nfsd uses do_readv_writev() to implement fops->read and fops->write.
do_readv_writev() will attempt to read/write using fops->aio_read and
fops->aio_write, but it will fallback to fops->read and fops->write when
AIO is not available. However, the fallback will perform a call for each
individual data page. Since our default recordsize is 128KB, sequential
operations on NFS will generate 32 DMU transactions where only 1
transaction was needed. That was unnecessary overhead and we implement
fops->aio_read and fops->aio_write to eliminate it.
ZFS originated in OpenSolaris, where the AIO API is entirely implemented
in userland's libc by intelligently mapping them to VOP_WRITE, VOP_READ
and VOP_FSYNC. Linux implements AIO inside the kernel itself. Linux
filesystems therefore must implement their own AIO logic and nearly all
of them implement fops->aio_write synchronously. Consequently, they do
not implement aio_fsync(). However, since the ZPL works by mapping
Linux's VFS calls to the functions implementing Illumos' VFS operations,
we instead implement AIO in the kernel by mapping the operations to the
VOP_READ, VOP_WRITE and VOP_FSYNC equivalents. We therefore implement
fops->aio_fsync.
One might be inclined to make our fops->aio_write implementation
synchronous to make software that expects this behavior safe. However,
there are several reasons not to do this:
1. Other platforms do not implement aio_write() synchronously and since
the majority of userland software using AIO should be cross platform,
expectations of synchronous behavior should not be a problem.
2. We would hurt the performance of programs that use POSIX interfaces
properly while simultaneously encouraging the creation of more
non-compliant software.
3. The broader community concluded that userland software should be
patched to properly use POSIX interfaces instead of implementing hacks
in filesystems to cater to broken software. This concept is best
described as the O_PONIES debate.
4. Making an asynchronous write synchronous is non sequitur.
Any software dependent on synchronous aio_write behavior will suffer
data loss on ZFSOnLinux in a kernel panic / system failure of at most
zfs_txg_timeout seconds, which by default is 5 seconds. This seems like
a reasonable consequence of using non-compliant software.
It should be noted that this is also a problem in the kernel itself
where nfsd does not pass O_SYNC on files opened with it and instead
relies on a open()/write()/close() to enforce synchronous behavior when
the flush is only guarenteed on last close.
Exporting any filesystem that does not implement AIO via NFS risks data
loss in the event of a kernel panic / system failure when something else
is also accessing the file. Exporting any file system that implements
AIO the way this patch does bears similar risk. However, it seems
reasonable to forgo crippling our AIO implementation in favor of
developing patches to fix this problem in Linux's nfsd for the reasons
stated earlier. In the interim, the risk will remain. Failing to
implement AIO will not change the problem that nfsd created, so there is
no reason for nfsd's mistake to block our implementation of AIO.
It also should be noted that `aio_cancel()` will always return
`AIO_NOTCANCELED` under this implementation. It is possible to implement
aio_cancel by deferring work to taskqs and use `kiocb_set_cancel_fn()`
to set a callback function for cancelling work sent to taskqs, but the
simpler approach is allowed by the specification:
```
Which operations are cancelable is implementation-defined.
```
http://pubs.opengroup.org/onlinepubs/009695399/functions/aio_cancel.html
The only programs on my system that are capable of using `aio_cancel()`
are QEMU, beecrypt and fio use it according to a recursive grep of my
system's `/usr/src/debug`. That suggests that `aio_cancel()` users are
rare. Implementing aio_cancel() is left to a future date when it is
clear that there are consumers that benefit from its implementation to
justify the work.
Lastly, it is important to know that handling of the iovec updates differs
between Illumos and Linux in the implementation of read/write. On Linux,
it is the VFS' responsibility whle on Illumos, it is the filesystem's
responsibility. We take the intermediate solution of copying the iovec
so that the ZFS code can update it like on Solaris while leaving the
originals alone. This imposes some overhead. We could always revisit
this should profiling show that the allocations are a problem.
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#223Closes#2373
Some nvlist_t could be leaked in error handling paths.
Also make sure cb argument to zfs_zevent_post() cannnot
be NULL.
Signed-off-by: Isaac Huang <he.huang@intel.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2158
4631 zvol_get_stats triggering too many reads
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Sebastien Roy <sebastien.roy@delphix.com>
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Approved by: Dan McDonald <danmcd@omniti.com>
References:
https://www.illumos.org/issues/4631https://github.com/illumos/illumos-gate/commit/bbfa8ea
Ported-by: Boris Protopopov <bprotopopov@hotmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2612Closes#2480
4976 zfs should only avoid writing to a failing non-redundant top-level vdev
4978 ztest fails in get_metaslab_refcount()
4979 extend free space histogram to device and pool
4980 metaslabs should have a fragmentation metric
4981 remove fragmented ops vector from block allocator
4982 space_map object should proactively upgrade when feature is enabled
4983 need to collect metaslab information via mdb
4984 device selection should use fragmentation metric
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Adam Leventhal <adam.leventhal@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
References:
https://www.illumos.org/issues/4976https://www.illumos.org/issues/4978https://www.illumos.org/issues/4979https://www.illumos.org/issues/4980https://www.illumos.org/issues/4981https://www.illumos.org/issues/4982https://www.illumos.org/issues/4983https://www.illumos.org/issues/4984https://github.com/illumos/illumos-gate/commit/2e4c998
Notes:
The "zdb -M" option has been re-tasked to display the new metaslab
fragmentation metric and the new "zdb -I" option is used to control
the maximum number of in-flight I/Os.
