that were freed
before, but are now overwritten.
Freed blocks are not TRIMed immediately. There is a tunable that defines
how many txg we should wait with TRIMming freed blocks (64 by default).
There is a low priority thread that TRIMs ranges when the time comes.
During TRIM we keep in-flight ranges on a list to detect colliding
writes - we have to delay writes that collide with in-flight TRIMs in
case something will be reordered and write will reached the disk before
the TRIM. We don't have to do the same for in-flight writes, as
colliding writes just remove ranges to TRIM.
Sponsored by: multiplay.co.uk
This work includes some important fixes and some improvements obtained
from the zfsonlinux project, including TRIMming entire vdevs on pool
create/add/attach and on pool import for spare and cache vdevs.
Obtained from: zfsonlinux
Submitted by: Etienne Dechamps <etienne.dechamps@ovh.net>
Do this by checking if spa_namespace_lock is already held and not taking
it again in that case.
Add a comment explaining why that is done and why it is safe.
Reviewed by: pjd
MFC after: 24 days
Since all attribute values start at 8-byte aligned boundary, we would
previously incorrectly calculate dn_bonuslen if any attribute but the
last had a variable-length value with length not multiple of 8.
Reported by: Nicolas Rachinsky <fbsd-mas-0@ml.turing-complete.org>
Tested by: Nicolas Rachinsky <fbsd-mas-0@ml.turing-complete.org>
Reviewed by: Matthew Ahrens <mahrens@delphix.com> (for upstream)
MFC after: 2 weeks
- skip length_idx index for a replaced variable-sized attribute
- skip length_idx index for a removed variable-sized attribute
- also re-arranged code to make sure that length_idx is always
incremented for variable-sized attributes
- additionally add an assertion that the number of actually produced
attributes is the same as the expected number of resulting
attributes
In cooperation with: Matthew Ahrens <mahrens@delphix.com>
Tested by: Trent Nelson <trent@snakebite.org>
Reviewed by: Matthew Ahrens <mahrens@delphix.com> (for upstream)
To do: get this upstreamed
MFC after: 2 weeks
When calling a revoke(2) on a dtrace device, dtrace_close() could be
called, even if threads are still stuck in the device. Defer the actual
deallocation of datastructures to the cdevpriv destructor.
While there, remove the unneeded D_TRACKCLOSE and D_NEEDMINOR flags. For
the helper device, we never need it. For the regular dtrace devices, we
only need these flags on FreeBSD pre-8.
MFC after: 1 month
1) It is not useful to call "devfs_clear_cdevpriv()" from
"d_close" callbacks, hence for example read, write, ioctl and
so on might be sleeping at the time of "d_close" being called
and then then freed private data can still be accessed.
Examples: dtrace, linux_compat, ksyms (all fixed by this patch)
2) In sys/dev/drm* there are some cases in which memory will
be freed twice, if open fails, first by code in the open
routine, secondly by the cdevpriv destructor. Move registration
of the cdevpriv to the end of the drm open routines.
3) devfs_clear_cdevpriv() is not called if the "d_open" callback
registered cdevpriv data and the "d_open" callback function
returned an error. Fix this.
Discussed with: phk
MFC after: 2 weeks
Update DTrace disassembler accordingly. The code to treat the prefixes
as null prefixes was already in place.
Although in practice compilers seem to generate only cs-prefix for use
in long NOPs, the same treatment is applied to all of cs, ds, es, ss for
consistency.
Reported by: emaste
Tested by: emaste
Obtained from: Illumos commit 13442:4adbe6de60c8 (+ local changes)
MFC after: 5 days
According to the AMD manual the whole range from 0x09 to 0x1f are NOPs.
Intel manual mentions only 0x1f. Use only Intel one for now, it seems
to be the one actually generated by compilers.
Use gdb mnemonic for the operation: "nopw".
