Commit Graph

670 Commits

Author SHA1 Message Date
Prakash Surya 77765b540b Remove "arc_meta_used" from arc_adjust calculation
Using "arc_meta_used" to determine if the arc's mru list is over it's
target value of "arc_p" doesn't seem correct. The size of the mru list
and the value of "arc_meta_used", although related, are completely
independent. Buffers contained in "arc_meta_used" may not even be
contained in the arc's mru list. As such, this patch removes
"arc_meta_used" from the calculation in arc_adjust.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
2014-02-21 16:10:49 -08:00
Prakash Surya 94520ca462 Prune metadata from ghost lists in arc_adjust_meta
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
2014-02-21 16:10:49 -08:00
Prakash Surya 1e3cb67b53 Revert "Return -1 from arc_shrinker_func()"
This reverts commit c11a12bc3b.

Out of memory events were fixed by reverting this patch.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
2014-02-21 16:10:49 -08:00
Prakash Surya 624227854e Disable arc_p adapt dampener by default
It's unclear why adjustments to arc_p need to be dampened as they are in
arc_adjust. With that said, it's removal significantly improves the arc's
ability to "warm up" to a given workload. Thus, I'm disabling by default
until its usefulness is better understood.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
2014-02-21 16:10:49 -08:00
Prakash Surya f521ce1b9c Allow "arc_p" to drop to zero or grow to "arc_c"
Setting a limit on the minimum value of "arc_p" has been shown to have
detrimental effects on the arc hit rate for certain "metadata" intensive
workloads. Specifically, this has been exhibited with a workload that
constantly dirties new "metadata" but also frequently touches a "small"
amount of mfu data (e.g. mkdir's).

What is seen is that the new anon data throttles the mfu list to a
negligible size (because arc_p > anon + mru in arc_get_data_buf), even
though the mfu ghost list receives a constant stream of hits. To remedy
this, arc_p is now allowed to drop to zero if the algorithm deems it
necessary.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
2014-02-21 16:10:27 -08:00
Prakash Surya 89c8cac493 Disable aggressive arc_p growth by default
For specific workloads consisting mainly of mfu data and new anon data
buffers, the aggressive growth of arc_p found in the arc_get_data_buf()
function can have detrimental effects on the mfu list size and ghost
list hit rate.

Running a workload consisting of two processes:

    * Process 1 is creating many small files
    * Process 2 is tar'ing a directory consisting of many small files

I've seen arc_p and the mru grow to their maximum size, while the mru
ghost list receives 100K times fewer hits than the mfu ghost list.

Ideally, as the mfu ghost list receives hits, arc_p should be driven
down and the size of the mfu should increase. Given the specific
workload I was testing with, the mfu list size should grow to a point
where almost no mfu ghost list hits would occur. Unfortunately, this
does not happen because the newly dirtied anon buffers constancy drive
arc_p to its maximum value and keep it there (effectively prioritizing
the mru list and starving the mfu list down to a negligible size).

The logic to increment arc_p from within the arc_get_data_buf() function
was introduced many years ago in this upstream commit:

    commit 641fbdae3a027d12b3c3dcd18927ccafae6d58bc
    Author: maybee <none@none>
    Date:   Wed Dec 20 15:46:12 2006 -0800

        6505658 target MRU size (arc.p) needs to be adjusted more aggressively

and since I don't fully understand the motivation for the change, I am
reluctant to completely remove it.

As a way to test out how it's removal might affect performance, I've
disabled that code by default, but left it tunable via a module option.
Thus, if its removal is found to be grossly detrimental for certain
workloads, it can be re-enabled on the fly, without a code change.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
2014-02-21 14:53:28 -08:00
Prakash Surya 39e055c44b Adjust arc_p based on "bytes" in arc_shrink
In an attempt to prevent arc_c from collapsing "too fast", the
arc_shrink() function was updated to take a "bytes" parameter by this
change:

    commit 302f753f16
    Author: Brian Behlendorf <behlendorf1@llnl.gov>
    Date:   Tue Mar 13 14:29:16 2012 -0700

        Integrate ARC more tightly with Linux

Unfortunately, that change failed to make a similar change to the way
that arc_p was updated. So, there still exists the possibility for arc_p
to collapse to near 0 when the kernel start calling the arc's shrinkers.