The new fragmentation metric is derived from the space map histogram
which has been rolled up to the vdev and pool level and is presented
to the user via "zpool list".
Add a number of module parameters related to the new metaslab weighting
logic.
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2595
Add a new 'overlay' property (default 'off') that controls whether the
filesystem should be mounted even if the mountpoint is busy or if it
should fail with a 'mountpoint not empty'.
Doing overlay mounts is the default mount behavior on Linux, but not
in ZFS. It have been decided that following the ZFS behavior should
be the default, but this overlay allows for site administrator to
override this decision on a per-dataset basis.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes: #2503
Most of the code base already uses va_list, which is specified by
iso-c. gcc/glibc provides 'typedef __gnuc_va_list va_list'. and
when not using gcc/glibc we can't expect to find __gnuc_va_list.
Signed-off-by: Alec Salazar <alec.j.salazar@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2588
Handle all iputs in zfs_purgedir() and zfs_inode_destroy()
asynchronously to prevent deadlocks. When the iputs are allowed
to run synchronously in the destroy call path deadlocks between
xattr directory inodes and their parent file inodes are possible.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Richard Yao <ryao@gentoo.org>
Closes#457
As part of commit e8b96c6 the search zio used by the
vdev_queue_io_to_issue() function was moved to the heap
to minimize stack usage. Functionally this is fine, but
to maximize performance it's best to minimize the number
of dynamic allocations.
To avoid this allocation temporary space for the search
zio has been reserved in the vdev_queue structure. All
access must be serialized through the vq_lock.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Closes#2572
4914 zfs on-disk bookmark structure should be named *_phys_t
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Richard Lowe <richlowe@richlowe.net>
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Approved by: Robert Mustacchi <rm@joyent.com>
References:
https://www.illumos.org/issues/4914https://github.com/illumos/illumos-gate/commit/7802d7b
Porting notes:
There were a number of zfsonlinux-specific uses of zbookmark_t which
needed to be updated. This should reduce the likelihood of further
problems like issue #2094 from occurring.
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2558
4390 i/o errors when deleting filesystem/zvol can lead to space map corruption
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Approved by: Dan McDonald <danmcd@omniti.com>
References:
https://www.illumos.org/issues/4390https://github.com/illumos/illumos-gate/commit/7fd05ac
Porting notes:
Previous stack-reduction efforts in traverse_visitb() caused a fair
number of un-mergable pieces of code. This patch should reduce its
stack footprint a bit more.
The new local bptree_entry_phys_t in bptree_add() is dynamically-allocated
using kmem_zalloc() for the purpose of stack reduction.
The new global zfs_free_leak_on_eio has been defined as an integer
rather than a boolean_t as was the case with the related zfs_recover
global. Also, zfs_free_leak_on_eio's definition has been inserted into
zfs_debug.c for consistency with the existing definition of zfs_recover.
Illumos placed it in spa_misc.c.
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2545
4757 ZFS embedded-data block pointers ("zero block compression")
4913 zfs release should not be subject to space checks
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Max Grossman <max.grossman@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Dan McDonald <danmcd@omniti.com>
References:
https://www.illumos.org/issues/4757https://www.illumos.org/issues/4913https://github.com/illumos/illumos-gate/commit/5d7b4d4
Porting notes:
For compatibility with the fastpath code the zio_done() function
needed to be updated. Because embedded-data block pointers do
not require DVAs to be allocated the associated vdevs will not
be marked and therefore should not be unmarked.
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2544
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Richard Lowe <richlowe@richlowe.net>
Description from Matt Ahrens's bug report at Delphix:
Add a new zfs property, "redundant_metadata" which can have values
"all" or "most". The default will be "all", which is the current
behavior. Setting to "most" will cause us to only store 1 copy of
level-1 indirect blocks of user data files.
Additional notes:
The new man page section for this property states
"The exact behavior of which metadata blocks
are stored redundantly may change in future releases."
and:
"When set to most, ZFS stores an extra copy of most types of
metadata. This can improve performance of random writes,
because less metadata must be written."
The current implementation is as described above in Matt's blog.
It is controlled by a new global integer
"zfs_redundant_metadata_most_ditto_level", currently initialized
to 2. When "redundant_metadata" is set to "most", only indirect
blocks of the specified level and higher will have additional ditto
blocks created.
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2542
4754 io issued to near-full luns even after setting noalloc threshold
4755 mg_alloc_failures is no longer needed
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Dan McDonald <danmcd@omniti.com>
References:
https://www.illumos.org/issues/4754https://www.illumos.org/issues/4755https://github.com/illumos/illumos-gate/commit/b6240e8
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2533
4374 dn_free_ranges should use range_tree_t
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Max Grossman <max.grossman@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com
Reviewed by: Garrett D'Amore <garrett@damore.org>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Dan McDonald <danmcd@omniti.com>
References:
https://www.illumos.org/issues/4374https://github.com/illumos/illumos-gate/commit/bf16b11
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2531
4370 avoid transmitting holes during zfs send
4371 DMU code clean up
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Approved by: Garrett D'Amore <garrett@damore.org>a
References:
https://www.illumos.org/issues/4370https://www.illumos.org/issues/4371https://github.com/illumos/illumos-gate/commit/43466aa
Ported by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2529
4171 clean up spa_feature_*() interfaces
4172 implement extensible_dataset feature for use by other zpool features
Reviewed by: Max Grossman <max.grossman@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Jerry Jelinek <jerry.jelinek@joyent.com>
Approved by: Garrett D'Amore <garrett@damore.org>a
References:
https://www.illumos.org/issues/4171https://www.illumos.org/issues/4172https://github.com/illumos/illumos-gate/commit/2acef22
Ported-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2528
This loop in dmu_objset_write_ready():
for (i = 0; i < dnp->dn_nblkptr; i++)
bp->blk_fill += dnp->dn_blkptr[i].blk_fill;
invokes _undefined behavior_ for the (common) case of dn_nblkptr=3,
therefore, the compiler is free to do whatever it wants (such as
optimizing it away, or otherwise messing up your expections).