[1] AMD64 Architecture Programmer's Manual
Volume 3: General-Purpose and System Instructions
[2] Software Optimization Guide for AMD Family 10h Processors
[3] Intel(R) 64 and IA-32 Architectures Software Developer’s Manual
Volume 2 (2A, 2B & 2C): Instruction Set Reference, A-Z
Tested by: Fabian Keil <freebsd-listen@fabiankeil.de> (earlier version)
MFC after: 3 days
Bryan Cantrill implemented the equivalent of semi-log graph
paper for Dtrace so llquantize will use one logarithmic and
one linear scale.
Special thanks to Mark Peek for providing fix to an
assertion and to Fabian Keill for testing the port.
Illumos Revision: 13355:15b74a2a9a9d
Reference:
https://www.illumos/issues/905
Obtained from: Illumos
Tested by: Fabian Keill, mp
MFC after: 4 days
differentiate between an incremental and full stream.
Be sure not to generate guid equal to 0.
Reported by: someone who saw 0 being generated as 64bit random guid
MFC after: 3 days
The skew calculation here is exactly backwards. We were able to repro
it on a multi-package ESX server running a FreeBSD VM, where the TSCs
can be pretty evil.
MFC after: 1 week
Submitted by: Jeff Ford <jeffrey.ford2@isilon.com>
Reviewed by: avg, gnn
1948 zpool list should show more detailed pool information
Display per-vdev information with "zpool list -v".
The added expandsize property has currently no value on FreeBSD.
This changeset allows adding expansion support to individual vdevs
in the future.
References:
https://www.illumos.org/issues/1948
Obtained from: illumos (issue #1948)
MFC after: 2 weeks
vn_rlimit_fsize takes uio->uio_offset and uio->uio_resid into account
when determining whether given write would exceed RLIMIT_FSIZE.
When APPEND flag is specified, ZFS updates uio->uio_offset to point to the
end of file.
But this happens after a call to vn_rlimit_fsize, so vn_rlimit_fsize check
can be rendered ineffective by thread that opens some file with O_APPEND
and lseeks below RLIMIT_FSIZE before calling write.
Submitted by: Mateusz Guzik <mjguzik at gmail dot com>
MFC after: 2 weeks
2703 add mechanism to report ZFS send progress
If the zfs send command is used with the -v flag, the amount of bytes
transmitted is reported in per second updates.
References:
https://www.illumos.org/issues/2703
Obtained from: illumos (issue #2703)
MFC after: 2 weeks
to follow the example of OpenSolaris and its descendants, which implemented
cpu as an inline that took a value out of curthread. At certain points in
the FreeBSD scheduler curthread->td_oncpu will no longer be valid (in
particukar, just before the thread gets descheduled) so instead I have
implemented this as its own built-in variable.
Sponsored by: Sandvine Inc.
MFC after: 1 week
accesses of the cache member of vm_object objects.
- Use novel vm_page_is_cached() for checks outside of the vm subsystem.
Reviewed by: alc
MFC after: 2 weeks
X-MFC: r234039
On FreeBSD the direct ioctl argument is automatically copied in/out
as necesary by the kernel ioctl entry point.
PR: kern/164445
Submitted by: Luis Garces-Erice <lge@ieee.org>
Tested by: Attila Nagy <bra@fsn.hu>
MFC after: 5 days
loaded and unloaded, also have sdt.ko register callbacks with kern_sdt.c
that will be called when a newly loaded KLD module adds more probes or
a module with probes is unloaded.
This fixes two issues: first, if a module with SDT probes was loaded after
sdt.ko was loaded, those new probes would not be available in DTrace.
Second, if a module with SDT probes was unloaded while sdt.ko was loaded,
the kernel would panic the next time DTrace had cause to try and do
anything with the no-longer-existent probes.
This makes it possible to create SDT probes in KLD modules, although there
are still two caveats: first, any SDT probes in a KLD module must be part
of a DTrace provider that is defined in that module. At present DTrace
only destroys probes when the provider is destroyed, so you can still
panic the system if a KLD module creates new probes in a provider from a
different module(including the kernel) and then unload the the first module.