This change attempts to fix this, by decrementing arc_p by the "bytes"
parameter in the same way that arc_c is updated.

In addition, a minimum value of arc_p is attempted to be maintained,
similar to the way a minimum arc_p value is maintained in arc_adapt().

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2110
2014-02-21 14:53:08 -08:00
Brian Behlendorf 9141582592 Set zfs_arc_min to 4MB
Decrease the mimimum ARC size from 1/32 of total system memory
(or 64MB) to a much smaller 4MB.

1) Large systems with over a 1TB of memory are being deployed
   and reserving 1/32 of this memory (32GB) as the mimimum
   requirement is overkill.

2) Tiny systems like the raspberry pi may only have 256MB of
   memory in which case 64MB is far too large.

The ARC should be reclaimable if the VFS determines it needs
the memory for some other purpose.  If you want to ensure the
ARC is never completely reclaimed due to memory pressure you
may still set a larger value with zfs_arc_min.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Prakash Surya <surya1@llnl.gov>
Issue #2110
2014-02-21 14:52:02 -08:00
Richard Yao 4f2dcb3eee Add erratum for issue #2094
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
2014-02-21 12:10:40 -08:00
Brian Behlendorf ffe9d38275 Add generic errata infrastructure
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
2014-02-21 12:10:40 -08:00
Richard Yao ed9e8368d3 Revert changes to zbookmark_t
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
2014-02-21 12:10:39 -08:00
Tim Chase 98fad86293 Propagate errors when registering "relatime" property callback.
Various errors can occur when registering property callbacks.  As the
author's comments indicate, the code is very paranoid about preserving
the first-seen error when registering callbacks.  This patch causes an
error seen while registering the "relatime" callback to not clobber a
previously-seen error.

Reported-by: Jorgen Lundman <lundman@lundman.net>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2117
2014-02-12 09:38:28 -08:00
Brian Behlendorf c5cb66addc Fix corrupted l2_asize in arcstats
Commit e0b0ca9 accidentally corrupted the l2_asize displayed in
arcstats.  This was caused by changing the l2arc_buf_hdr.b_asize
member from an int to uint32_t type.  There are places in the
code where this field is cast to a uint64_t resulting in the
b_hits member being treated as part of b_asize.

To resolve the issue the type has been changed to a uint64_t,
and the b_hits member is placed after the enum to prevent the
size of the structure from increasing.

This is a good example of exactly why it's a bad idea to use
ambiguous types (int) in these structures.

Signed-off-by: DHE <git@dehacked.net>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1990
2014-02-05 12:24:53 -08:00
Matthew Ahrens 2e7b7657cd 4188 assertion failed in dmu_tx_hold_free(): dn_datablkshift != 0
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>

Refences:
  https://www.illumos.org/issues/4188
  illumos/illumos-gate@bb411a08b0

Ported-by: Chris Dunlop <chris@onthe.net.au>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2091
2014-01-31 10:49:34 -08:00
Matthew Ahrens 8b4646494c Illumos 4504 traverse_visitbp: visit group before user
4504 traverse_visitbp: visit DMU_GROUPUSED_OBJECT before DMU_USERUSED_OBJECT

Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>

References:
  https://illumos.org/issues/4504
  http://code.delphix.com/illumos-4504
  http://svnweb.freebsd.org/base?view=revision&revision=260812

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tim Chase <tim@chase2k.com>
Closes #2079
2014-01-29 15:50:49 -08:00
Tim Chase 6d111134c0 Implement relatime.
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 #2064
Closes #1917
2014-01-29 15:50:44 -08:00
Cyril Plisko 01b738f457 Call gethrtime() only once per new txg creation
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
2014-01-23 13:31:51 -08:00
Igor Lvovsky 478d64fdae Add additional state TXG_STATE_WAIT_FOR_SYNC for txg.
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
2014-01-23 13:31:51 -08:00
Shen Yan 93292b3081 Use enum type(zfetch_dirn_t) instead
Fix code with zfetch_dirn_t, which is more readable and clear.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2068
2014-01-23 12:56:33 -08:00
Tim Chase 4461aa6118 Allow chown/chgrp when no ACL SAs exist.
From the comment in the commit:

Some ZFS implementations (ZEVO) create neither a ZNODE_ACL nor a DACL_ACES
SA in which case ENOENT is returned from zfs_acl_node_read() when the
SA can't be located.  Allow chown/chgrp to succeed in these cases rather
than returning an error that makes no sense in the context of the caller.

Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue zfs-osx/zfs#86
Closes #1911
Closes #2029
2014-01-23 11:07:29 -08:00
Ned Bass 04aa2de8f7 vdev_file_io_start() to use taskq_dispatch(TQ_PUSHPAGE)
The vdev_file_io_start() function may be processing a zio that the
txg_sync thread is waiting on.  In this case it is not safe to perform
memory allocations that may generate new I/O since this could cause a
deadlock.  To avoid this, call taskq_dispatch() with TQ_PUSHPAGE
instead of TQ_SLEEP.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1928
2014-01-23 09:58:07 -08:00
Brian Behlendorf 35d3e32274 Use long holds in zvol_set_volsize()
Under Linux the zvol_set_volsize() function was originally written
to use dmu_objset_hold()/dmu_objset_rele().  Subsequently, the
dmu_objset_own()/dmu_objset_disown() interfaces were added but
the ZVOL code wasn't updated to take advantage of them.

This was never an issue but after the dsl_pool_config changes
the code now takes the config lock twice.  The cleanest solution
is to shift to using dmu_objset_own() which takes a long hold
on the dataset and does not hold the dsl pool lock.

This patch also slightly restructures the existing code such
that it more closely resembles the upstream Illumos code.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2039
2014-01-14 14:46:12 -08:00
Brian Behlendorf fd23720ae1 Drain iput taskq outside z_teardown_lock
It's unsafe to drain the iput taskq while holding the z_teardown_lock
as a writer.  This is because when the last reference on an inode is
dropped it may still have pages which need to be written to disk.
This will be done through zpl_writepages which will acquire the
z_teardown_lock as a reader in ZFS_ENTER.  Therefore, if we're
holding the lock as a writer in zfs_sb_teardown the unmount will
deadlock.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chris Dunlop <chris@onthe.net.au>
Closes #1988
2014-01-09 15:54:08 -08:00
Brian Behlendorf 4fcc43790c Force LZ4_FORCE_SW_BITCOUNT for Sparc
This change was proposed for Sparc but it's not clear to me
why it's required.  Proper support exists in the lz4 code to
detect the endianness and the required builtins are available
for gcc.  Still I'm including the patch because it will only
impact Sparc and it may resolve a case which hasn't occured
to me.

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
2014-01-09 15:54:03 -08:00
Brian Behlendorf b585bc4afa Fix zfs_getattr_fast types
On Sparc sp->blksize will be a 64-bit value which is then cast
incorrectly to a 32-bit value.  For big endian systems this
results in an incorrect value for sp->blksize.  To resolve the
problem local variables of the correct size are used and then
assigned to sp->blksize.

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
2014-01-09 15:50:23 -08:00
Brian Behlendorf aa0218d6a1 Fix nvlist 'Bus Error' for Sparc
The mis-aligned memory accesses in nvpair_native_embedded() and
nvpair_native_embedded_array() will cause a 'Bus Error' for
architectures such as Sparc which not fully byte addressible.
To avoid this issue care is taken to avoid dereferencing the
potentially mis-aligned packed nvlist_t.

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
2014-01-09 15:50:15 -08:00
Brian Behlendorf 7f89ae6ba0 Use local variable to read zp->z_mode
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
2014-01-09 15:50:11 -08:00
John Layman ecf3d9b8e6 Add ddt, ddt_entry, and l2arc_hdr caches
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
2014-01-07 10:33:11 -08:00
Tim Chase fb8e608d9d Fix the creation of ZPOOL_HIST_CMD pool history entries.
Move the libzfs_fini() after the zpool_log_history() call so the
ZPOOL_HIST_CMD entry can get written.