The fix is to be honest about the array size.
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2511Closes#2010
4101 metaslab_debug should allow for fine-grained control
4102 space_maps should store more information about themselves
4103 space map object blocksize should be increased
4105 removing a mirrored log device results in a leaked object
4106 asynchronously load metaslab
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Sebastien Roy <seb@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
Prior to this patch, space_maps were preferred solely based on the
amount of free space left in each. Unfortunately, this heuristic didn't
contain any information about the make-up of that free space, which
meant we could keep preferring and loading a highly fragmented space map
that wouldn't actually have enough contiguous space to satisfy the
allocation; then unloading that space_map and repeating the process.
This change modifies the space_map's to store additional information
about the contiguous space in the space_map, so that we can use this
information to make a better decision about which space_map to load.
This requires reallocating all space_map objects to increase their
bonus buffer size sizes enough to fit the new metadata.
The above feature can be enabled via a new feature flag introduced by
this change: com.delphix:spacemap_histogram
In addition to the above, this patch allows the space_map block size to
be increase. Currently the block size is set to be 4K in size, which has
certain implications including the following:
* 4K sector devices will not see any compression benefit
* large space_maps require more metadata on-disk
* large space_maps require more time to load (typically random reads)
Now the space_map block size can adjust as needed up to the maximum size
set via the space_map_max_blksz variable.
A bug was fixed which resulted in potentially leaking an object when
removing a mirrored log device. The previous logic for vdev_remove() did
not deal with removing top-level vdevs that are interior vdevs (i.e.
mirror) correctly. The problem would occur when removing a mirrored log
device, and result in the DTL space map object being leaked; because
top-level vdevs don't have DTL space map objects associated with them.
References:
https://www.illumos.org/issues/4101https://www.illumos.org/issues/4102https://www.illumos.org/issues/4103https://www.illumos.org/issues/4105https://www.illumos.org/issues/4106https://github.com/illumos/illumos-gate/commit/0713e23
Porting notes:
A handful of kmem_alloc() calls were converted to kmem_zalloc(). Also,
the KM_PUSHPAGE and TQ_PUSHPAGE flags were used as necessary.
Ported-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2488
Update the current code to ensure inodes are never dirtied if they are
part of a read-only file system or snapshot. If they do somehow get
dirtied an attempt will make made to write them to disk. In the case
of snapshots, which don't have a ZIL, this will result in a NULL
dereference in zil_commit().
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2405
Restructure the zfsdev_state_list to allow for lock-free reading by
converting to a simple singly-linked list from which items are never
deleted and over which only forward iterations are performed. It depends
on, among other things, the atomicity of accessing the zs_minor integer
and zs_next pointer.
This fixes a lock inversion in which the zfsdev_state_lock is used by
both the sync task (txg_sync) and indirectly by any user program which
uses /dev/zfs; the zfsdev_release method uses the same lock and then
blocks on the sync task.
The most typical failure scenerio occurs when the sync task is cleaning
up a user hold while various concurrent "zfs" commands are in progress.
Neither Illumos nor Solaris are affected by this issue because they use
DDI interface which provides lock-free reading of device state via the
ddi_get_soft_state() function.
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2301
Originally, vdev_file used system_taskq. This would cause a deadlock,
especially on system with few CPUs. The reason is that the prefetcher
threads, which are on system_taskq, will sometimes be blocked waiting
for I/O to finish. If the prefetcher threads consume all the tasks in
system_taskq, the I/O cannot be served and thus results in a deadlock.
We fix this by creating a dedicated vdev_file_taskq for vdev_file I/O.
Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2270
When fetching property values of snapshots, a check against the head
dataset type must be performed. Previously, this additional check was
performed only when fetching "version", "normalize", "utf8only" or "case".
This caused the ZPL properties "acltype", "exec", "devices", "nbmand",
"setuid" and "xattr" to be erroneously displayed with meaningless values
for snapshots of volumes. It also did not allow for the display of
"volsize" of a snapshot of a volume.
This patch adds the headcheck flag paramater to zfs_prop_valid_for_type()
and zprop_valid_for_type() to indicate the check is being done
against a head dataset's type in order that properties valid only for
snapshots are handled correctly. This allows the the head check in
get_numeric_property() to be performed when fetching a property for
a snapshot.
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2265
This implements a subset of the LWP rwlock interface by wrapping the
equivalent POSIX thread interface. It is a superset of the features
needed by ztest.
The missing bits are {,_}rw_read_held() and {,_}rw_write_held().
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1970
Also, make sure we use clock_t for ddi_get_lbolt to prevent type conversion
from screwing things.
Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2142
When a vdev starts getting I/O or checksum errors it is now
possible to automatically rebuild to a hot spare device.