Second, the system will panic if you unload a module containing SDT probes
while there is an active D script that has enabled those probes.
MFC after: 1 month
- header and stub .c file for fasttrap module. It's not supported on
MIPS yet, but there is no way to disable support completely
- Do as amd64 trying to limit allocated memory
allow.mount.zfs:
allow mounting the zfs filesystem inside a jail
This way the permssions for mounting all current VFCF_JAIL filesystems
inside a jail are controlled wia allow.mount.* jail parameters.
Update sysctl descriptions.
Update jail(8) and zfs(8) manpages.
TODO: document the connection of allow.mount.* and VFCF_JAIL for kernel
developers
MFC after: 10 days
Add the sysctl debug.iosize_max_clamp, enabled by default. Setting the
sysctl to zero allows to perform the SSIZE_MAX-sized i/o requests from
the usermode.
Discussed with: bde, das (previous versions)
MFC after: 1 month
excluding other allocations including UMA now entails the addition of
a single flag to kmem_alloc or uma zone create
Reviewed by: alc, avg
MFC after: 2 weeks
names and wants to sort them by name, ie. when executes:
# zfs list -t snapshot -o name -s name
Because only name is needed we don't have to read all snapshot properties.
Below you can find how long does it take to list 34509 snapshots from a single
disk pool before and after this change with cold and warm cache:
before:
# time zfs list -t snapshot -o name -s name > /dev/null
cold cache: 525s
warm cache: 218s
after:
# time zfs list -t snapshot -o name -s name > /dev/null
cold cache: 1.7s
warm cache: 1.1s
MFC after: 1 week
the allocated memory before calling mtx_init(9) on mtx pointing to it.
Otherwize, random contents of uninitialized memory might occasionally
trigger the assertion.
Reported by: Pavel Polyakov <bsd kobyla org>
Reviewed by: pjd
MFC after: 1 week
the number of links against LINK_MAX (which is INT16_MAX), not against
UINT32_MAX. Otherwise, the constant would implicitly be converted to
-1.
Reviewed by: pjd
MFC after: 1 week
(ss == NULL) on pool import. I had such a panic recently. With current version
of ZFS it is still possible to import the pool in readonly mode and backup
all the data, but in case it is impossible for some reason add tunable
vfs.zfs.space_map_last_hope, which when set to '1' will tell ZFS to remove
colliding range and retry. This seems to have worked for me, but I consider
it highly risky to use.
MFC after: 1 week
lock_success/lock_failure, introduced in r228424, by directly skipping
in dtrace_probe.
This mainly helps in avoiding namespace pollution and thus lockstat.h
dependency by systm.h.
As an added bonus, this also helps in MFC case.
Reviewed by: avg
MFC after: 3 months (or never)
X-MFC: r228424
nullfs. The problem is that resulting vnode is only required to be
held on return from the successfull call to vop, instead of being
referenced.
Nullfs VOP_INACTIVE() method reclaims the vnode, which in combination
with the VOP_VPTOCNP() interface means that the directory vnode
returned from VOP_VPTOCNP() is reclaimed in advance, causing
vn_fullpath() to error with EBADF or like.
Change the interface for VOP_VPTOCNP(), now the dvp must be
referenced. Convert all in-tree implementations of VOP_VPTOCNP(),
which is trivial, because vhold(9) and vref(9) are similar in the
locking prerequisites. Out-of-tree fs implementation of VOP_VPTOCNP(),
if any, should have no trouble with the fix.
Tested by: pho
Reviewed by: mckusick
MFC after: 3 weeks (subject of re approval)
certain instructions in a function prologue or epilogue. DTrace has a
hook into the invalid opcode fault handler that checks whether the fault
was due to an probe and if so, runs the DTrace magic.