Fix the handling of saved_poolname in zfsdev_ioctl()
which was broken as part of the stack-reduction work in
a168788053.

Since ZoL destroys the TSD data in which the previously successful
ioctl()'s pool name is stored following every vop, the ZFS_IOC_LOG_HISTORY
ioctl has a very important restriction: it can only successfully write
a long entry following a successful ioctl() if no intervening vops have
been performed.  Some of zfs subcommands do perform intervening vops and
to do the logging themselves. At the moment, the "create" and "clone"
subcommands have been modified appropriately.

Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1998
2014-01-07 09:00:26 -08:00
Tim Chase 5d862cb0d9 Properly handle updates of variably-sized SA entries.
During the update process in sa_modify_attrs(), the sizes of existing
variably-sized SA entries are obtained from sa_lengths[]. The case where
a variably-sized SA was being replaced neglected to increment the index
into sa_lengths[], so subsequent variable-length SAs would be rewritten
with the wrong length. This patch adds the missing increment operation
so all variably-sized SA entries are stored with their correct lengths.

Previously, a size-changing update of a variably-sized SA that occurred
when there were other variably-sized SAs in the bonus buffer would cause
the subsequent SAs to be corrupted.  The most common case in which this
would occur is when a mode change caused the ZPL_DACL_ACES entry to
change size when a ZPL_DXATTR (SA xattr) entry already existed.

The following sequence would have caused a failure when xattr=sa was in
force and would corrupt the bonus buffer:

	open(filename, O_WRONLY | O_CREAT, 0600);
	...
	lsetxattr(filename, ...);	/* create xattr SA */
	chmod(filename, 0650);		/* enlarges the ACL */

Signed-off-by: Chris Dunlop <chris@onthe.net.au>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1978
2013-12-20 13:52:33 -08:00
Brian Behlendorf ac0340970c Register correct handlers for nvlist_{dup,pack,unpack}
This change is related to commit 81eaf15 which ensured the correct
allocation handlers were installed for nvlist_alloc().  The nvlist
functions nvlist_dup(), nvlist_pack(), and nvlist_unpack() suffer
from the same issue and have been updated accordingly.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1937
2013-12-20 13:52:28 -08:00
Matthew Thode 11b9ec23b9 Add full SELinux support
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
2013-12-19 10:37:31 -08:00
Michael Kjorling d1d7e2689d cstyle: Resolve C style issues
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
2013-12-18 16:46:35 -08:00
Turbo Fredriksson fd8febbd1e Add zfs_send_corrupt_data module option
Tuning setting to ignore read/checksum errors when sending data.

Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1982
Issue #1897
2013-12-18 16:46:35 -08:00
Chunwei Chen 7dc71949f2 Fix z_sync_cnt decrement in zfs_close
The comment in zfs_close states that "Under Linux the zfs_close() hook
is not symmetric with zfs_open()". This is not true. zfs_open/zfs_close
is associated with every successful struct file creation/deletion, which
should always be balanced.

Here is an example of what's wrong:

Process A		B
	open(O_SYNC)
	z_sync_cnt = 1
			open(O_SYNC)
			z_sync_cnt = 2
	close()
	z_sync_cnt = 0

So z_sync_cnt is 0 even if B still has the file with O_SYNC.

Also moves the generic_file_open call before zfs_open to ensure that in
the case generic_file_open fails z_sync_cnt is not incremented.  This
is safe because generic_file_open has no side effects.

Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1962
2013-12-17 10:28:27 -08:00
Brian Behlendorf ce37ebd2eb cstyle: zvol.c
Update zvol.c to conform to the style guidelines, verified by
running cstyle.pl on the source file.  This patch contains
no functional changes.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Tim Chase <tim@chase2k.com>
Issue #1821
2013-12-16 09:41:45 -08:00
Brian Behlendorf 2e0358cbca Sync /dev/zfs ioctl ordering
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
2013-12-16 09:41:39 -08:00
Brian Behlendorf ba6a24026c Remove ZFC_IOC_*_MINOR ioctl()s
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
2013-12-16 09:15:57 -08:00
George Wilson dda12da9f1 Illumos #4121 vdev_label_init read only
4121 vdev_label_init should treat request as succeeded when pool
     is read only
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Approved by: Richard Lowe <richlowe@richlowe.net>

References:
  https://www.illumos.org/issues/4121
  illumos/illumos-gate@973c78e94b

Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1863
2013-12-12 10:24:01 -08:00
Tim Chase 84b0aac5fd Fix atime handling.
Previously, the atime-modifying vnops called ZFS_ACCESSTIME_STAMP()
followed by zfs_inode_update() to update the atime.  However, since atimes
are cached in the znode for delayed writing, the zfs_inode_update()
function would effectively ignore the cached atime by reading it from
the SA.

This commit moves the updating of the atime in the inode into
zfs_tstamp_update_setup() which is called by the ZFS_ACCESSTIME_STAMP()
macro and eliminates the call to zfs_inode_update() in the atime-modifying
vnops.

It's possible the same thing could have been done directly in
zfs_inode_update() but I wasn't sure that it was safe in all cases where
it is called.

The effect is that atime handling is as if "strictatime" were selected;
even if the filesystem is mounted with "relatime".

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1949
2013-12-12 10:23:58 -08:00
david.chen be5db977ea Remove MAX when initializing arc_c_max
The MAX when initializing arc_c_max doesn't make any sense because
it hasn't been set anywhere before. Though, arc_c_max should be
implicitly set to zero when initializing arc_stats, so the MAX
doesn't make any difference.

The MAX was mistakenly left if place when the Illumos default
values were changed for Linux.

Signed-off-by: david.chen <tuxoko@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1941
2013-12-10 10:05:40 -08:00
Ned Bass b6e335bfc4 Revert "Use directory xattrs for symlinks"
This reverts commit 6a7c0ccca4.

A proper fix for Issue #1648 was landed under Issue #1890, so this is no
longer needed.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1648
2013-12-10 09:48:30 -08:00
James Pan 472e7c6085 sa_find_sizes() may compute wrong SA header size
Under the right conditions sa_find_sizes() will compute an incorrect
size of the system attribute (SA) header.  This causes a failed assertion
when the SA_HDR_SIZE_MATCH_LAYOUT() test returns false, and may lead
to corruption of SA data.

The bug presents itself when there are more than two variable-length SAs
of just the right size to fit in the bonus buffer of a dnode.  The
existing logic fails to account for the SA header space needed to store
the sizes of all the variable-length SAs.

A reproducer was possible on Linux by setting the xattr=sa dataset
property and storing xattrs on symbolic links (Issue #1648).  Note the
corrupt link target name:

$ zfs set xattr=sa tank/fish
$ cd /tank/fish
$ ln -fs 12345678901234567 link
$ setfattr -n trusted.0000000000000000000 -v 0x000000000000000000000000 -h link
$ setfattr -n trusted.1111111111111111111 -v 0x000000000000000000000000 -h link
$ ls -l link
lrwxrwxrwx 1 root root 17 Dec  6 15:40 link -> 90123456701234567

Commit 6a7c0ccca4 worked around this bug
by forcing xattr's on symlinks to be stored in directory format.  This
change implements a proper fix, so the workaround can now be reverted.

The reference link below contains a reproducer for FreeBSD.

References:
  http://lists.open-zfs.org/pipermail/developer/2013-November/000306.html

Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1890
2013-12-10 09:48:15 -08:00
Brian Behlendorf 90ee9ed32f Fix 'zfs diff' shares error
When creating a dataset with ZoL a zsb->z_shares_dir ZAP object
will not be created because shares are unimplemented.  Instead ZoL
just sets zsb->z_shares_dir to zero to indicate there are no shares.

However, if you import a pool which was created with a different
ZFS implementation then the shares ZAP object may exist.  Code was
added to handle this case but it clearly wasn't sufficiently tested
with other ZFS pools.

There was a bug in the zpl_shares_getattr() function which passed
the wrong inode to zfs_getattr_fast() for the case where are shares
ZAP object does exist.  This causes an EIO to be returned to stat64()
which in turn causes 'zfs diff' to fail.