To cleanly support this functionality in a shell script some
additional information was added to all zevent ereports which
include a vdev. This covers both io and checksum zevents but
may be used but other scripts.
In the Illumos FMA solution the same information is required
but it is retrieved through the libzfs library interface.
Specifically the following members were added:
vdev_spare_paths - List of vdev paths for all hot spares.
vdev_spare_guids - List of vdev guids for all hot spares.
vdev_read_errors - Read errors for the problematic vdev
vdev_write_errors - Write errors for the problematic vdev
vdev_cksum_errors - Checksum errors for the problematic vdev.
By default the required hot spare scripts are installed but this
functionality is disabled. To enable hot sparing uncomment the
ZED_SPARE_ON_IO_ERRORS and ZED_SPARE_ON_CHECKSUM_ERRORS in the
/etc/zfs/zed.d/zed.rc configuration file.
These scripts do no add support for the autoexpand property. At
a minimum this requires adding a new udev rule to detect when
a new device is added to the system. It also requires that the
autoexpand policy be ported from Illumos, see:
https://github.com/illumos/illumos-gate/blob/master/usr/src/cmd/syseventd/modules/zfs_mod/zfs_mod.c
Support for detecting the correct name of a vdev when it's not
a whole disk was added by Turbo Fredriksson.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov>
Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Issue #2
zpool_events_next() can be called in blocking mode by specifying a
non-zero value for the "block" parameter. However, the design of
the ZFS Event Daemon (zed) requires additional functionality from
zpool_events_next(). Instead of adding additional arguments to the
function, it makes more sense to use flags that can be bitwise-or'd
together.
This commit replaces the zpool_events_next() int "block" parameter with
an unsigned bitwise "flags" parameter. It also defines ZEVENT_NONE
to specify the default behavior. Since non-blocking mode can be
specified with the existing ZEVENT_NONBLOCK flag, the default behavior
becomes blocking mode. This, in effect, inverts the previous use
of the "block" parameter. Existing callers of zpool_events_next()
have been modified to check for the ZEVENT_NONBLOCK flag.
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2
The ZFS_IOC_EVENTS_SEEK ioctl was added to allow user space callers
to seek around the zevent file descriptor by EID. When a specific
EID is passed and it exists the cursor will be positioned there.
If the EID is no longer cached by the kernel ENOENT is returned.
The caller may also pass ZEVENT_SEEK_START or ZEVENT_SEEK_END to seek
to those respective locations.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov>
Issue #2
Tagging each zevent with a unique monotonically increasing EID
(Event IDentifier) provides the required infrastructure for a user
space daemon to reliably process zevents. By writing the EID to
persistent storage the daemon can safely resume where it left off
in the event stream when it's restarted.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov>
Issue #2
Originally, users had to handle spa namespace collisions by either
exporting the already imported pool or by specifying a new name for the
pool with a conflicting name. In the case of root pools from virtual
guests, neither approach to collision resolution is reasonable. This is
addressed by extending the new name syntax with a -t option to specify
that the new name is temporary. When specified, this sets an internal
flag that is passed into the kernel to tell it that all label updates
should refer to the name used in the original label. Consequently, the
original pool name will be retained on export.
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2189
The nreserved column in the txgs kstat file always contains 0
following the write throttle restructuring of commit
e8b96c6007.
Prior to that commit, the nreserved column showed the number of bytes
temporarily reserved in the pool by a transaction group at sync time.
The new write throttle did away with temporary reservations and uses
the amount of dirty data instead. To approximate the old output of
the txgs kstat, the number of dirty bytes per-txg was passed in as
the nreserved value to spa_txg_history_set_io(). This approach did
not work as intended, because the per-txg dirty value is decremented
as data is written out to disk, so it is zero by the time we call
spa_txg_history_set_io(). To fix this, save the number of dirty
bytes before calling spa_sync(), and pass this value in to
spa_txg_history_set_io().
Also, since the name "nreserved" is now a misnomer, the column
heading is now labeled "ndirty".
Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1696
A few counters in the dmu_tx kstats are obsolete or no longer
bumped properly.
- The sync task restructuring commit
13fe019870 removed the code
that bumpted dmu_tx_quota. The counter is now bumped in two
cases, instead of just the one case as before (after the result
of dsl_dataset_check_quota call). The second case is where
we check the requested reservation against the actual pool size,
as this is an implicit quota of sorts.
- The write throttle restructuring commit
e8b96c6007 makes dmu_tx_how and
dmu_tx_inflight obsolete, so they are removed.
Signed-off-by: Kohsuke Kawaguchi <kk@kohsuke.org>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1914
Previously, the "data_size" field in the arcstats kstat contained the
amount of cached "metadata" and "data" in the ARC. The problem is this
then made it difficult to extract out just the "metadata" size, or just
the "data" size.
To make it easier to distinguish the two values, "data_size" has been
modified to count only buffers of type ARC_BUFC_DATA, and "meta_size"
was added to count only buffers of type ARC_BUFC_METADATA. If one wants
the old "data_size" value, simply sum the new "data_size" and
"meta_size" values.
Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
To maintain a strict limit on the metadata contained in the arc, while
preventing the arc buffer headers from completely consuming the
"arc_meta_used" space, we need to evict metadata buffers from the arc's
ghost lists along with the regular lists.
This change modifies arc_adjust_meta such that it more closely models
the adjustments made in arc_adjust. "arc_meta_used" is used similarly to
"arc_size", and "arc_meta_limit" is used similarly to "arc_c".