Upon returning from an invalid opcode fault caused by a probe, DTrace must
emulate the instruction that was replaced with the invalid opcode and then
return control to the instruction following the invalid opcode.
There were a pair of related bugs in the emulation for the leave
instruction. The leave instruction is used to pop off a stack frame prior
to returning from a function. The emulation for this instruction must
move the trap frame for the invalid opcode fault down the stack to the
bottom of the stack frame that is being removed, and then execute an iret.
At two points in this process, the emulation code was storing values above
the current value of the stack pointer. This opened up a window in which
if we were two take an interrupt, the trap frame for the interrupt would
overwrite the values stored on the stack, causing the system to panic
later.
The first bug was that at one point the emulation code saves the new value
for $esp above the current stack pointer value. The fix is to save this
value instead inside of the original trap frame. At this point we do
not need the original trap frame so this is safe.
The second bug is that when the emulate code loads $esp from the stack, it
points part-way through the new trap frame instead of at its beginning.
The emulation code adjusts the stack pointer to the correct value
immediately afterwards, but this still leaves a one instruction window in
which an interrupt would corrupt this trap frame. Fix this by adjusting
the stack frame value before loading it into $esp.
This fixes panics in invop_leave on i386 when using fbt return probes.
Reviewed by: rpaulo, attilio
MFC after: 1 week
ZFS is trying to open and taste ZVOL as its VDEV. This is not supported,
so return an error instead of panicing on spa_namespace_lock recursion.
Reported by: Robert Millan <rmh@debian.org>
PR: kern/162008
MFC after: 3 days
of the CDDL licence explicitly requires every Contributor to add
a copyright notice.
This also reflects the copyright notices for the changes recently
added by Illumos.
MFC after: 3 days
It is possible for file systems with 'mountpoint' preperty set to 'legacy'
or 'none' - we don't have to change mount directory for them.
Currently such file systems are unmounted on rename and not even mounted back.
This introduces layering violation, as we need to update 'f_mntfromname'
field in statfs structure related to mountpoint (for the dataset we are
renaming and all its children).
In my opinion it is worth it, as it allow to update FreeBSD in even cleaner
way - in ZFS-only configuration root file system is ZFS file system with
'mountpoint' property set to 'legacy'. If root dataset is named system/rootfs,
we can snapshot it (system/rootfs@upgrade), clone it (system/oldrootfs),
update FreeBSD and if it doesn't boot we can boot back from system/oldrootfs
and rename it back to system/rootfs while it is mounted as /. Before it was
not possible, because unmounting / was not possible.
MFC after: 2 weeks
This allows to see processes I/O activity in 'top -m io' output.
PR kern/156218
Reported by: Marcus Reid <marcus@blazingdot.com>
Patch by: avg
MFC after: 3 days
- Decompress assembled gang block data if compressed.
- Verify checksum of a gang header.
- Verify checksum of assembled gang block data.
- Verify checksum of uber block.
Submitted by: avg
MFC after: 3 days
physical block size declared in bp may not always be what we want.
For example in case of gang block header physical block size declared
in bp is much larger than SPA_GANGBLOCKSIZE (512 bytes) and checksum
calculation failed. This bug could lead to accessing unallocated
memory and resets/failures during boot.
MFC after: 3 days
When calculating space needed for SA_BONUS buffers,
hdrsize is always rounded up to next 8-aligned boundary.
However, in two places the round up was done against
sum of 'total' plus hdrsize. On the other hand,
hdrsize increments by 4 each time, which means in
certain conditions, we would end up returning with
will_spill == 0 and (total + hdrsize) larger than
full_space, leading to a failed assertion because
it's invalid for dmu_set_bonus.
Sponsored by: iXsystems, Inc.