This fix is the pass the correct inode after a sucessful zfs_zget().
Additionally, only put away the references if we were able to get one.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Graham Booker <https://github.com/gbooker>
Signed-off-by: timemaster67 <https://github.com/timemaster67>
Closes #1426
Closes #481
2013-12-06 09:42:39 -08:00
Brian Behlendorf 99e349db92 Add module versioning
Use the standard Linux MODULE_VERSION macro to expose the installed
zavl, znvpair, zunicode, zcommon, zfs, and zpios module versions.
This will also automatically add a checksum of the .c files and
headers in "srcversion".  See:

  /sys/module/zavl/version
  /sys/module/zavl/srcversion
  /sys/module/znvpair/version
  /sys/module/znvpair/srcversion
  /sys/module/zunicode/version
  /sys/module/zunicode/srcversion
  /sys/module/zcommon/version
  /sys/module/zcommon/srcversion
  /sys/module/zfs/version
  /sys/module/zfs/srcversion
  /sys/module/zpios/version
  /sys/module/zpios/srcversion

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1923
2013-12-06 09:34:41 -08:00
Matthew Ahrens e8b96c6007 Illumos #4045 write throttle & i/o scheduler performance work
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/4045
  illumos/illumos-gate@69962b5647

Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1913
2013-12-06 09:32:43 -08:00
Matthew Ahrens 384f8a09f8 Illumos #4347 ZPL can use dmu_tx_assign(TXG_WAIT)
Fix a lock contention issue by allowing threads not holding
ZPL locks to block when waiting to assign a transaction.

Porting Notes:

zfs_putpage() still uses TXG_NOWAIT, unlike the upstream version.  This
case may be a contention point just like zfs_write(), however it is not
safe to block here since it may be called during memory reclaim.

Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@nexenta.com>
Reviewed by: Boris Protopopov <boris.protopopov@nexenta.com>
Approved by: Dan McDonald <danmcd@nexenta.com>

References:
  https://www.illumos.org/issues/4347
  illumos/illumos-gate@e722410c49

Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2013-12-06 09:30:51 -08:00
Brian Behlendorf 2e40f09410 Remove incorrect ASSERT in zfs_sb_teardown()
As part of zfs_sb_teardown() there is an assertion that all inodes
which are part of the zsb->z_all_znodes list have at least one
reference on them.  This is always true for the standard unmount
case but there are two other cases where it is not strictly true.

* zfs_ioc_rollback() - This is the most common case and it results
  from the fact that we aren't unmounting the filesystem.  During a
  normal unmount the MS_ACTIVE flag will be cleared on the super block
  causing iput_final() to evict the inode when its reference count
  drops to zero.  However, during a rollback MS_ACTIVE remains set
  since we're rolling back a live filesystem and need to preserve the
  existing super block.  This allows inodes with a zero reference count
  to stay in the cache thereby violating the assertion.

* destroy_inode() / zfs_sb_teardown() - There exists a small race
  between dropping the last reference on an inode and removing it from
  the zsb->z_all_znodes list.  This is unlikely to occur but could also
  trigger the assertion which is incorrect.  The inode may safely have
  a zero reference count in this case.

Since allowing a zero reference count on the inode is expected and
safe for both of these cases the simplest thing to do is remove the
ASSERT.  This code is only enabled for default builds so removing
this entirely is a very safe change.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chris Dunlop <chris@onthe.net.au>
Signed-off-by: Tim Chase <tim@chase2k.com>
Closes #1417
Closes #1536
2013-12-02 15:58:58 -08:00
Tim Chase f707635fa5 Some nvlist allocations in hold processing need to use KM_PUSHPAGE.
This should hopefully catch the rest of the allocations in the
user hold/release processing that were missed by commit
65c67ea86e.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1852
Closes #1855
2013-12-02 14:02:46 -08:00
Etienne Dechamps 119a394ab0 Only commit the ZIL once in zpl_writepages() (msync() case).
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 #1849
Closes #907
2013-11-23 15:08:29 -08:00