Testing metadata intensive workloads (e.g. creating, copying, and
removing millions of small files and/or directories) has shown this
change to make a dramatic improvement to the hit rate maintained in the
arc. While I think there is still room for improvement, this is a big
step in the right direction.
In addition, zpl_free_cached_objects was made into a no-op as I'm not
yet sure how to properly implement that function.
Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
ZoL commit 1421c89 unintentionally changed the disk format in a forward-
compatible, but not backward compatible way. This was accomplished by
adding an entry to zbookmark_t, which is included in a couple of
on-disk structures. That lead to the creation of pools with incorrect
dsl_scan_phys_t objects that could only be imported by versions of ZoL
containing that commit. Such pools cannot be imported by other versions
of ZFS or past versions of ZoL.
The additional field has been removed by the previous commit. However,
affected pools must be imported and scrubbed using a version of ZoL with
this commit applied. This will return the pools to a state in which they
may be imported by other implementations.
The 'zpool import' or 'zpool status' command can be used to determine if
a pool is impacted. A message similar to one of the following means your
pool must be scrubbed to restore compatibility.
$ zpool import
pool: zol-0.6.2-173
id: 1165955789558693437
state: ONLINE
status: Errata #1 detected.
action: The pool can be imported using its name or numeric identifier,
however there is a compatibility issue which should be corrected
by running 'zpool scrub'
see: http://zfsonlinux.org/msg/ZFS-8000-ER
config:
...
$ zpool status
pool: zol-0.6.2-173
state: ONLINE
scan: pool compatibility issue detected.
see: https://github.com/zfsonlinux/zfs/issues/2094
action: To correct the issue run 'zpool scrub'.
config:
...
If there was an async destroy in progress 'zpool import' will prevent
the pool from being imported. Further advice on how to proceed will be
provided by the error message as follows.
$ zpool import
pool: zol-0.6.2-173
id: 1165955789558693437
state: ONLINE
status: Errata #2 detected.
action: The pool can not be imported with this version of ZFS due to an
active asynchronous destroy. Revert to an earlier version and
allow the destroy to complete before updating.
see: http://zfsonlinux.org/msg/ZFS-8000-ER
config:
...
Pools affected by the damaged dsl_scan_phys_t can be detected prior to
an upgrade by running the following command as root:
zdb -dddd poolname 1 | grep -P '^\t\tscan = ' | sed -e 's;scan = ;;' | wc -w
Note that `poolname` must be replaced with the name of the pool you wish
to check. A value of 25 indicates the dsl_scan_phys_t has been damaged.
A value of 24 indicates that the dsl_scan_phys_t is normal. A value of 0
indicates that there has never been a scrub run on the pool.
The regression caused by the change to zbookmark_t never made it into a
tagged release, Gentoo backports, Ubuntu, Debian, Fedora, or EPEL
stable respositorys. Only those using the HEAD version directly from
Github after the 0.6.2 but before the 0.6.3 tag are affected.
This patch does have one limitation that should be mentioned. It will not
detect errata #2 on a pool unless errata #1 is also present. It expected
this will not be a significant problem because pools impacted by errata #2
have a high probably of being impacted by errata #1.
End users can ensure they do no hit this unlikely case by waiting for all
asynchronous destroy operations to complete before updating ZoL. The
presence of any background destroys on any imported pools can be checked
by running `zpool get freeing` as root. This will display a non-zero
value for any pool with an active asynchronous destroy.
Lastly, it is expected that no user data has been lost as a result of
this erratum.
Original-patch-by: Tim Chase <tim@chase2k.com>
Reworked-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2094
From time to time it may be necessary to inform the pool administrator
about an errata which impacts their pool. These errata will by shown
to the administrator through the 'zpool status' and 'zpool import'
output as appropriate. The errata must clearly describe the issue
detected, how the pool is impacted, and what action should be taken
to resolve the situation. Additional information for each errata will
be provided at http://zfsonlinux.org/msg/ZFS-8000-ER.
To accomplish the above this patch adds the required infrastructure to
allow the kernel modules to notify the utilities that an errata has
been detected. This is done through the ZPOOL_CONFIG_ERRATA uint64_t
which has been added to the pool configuration nvlist.
To add a new errata the following changes must be made:
* A new errata identifier must be assigned by adding a new enum value
to the zpool_errata_t type. New enums must be added to the end to
preserve the existing ordering.
* Code must be added to detect the issue. This does not strictly
need to be done at pool import time but doing so will make the
errata visible in 'zpool import' as well as 'zpool status'. Once
detected the spa->spa_errata member should be set to the new enum.
* If possible code should be added to clear the spa->spa_errata member
once the errata has been resolved.
* The show_import() and status_callback() functions must be updated
to include an informational message describing the errata. This
should include an action message describing what an administrator
should do to address the errata.
* The documentation at http://zfsonlinux.org/msg/ZFS-8000-ER must be
updated to describe the errata. This space can be used to provide
as much additional information as needed to fully describe the errata.
A link to this documentation will be automatically generated in the
output of 'zpool import' and 'zpool status'.
Original-idea-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Richard Yao <ryao@gentoo.or
Issue #2094
Commit 1421c89142 added a field to
zbookmark_t that unintentinoally caused a disk format change. This
negatively affected backward compatibility and platform portability.
Therefore, this field is being removed.
The function that field permitted is left unimplemented until a later
patch that will reimplement the field in a way that does not affect the
disk format.