Reviewed by: mm
MFC after: 3 days
patch modifies makesyscalls.sh to prefix all of the non-compatibility
calls (e.g. not linux_, freebsd32_) with sys_ and updates the kernel
entry points and all places in the code that use them. It also
fixes an additional name space collision between the kernel function
psignal and the libc function of the same name by renaming the kernel
psignal kern_psignal(). By introducing this change now we will ease future
MFCs that change syscalls.
Reviewed by: rwatson
Approved by: re (bz)
flags field. Updates to the atomic flags are performed using the atomic
ops on the containing word, do not require any vm lock to be held, and
are non-blocking. The vm_page_aflag_set(9) and vm_page_aflag_clear(9)
functions are provided to modify afalgs.
Document the changes to flags field to only require the page lock.
Introduce vm_page_reference(9) function to provide a stable KPI and
KBI for filesystems like tmpfs and zfs which need to mark a page as
referenced.
Reviewed by: alc, attilio
Tested by: marius, flo (sparc64); andreast (powerpc, powerpc64)
Approved by: re (bz)
Remove mapped pages for all dataset vnodes in zfs_rezget() using
new vn_pages_remove() to fix mmapped files changed by
zfs rollback or zfs receive -F.
PR: kern/160035, kern/156933
Reviewed by: kib, pjd
Approved by: re (kib)
MFC after: 1 week
zvol.c: fix calling of dmu_objset_prefetch() in zvol_create_minors()
by passing full instead of relative dataset name and prefetching all
visible datasets to be processed later instead of just the pool name
Reviewed by: pjd
Approved by: re (kib)
MFC after: 1 week
> Reviewed by: If someone else reviewed your modification.
> Approved by: If you needed approval for this commit.
> Obtained from: If the change is from a third party.
> MFC after: N [day[s]|week[s]|month[s]]. Request a reminder email.
> Security: Vulnerability reference (one per line) or description.
> Empty fields above will be automatically removed.
M opensolaris/uts/common/fs/zfs/zfs_ioctl.c
M opensolaris/uts/common/fs/zfs/zvol.c
zfs_ioc_dataset_list_next() and dsl_dir_destroy_check() indirectly
invoked from dmu_recv_existing_end() via dsl_dataset_destroy() by not
prefetching temporary clones, as these count as always inconsistent.
In addition, do not prefetch hidden datasets at all as we are not
going to process these later.
Filed as Illumos Bug #1346
PR: kern/157728
Tested by: Borja Marcos <borjam@sarenet.es>, mm
Reviewed by: pjd
Approved by: re (kib)
MFC after: 1 week
spa_namespace_lock. This fixes LOR between the spa_namespace_lock and
spa_config lock. LOR can cause deadlock on vdevs removal/insertion.
Reported by: gibbs, delphij
Tested by: delphij
Approved by: re (kib)
MFC after: 1 week
kernel for FreeBSD 9.0:
Add a new capability mask argument to fget(9) and friends, allowing system
call code to declare what capabilities are required when an integer file
descriptor is converted into an in-kernel struct file *. With options
CAPABILITIES compiled into the kernel, this enforces capability
protection; without, this change is effectively a no-op.
Some cases require special handling, such as mmap(2), which must preserve
information about the maximum rights at the time of mapping in the memory
map so that they can later be enforced in mprotect(2) -- this is done by
narrowing the rights in the existing max_protection field used for similar
purposes with file permissions.
In namei(9), we assert that the code is not reached from within capability
mode, as we're not yet ready to enforce namespace capabilities there.
This will follow in a later commit.
Update two capability names: CAP_EVENT and CAP_KEVENT become
CAP_POST_KEVENT and CAP_POLL_KEVENT to more accurately indicate what they
represent.
Approved by: re (bz)
Submitted by: jonathan
Sponsored by: Google Inc
zfsvfs->z_log before calling zil_commit(). [1]
Do not call zfs_read() from zfs_getextattr() with the IO_SYNC flag.
Submitted by: Alexander Zagrebin <alex@zagrebin.ru> [1]
Reviewed by: pjd@
Approved by: re (kib)
MFC after: 3 days
in the case of a held dataset during remount.