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2094
Add the "relatime" property. When set to "on", a file's atime will only
be updated if the existing atime at least a day old or if the existing
ctime or mtime has been updated since the last access. This behavior
is compatible with the Linux "relatime" mount option.
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2064Closes#1917
When transitioning current open TXG into QUIESCE state and opening
a new one txg_quiesce() calls gethrtime():
- to mark the birth time of the new TXG
- to record the SPA txg history kstat
- implicitely inside spa_txg_history_add()
These timestamps are practically the same, so that the first one
can be used instead of the other two. The only visible difference
is that inside spa_txg_history_add() the time spent in kmem_zalloc()
will be counted towards the opened TXG.
Since at this point the new TXG already exists (tx->tx_open_txg
has been already incremented) it is actually a correct accounting.
In any case this extra work is only happening when spa_txg_history
kstat is activated (i.e. zfs_txg_history > 0) and doesn't affect
the normal processing in any way.
Signed-off-by: Cyril Plisko <cyril.plisko@mountall.com>
Issue #2075
In several cases when digging into kstats we can found two txgs
in SYNC state, e.g.
txg birth state nreserved nread nwritten ...
985452 258127184872561 C 0 373948416 2376272384 ...
985453 258129016180616 C 0 378173440 28793344 ...
985454 258129016271523 S 0 0 0 ...
985455 258130864245986 S 0 0 0 ...
985456 258130867458851 O 0 0 0 ...
However only first txg (985454) is really syncing at this moment.
The other one (985455) marked as SYNCED is actually in a post-QUIESCED
state and waiting to start sync. So, the new TXG_STATE_WAIT_FOR_SYNC
state between TXG_STATE_QUIESCED and TXG_STATE_SYNCED was added to
reveal this situation.
txg birth state nreserved nread nwritten ...
1086896 235261068743969 C 0 163577856 8437248 ...
1086897 235262870830801 C 0 280625152 822594048 ...
1086898 235264172219064 S 0 0 0 ...
1086899 235264936134407 W 0 0 0 ...
1086900 235264936296156 O 0 0 0 ...
Signed-off-by: Igor Lvovsky <ilvovsky@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2075
When accessing the zp->z_mode through the SA bulk interface we
expect that 64-bits are available to hold the result. However,
on 32-bit platforms mode_t will only be 32-bits so we cannot
pass it to SA_ADD_BULK_ATTR(). Instead a local uint64_t variable
must be used and the result assigned to zp->z_mode.
This went unnoticed on 32-bit little endian platforms because
the bytes happen to end up in the correct 32-bits. But on big
endian platforms like Sparc the zp->z_mode will always end up
set to zero.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: marku89 <mar42@kola.li>
Issue #1700
Back the allocations for ddt tables+entries and l2arc headers with
kmem caches. This will reduce the cost of allocating these commonly
used structures and allow for greater visibility of them through the
/proc/spl/kmem/slab interface.
Signed-off-by: John Layman <jlayman@sagecloud.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1893
Four new dataset properties have been added to support SELinux. They
are 'context', 'fscontext', 'defcontext' and 'rootcontext' which map
directly to the context options described in mount(8). When one of
these properties is set to something other than 'none'. That string
will be passed verbatim as a mount option for the given context when
the filesystem is mounted.
For example, if you wanted the rootcontext for a filesystem to be set
to 'system_u:object_r:fs_t' you would set the property as follows:
$ zfs set rootcontext="system_u:object_r:fs_t" storage-pool/media
This will ensure the filesystem is automatically mounted with that
rootcontext. It is equivalent to manually specifying the rootcontext
with the -o option like this:
$ zfs mount -o rootcontext=system_u:object_r:fs_t storage-pool/media
By default all four contexts are set to 'none'. Further information
on SELinux contexts is detailed in mount(8) and selinux(8) man pages.
Signed-off-by: Matthew Thode <prometheanfire@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Richard Yao <ryao@gentoo.org>
Closes#1504
The vast majority of these changes are in Linux specific code.
They are the result of not having an automated style checker to
validate the code when it was originally written. Others were
caused when the common code was slightly adjusted for Linux.
This patch contains no functional changes. It only refreshes
the code to conform to style guide.
Everyone submitting patches for inclusion upstream should now
run 'make checkstyle' and resolve any warning prior to opening
a pull request. The automated builders have been updated to
fail a build if when 'make checkstyle' detects an issue.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1821
In order to minimize any future disruption caused by the addition
and removal /dev/zfs ioctls this patch makes the following changes.
1) Sync ZoL's ioctl ordering such that it matches Illumos. For
historic reasons the ZFS_IOC_DESTROY_SNAPS and ZFS_IOC_POOL_REGUID
ioctls were out of order.
2) Move Linux and FreeBSD specific ioctls in to their own reserved
ranges. This allows us to preserve the existing ordering when
new ioctls are added by either Illumos or FreeBSD. When an
ioctl is no longer needed it should be retired in place.
This change alters the ZFS user/kernel ABI so make sure you rebuild
both your user and kernel modules. However, it should allow for a
much stabler interface going forward.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Closes#1973
Early versions of ZFS coordinated the creation and destruction
of device minors from userspace. This was inherently racy and
in late 2009 these ioctl()s were removed leaving everything up
to the kernel. This significantly simplified the code.
However, we never picked up these changes in ZoL since we'd
already significantly adjusted this code for Linux. This patch
aims to rectify that by finally removing ZFC_IOC_*_MINOR ioctl()s
and moving all the functionality down in to the kernel. Since
this cleanup will change the kernel/user ABI it's being done
in the same tag as the previous libzfs_core ABI changes. This
will minimize, but not eliminate, the disruption to end users.
Once merged ZoL, Illumos, and FreeBSD will basically be back
in sync in regards to handling ZVOLs in the common code. While
each platform must have its own custom zvol.c implemenation the
interfaces provided are consistent.
NOTES:
1) This patch introduces one subtle change in behavior which
could not be easily avoided. Prior to this change callers
of 'zfs create -V ...' were guaranteed that upon exit the
/dev/zvol/ block device link would be created or an error
returned. That's no longer the case. The utilities will no
longer block waiting for the symlink to be created. Callers
are now responsible for blocking, this is why a 'udev_wait'
call was added to the 'label' function in scripts/common.sh.
2) The read-only behavior of a ZVOL now solely depends on if
the ZVOL_RDONLY bit is set in zv->zv_flags. The redundant
policy setting in the gendisk structure was removed. This
both simplifies the code and allows us to safely leverage
set_disk_ro() to issue a KOBJ_CHANGE uevent. See the
comment in the code for futher details on this.
3) Because __zvol_create_minor() and zvol_alloc() may now be
called in a sync task they must use KM_PUSHPAGE.
References:
illumos/illumos-gate@681d9761e8
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Tim Chase <tim@chase2k.com>
Closes#1969
The DMU zfetch code organizes streams with lists not avl trees. A
avl_node_t was mistakenly used for a list_node_t in the zstream_t
type. This is incorrect (but harmless) and when unnoticed because:
1) The list functions explicitly cast the value preventing a warning,
2) sizeof(avl_node_t) >= sizeof(list_node_t) so no overrun occurs, and
3) The calculated offset is the same regardless of the type.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1946
4045 zfs write throttle & i/o scheduler performance work
1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync
read, sync write, async read, async write, and scrub/resilver. The scheduler
issues a number of concurrent i/os from each class to the device. Once a class
has been selected, an i/o is selected from this class using either an elevator
algorithem (async, scrub classes) or FIFO (sync classes). The number of
concurrent async write i/os is tuned dynamically based on i/o load, to achieve
good sync i/o latency when there is not a high load of writes, and good write
throughput when there is. See the block comment in vdev_queue.c (reproduced
below) for more details.
2. The write throttle (dsl_pool_tempreserve_space() and
txg_constrain_throughput()) is rewritten to produce much more consistent delays
when under constant load. The new write throttle is based on the amount of
dirty data, rather than guesses about future performance of the system. When
there is a lot of dirty data, each transaction (e.g. write() syscall) will be
delayed by the same small amount. This eliminates the "brick wall of wait"
that the old write throttle could hit, causing all transactions to wait several
seconds until the next txg opens. One of the keys to the new write throttle is
decrementing the amount of dirty data as i/o completes, rather than at the end
of spa_sync(). Note that the write throttle is only applied once the i/o
scheduler is issuing the maximum number of outstanding async writes. See the
block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for
more details.
This diff has several other effects, including:
* the commonly-tuned global variable zfs_vdev_max_pending has been removed;
use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead.
* the size of each txg (meaning the amount of dirty data written, and thus the
time it takes to write out) is now controlled differently. There is no longer
an explicit time goal; the primary determinant is amount of dirty data.
Systems that are under light or medium load will now often see that a txg is
always syncing, but the impact to performance (e.g. read latency) is minimal.
Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this.
* zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression,
checksum, etc. This improves latency by not allowing these CPU-intensive tasks
to consume all CPU (on machines with at least 4 CPU's; the percentage is
rounded up).
--matt
APPENDIX: problems with the current i/o scheduler
The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem
with this is that if there are always i/os pending, then certain classes of
i/os can see very long delays.
For example, if there are always synchronous reads outstanding, then no async
writes will be serviced until they become "past due". One symptom of this
situation is that each pass of the txg sync takes at least several seconds
(typically 3 seconds).
If many i/os become "past due" (their deadline is in the past), then we must
service all of these overdue i/os before any new i/os. This happens when we
enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in
the future. If we can't complete all the i/os in 2.5 seconds (e.g. because
there were always reads pending), then these i/os will become past due. Now we
must service all the "async" writes (which could be hundreds of megabytes)
before we service any reads, introducing considerable latency to synchronous
i/os (reads or ZIL writes).
Notes on porting to ZFS on Linux:
- zio_t gained new members io_physdone and io_phys_children. Because
object caches in the Linux port call the constructor only once at
allocation time, objects may contain residual data when retrieved
from the cache. Therefore zio_create() was updated to zero out the two
new fields.
- vdev_mirror_pending() relied on the depth of the per-vdev pending queue
(vq->vq_pending_tree) to select the least-busy leaf vdev to read from.
This tree has been replaced by vq->vq_active_tree which is now used
for the same purpose.
- vdev_queue_init() used the value of zfs_vdev_max_pending to determine
the number of vdev I/O buffers to pre-allocate. That global no longer
exists, so we instead use the sum of the *_max_active values for each of
the five I/O classes described above.
- The Illumos implementation of dmu_tx_delay() delays a transaction by
sleeping in condition variable embedded in the thread
(curthread->t_delay_cv). We do not have an equivalent CV to use in
Linux, so this change replaced the delay logic with a wrapper called
zfs_sleep_until(). This wrapper could be adopted upstream and in other
downstream ports to abstract away operating system-specific delay logic.