Detailed description is available at:
https://www.illumos.org/issues/883
illumos-gate revision: 13380:161b964a0e10
Reviewed by: pjd
Approved by: re (kib)
Obtained from: Illumos (Bug #883)
MFC after: 3 days
which does not require change in the znode structure.
Specifically, it queries rdev from the znode in the
same sa_bulk_lookup already done in zfs_getattr().
Submitted by: pjd (with some revisions)
Reviewed by: pjd, mm
Approved by: re (kib)
devices are imbalanced zfs will lots of CPU searching for space on devices
which tend to be pretty full. It should instead fail quickly on the full
devices and move onto devices which have more availability.
New loader tunable: vfs.zfs.mg_alloc_failures (min = 8)
Illumos-gate changeset: 13379:4df42cc92254
Obtained from: Illumos (Bug #1051)
MFC after: 2 weeks
mask of CPUs, pc_other_cpus and pc_cpumask become highly inefficient.
Remove them and replace their usage with custom pc_cpuid magic (as,
atm, pc_cpumask can be easilly represented by (1 << pc_cpuid) and
pc_other_cpus by (all_cpus & ~(1 << pc_cpuid))).
This change is not targeted for MFC because of struct pcpu members
removal and dependency by cpumask_t retirement.
MD review by: marcel, marius, alc
Tested by: pluknet
MD testing by: marcel, marius, gonzo, andreast
For snapshots, this is the same as COMPRESSRATIO, but for
filesystems/volumes, the COMPRESSRATIO is based on the data "USED" (ie,
includes blocks in children, but not blocks shared with the origin).
This is needed to figure out how much space a filesystem would use if it
were not compressed (ignoring snapshots).
Illumos-gate revision: 13387
Obtained from: Illumos (Feature #1092)
MFC after: 2 weeks
The vdev cache is very underutilized (hit ratio 30%-70%) and may consume
excessive memory on systems with many vdevs.
Illumos-gate revision: 13346
Obtained from: Illumos (Bug #175)
MFC after: 1 week
OpenSolaris and ZFS header files. These changes are sufficient
to allow a C++ program to use the libzfs library.
Note: The majority of these files already included 'extern "C"'
declarations, so the intention of providing C++ compatibility
already existed even if it wasn't provided.
cddl/compat/opensolaris/include/assert.h:
Wrap our compatibility assert implementation in
'extern "C"'. Since this is a compatibility header
I matched the Solaris style of doing this explicitly
rather than rely on FreeBSD's __BEGIN/END_DECLS macro.
sys/cddl/compat/opensolaris/sys/kstat.h:
sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/arc.h:
sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dsl_pool.h:
sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/ddt.h:
sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/spa.h:
sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h:
Rename parameters in function declarations that conflict
with C++ keywords. This was the solution preferred by
members of the Illumos community.
sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_ioctl.h:
In C, nested structures are visible in the global namespace,
but in C++, they take on the namespace of the structure in
which they are contained. Flatten nested structure
definitions within struct zfs_cmd so these structures are
visible in the global namespace when compiled in both
languages.
Sponsored by: Spectra Logic Corporation
User upgrades his system to fix the problem, but if he has any ZFS snapshots
for the file system which contains problematic binary, any user can mount the
snapshot and execute vulnerable binary.
Prevent this from happening by always mounting snapshots with setuid turned off.
MFC after: 2 weeks
The task structure might be no longer available.
This also allows to eliminates the need for two tasks in the zio structure.
Submitted by: anonymous
MFC after: 2 weeks
method, so that callers can indicate the minimum vnode
locking requirement. This will allow some file systems to choose
to return a LK_SHARED locked vnode when LK_SHARED is specified
for the flags argument. This patch only adds the flag. It
does not change any file system to use it and all callers
specify LK_EXCLUSIVE, so file system semantics are not changed.