- These tunables are added as module parameters, and descriptions added
to the zfs-module-parameters.5 man page.
spa_asize_inflation
zfs_deadman_synctime_ms
zfs_vdev_max_active
zfs_vdev_async_write_active_min_dirty_percent
zfs_vdev_async_write_active_max_dirty_percent
zfs_vdev_async_read_max_active
zfs_vdev_async_read_min_active
zfs_vdev_async_write_max_active
zfs_vdev_async_write_min_active
zfs_vdev_scrub_max_active
zfs_vdev_scrub_min_active
zfs_vdev_sync_read_max_active
zfs_vdev_sync_read_min_active
zfs_vdev_sync_write_max_active
zfs_vdev_sync_write_min_active
zfs_dirty_data_max_percent
zfs_delay_min_dirty_percent
zfs_dirty_data_max_max_percent
zfs_dirty_data_max
zfs_dirty_data_max_max
zfs_dirty_data_sync
zfs_delay_scale
The latter four have type unsigned long, whereas they are uint64_t in
Illumos. This accommodates Linux's module_param() supported types, but
means they may overflow on 32-bit architectures.
The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most
likely to overflow on 32-bit systems, since they express physical RAM
sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to
2^32 which does overflow. To resolve that, this port instead initializes
it in arc_init() to 25% of physical RAM, and adds the tunable
zfs_dirty_data_max_max_percent to override that percentage. While this
solution doesn't completely avoid the overflow issue, it should be a
reasonable default for most systems, and the minority of affected
systems can work around the issue by overriding the defaults.
- Fixed reversed logic in comment above zfs_delay_scale declaration.
- Clarified comments in vdev_queue.c regarding when per-queue minimums take
effect.
- Replaced dmu_tx_write_limit in the dmu_tx kstat file
with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts
how many times a transaction has been delayed because the pool dirty
data has exceeded zfs_delay_min_dirty_percent. The latter counts how
many times the pool dirty data has exceeded zfs_dirty_data_max (which
we expect to never happen).
- The original patch would have regressed the bug fixed in
zfsonlinux/zfs@c418410, which prevented users from setting the
zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE.
A similar fix is added to vdev_queue_aggregate().
- In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the
heap instead of the stack. In Linux we can't afford such large
structures on the stack.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Ned Bass <bass6@llnl.gov>
Reviewed by: Brendan Gregg <brendan.gregg@joyent.com>
Approved by: Robert Mustacchi <rm@joyent.com>
References:
http://www.illumos.org/issues/4045illumos/illumos-gate@69962b5647
Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1913
Currently, using msync() results in the following code path:
sys_msync -> zpl_fsync -> filemap_write_and_wait_range -> zpl_writepages -> write_cache_pages -> zpl_putpage
In such a code path, zil_commit() is called as part of zpl_putpage().
This means that for each page, the write is handed to the DMU, the ZIL
is committed, and only then do we move on to the next page. As one might
imagine, this results in atrocious performance where there is a large
number of pages to write: instead of committing a batch of N writes,
we do N commits containing one page each. In some extreme cases this
can result in msync() being ~700 times slower than it should be, as well
as very inefficient use of ZIL resources.
This patch fixes this issue by making sure that the requested writes
are batched and then committed only once. Unfortunately, the
implementation is somewhat non-trivial because there is no way to run
write_cache_pages in SYNC mode (so that we get all pages) without
making it wait on the writeback tag for each page.
The solution implemented here is composed of two parts:
- I added a new callback system to the ZIL, which allows the caller to
be notified when its ITX gets written to stable storage. One nice
thing is that the callback is called not only in zil_commit() but
in zil_sync() as well, which means that the caller doesn't have to
care whether the write ended up in the ZIL or the DMU: it will get
notified as soon as it's safe, period. This is an improvement over
dmu_tx_callback_register() that was used previously, which only
supports DMU writes. The rationale for this change is to allow
zpl_putpage() to be notified when a ZIL commit is completed without
having to block on zil_commit() itself.
- zpl_writepages() now calls write_cache_pages in non-SYNC mode, which
will prevent (1) write_cache_pages from blocking, and (2) zpl_putpage
from issuing ZIL commits. zpl_writepages() will issue the commit
itself instead of relying on zpl_putpage() to do it, thus nicely
batching the writes. Note, however, that we still have to call
write_cache_pages() again in SYNC mode because there is an edge case
documented in the implementation of write_cache_pages() whereas it
will not give us all dirty pages when running in non-SYNC mode. Thus
we need to run it at least once in SYNC mode to make sure we honor
persistency guarantees. This only happens when the pages are
modified at the same time msync() is running, which should be rare.
In most cases there won't be any additional pages and this second
call will do nothing.
Note that this change also fixes a bug related to #907 whereas calling
msync() on pages that were already handed over to the DMU in a previous
writepages() call would make msync() block until the next TXG sync
instead of returning as soon as the ZIL commit is complete. The new
callback system fixes that problem.
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1849Closes#907
Because ZFS bypasses the page cache we don't inherit per-task I/O
accounting for free. However, the Linux kernel does provide helper
functions allow us to perform our own accounting. These are most
commonly used for direct IO which also bypasses the page cache, but
they can be used for the common read/write call paths as well.
Signed-off-by: Pavel Snajdr <snajpa@snajpa.net>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#313Closes#1275