Reviewed by: kib
(zfs destroy -r pool/dataset@snapshot)
To destroy all descendent snapshots with the same name the top level
snapshot was not required to exist. So if the top level snapshot does
not exist, check permissions of the parent dataset instead.
Filed as Illumos Bug #1043
Reviewed by: delphij
Approved by: pjd
MFC after: together with v28
Now in the case when one-shot timers are used cyclic events should fire
closer to theier scheduled times. As the cyclic is currently used only
to drive DTrace profile provider, this is the area where the change
makes a difference.
Reviewed by: mav (earlier version, a while ago)
X-MFC after: clocksource/eventtimer subsystem
cpuset_t objects.
That is going to offer the underlying support for a simple bump of
MAXCPU and then support for number of cpus > 32 (as it is today).
Right now, cpumask_t is an int, 32 bits on all our supported architecture.
cpumask_t on the other side is implemented as an array of longs, and
easilly extendible by definition.
The architectures touched by this commit are the following:
- amd64
- i386
- pc98
- arm
- ia64
- XEN
while the others are still missing.
Userland is believed to be fully converted with the changes contained
here.
Some technical notes:
- This commit may be considered an ABI nop for all the architectures
different from amd64 and ia64 (and sparc64 in the future)
- per-cpu members, which are now converted to cpuset_t, needs to be
accessed avoiding migration, because the size of cpuset_t should be
considered unknown
- size of cpuset_t objects is different from kernel and userland (this is
primirally done in order to leave some more space in userland to cope
with KBI extensions). If you need to access kernel cpuset_t from the
userland please refer to example in this patch on how to do that
correctly (kgdb may be a good source, for example).
- Support for other architectures is going to be added soon
- Only MAXCPU for amd64 is bumped now
The patch has been tested by sbruno and Nicholas Esborn on opteron
4 x 12 pack CPUs. More testing on big SMP is expected to came soon.
pluknet tested the patch with his 8-ways on both amd64 and i386.
Tested by: pluknet, sbruno, gianni, Nicholas Esborn
Reviewed by: jeff, jhb, sbruno
safer for i386 because it can be easily over 4 GHz now. More worse, it can
be easily changed by user with 'machdep.tsc_freq' tunable (directly) or
cpufreq(4) (indirectly). Note it is intentionally not used in performance
critical paths to avoid performance regression (but we should, in theory).
Alternatively, we may add "virtual TSC" with lower frequency if maximum
frequency overflows 32 bits (and ignore possible incoherency as we do now).
VFS where we know if this is truncate(2) or ftruncate(2). If this is the
latter we should depend on the mode the file was opened and not on the current
permission.
PR: standards/154873
Reported by: Mark Martinec <Mark.Martinec@ijs.si>
Discussed with: Eric Schrock <eric.schrock@delphix.com>
Discussed with: Mark Maybee <Mark.Maybee@Oracle.COM>
MFC after: 1 month
Add systrace_linux32 and systrace_freebsd32 modules which provide
support for tracing compat system calls in addition to native system
call tracing provided by systrace module.
Provided that all the systrace modules are loaded now you can select
what syscalls to trace in the following manner:
syscall::xxx:yyy - work on all system calls that match the specification
syscall:freebsd:xxx:yyy - only native system calls
syscall:linux32:xxx:yyy - linux32 compat system calls
syscall:freebsd32:xxx:yyy - freebsd32 compat system calls on amd64
PR: kern/152822
Submitted by: Artem Belevich <fbsdlist@src.cx>
Reviewed by: jhb (earlier version)
MFC after: 3 weeks
Few new things available from now on:
- Data deduplication.
- Triple parity RAIDZ (RAIDZ3).
- zfs diff.
- zpool split.
- Snapshot holds.
- zpool import -F. Allows to rewind corrupted pool to earlier
transaction group.
- Possibility to import pool in read-only mode.
MFC after: 1 month
