| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3829 fix for 3740 changed behavior of zfs destroy/hold/release ioctl
Reviewed by: Matt Amdur <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Approved by: Richard Lowe <[email protected]>
References:
https://www.illumos.org/issues/3829
illumos/illumos-gate@bb6e70758d0c30c09f148026d6e686e21cfc8d18
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3818 zpool status -x should report pools with removed l2arc devices
Reviewed by: Saso Kiselkov <[email protected]>
Reviewed by: George Wilson <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3818
illumos/illumos-gate@7f2416ef64fb43dab18d9b36c0da64bea37c0df3
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3740 Poor ZFS send / receive performance due to snapshot
hold / release processing
Reviewed by: Matthew Ahrens <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3740
illumos/illumos-gate@a7a845e4bf22fd1b2a284729ccd95c7370a0438c
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Porting notes:
1. 13fe019870c8779bf2f5b3ff731b512cf89133ef introduced a merge conflict
in dsl_dataset_user_release_tmp where some variables were moved
outside of the preprocessor directive.
2. dea9dfefdd747534b3846845629d2200f0616dad made the previous merge
conflict worse by switching KM_SLEEP to KM_PUSHPAGE. This is notable
because this commit refactors the code, adding a new KM_SLEEP
allocation. It is not clear to me whether this should be converted
to KM_PUSHPAGE.
3. We had a merge conflict in libzfs_sendrecv.c because of copyright
notices.
4. Several small C99 compatibility fixed were made.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3745 zpool create should treat -O mountpoint and -m the same
3811 zpool create -o altroot=/xyz -O mountpoint=/mnt ignores
the mountpoint option
Reviewed by: Matthew Ahrens <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3745
https://www.illumos.org/issues/3811
illumos/illumos-gate@8b713775314bbbf24edd503b4869342d8711ce95
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3744 zfs shouldn't ignore errors unmounting snapshots
Reviewed by: Matthew Ahrens <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3744
illumos/illumos-gate@fc7a6e3fefc649cb65c8e2a35d194781445008b0
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Porting notes:
1. There is no clear way to distinguish between a failure when we
tried to unmount the snapdir of a zvol (which does not exist)
and the failure when we try to unmount a snapdir of a dataset,
so the changes to zfs_unmount_snap() were dropped in favor of
an altered Linux function that unconditionally returns 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3743 zfs needs a refcount audit
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: Eric Schrock <[email protected]>
Reviewed by: George Wilson <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3743
illumos/illumos-gate@b287be1ba86043996f49b1cc34c80cc620f9b841
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3742 zfs comments need cleaner, more consistent style
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Eric Schrock <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3742
illumos/illumos-gate@f7170741490edba9d1d9c697c177c887172bc741
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Porting notes:
1. The change to zfs_vfsops.c was dropped because it involves
zfs_mount_label_policy, which does not exist in the Linux port.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3741 zfs needs better comments
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: Eric Schrock <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3741
illumos/illumos-gate@3e30c24aeefdee1631958ecf17f18da671781956
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3699 zfs hold or release of a non-existent snapshot does not output error
3739 cannot set zfs quota or reservation on pool version < 22
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: Eric Shrock <[email protected]>
Approved by: Dan McDonald <[email protected]>
References:
https://www.illumos.org/issues/3699
https://www.illumos.org/issues/3739
illumos/illumos-gate@013023d4ed2f6d0cf75380ec686a4aac392b4e43
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3582 zfs_delay() should support a variable resolution
3584 DTrace sdt probes for ZFS txg states
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Reviewed by: Dan McDonald <[email protected]>
Reviewed by: Richard Elling <[email protected]>
Approved by: Garrett D'Amore <[email protected]>
References:
https://www.illumos.org/issues/3582
illumos/illumos-gate@0689f76
Ported by: Ned Bass <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
| |
References:
illumos/illumos-gate@44bffe012cad6481c82ad67bacd6b40bd29def2b
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
References:
illumos/illumos-gate@d39ee142a97a7c58f60f7b52c62409f2ff64b234
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Porting notes:
1. This commit was so old that only two lines applied to the modern
code base.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3642 dsl_scan_active() should not issue I/O to determine if async
destroying is active
3643 txg_delay should not hold the tc_lock
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Approved by: Gordon Ross <[email protected]>
References:
https://www.illumos.org/issues/3642
https://www.illumos.org/issues/3643
illumos/illumos-gate@4a92375985c37d61406d66cd2b10ee642eb1f5e7
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Porting Notes:
1. The alignment assumptions for the tx_cpu structure assume that
a kmutex_t is 8 bytes. This isn't true under Linux but tc_pad[]
was adjusted anyway for consistency since this structure was
never carefully aligned in ZoL. If careful alignment does impact
performance significantly this should be reworked to be portable.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3645 dmu_send_impl: possibilty of pool hold leak
3692 Panic on zfs receive of a recursive deduplicated stream
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Reviewed by: Dan McDonald <[email protected]>
Approved by: Richard Lowe <[email protected]>
References:
https://www.illumos.org/issues/3645
https://www.illumos.org/issues/3692
illumos/illumos-gate@de8d9cff565e928d0ace86f3ea0e2b15094d61df
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1792
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3598 want to dtrace when errors are generated in zfs
Reviewed by: Dan Kimmel <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Approved by: Garrett D'Amore <[email protected]>
References:
https://www.illumos.org/issues/3598
illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Porting notes:
1. include/sys/zfs_context.h has been modified to render some new
macros inert until dtrace is available on Linux.
2. Linux-specific changes have been adapted to use SET_ERROR().
3. I'm NOT happy about this change. It does nothing but ugly
up the code under Linux. Unfortunately we need to take it to
avoid more merge conflicts in the future. -Brian
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3517 importing pool with autoreplace=on and "hole" vdevs crashes syseventd
Reviewed by: Albert Lee <[email protected]>
Reviewed by: Jeffry Molanus <[email protected]>
Reviewed by: George Wilson <[email protected]>
Approved by: Christopher Siden <[email protected]>
References:
https://www.illumos.org/issues/3517
illumos/illumos-gate@efb4a871d8fd510a833bdca610528dde5ed69e42
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3603 panic from bpobj_enqueue_subobj()
3604 zdb should print bpobjs more verbosely
3871 GCC 4.5.3 does not like issue 3604 patch
Reviewed by: Henrik Mattson <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Garrett D'Amore <[email protected]>
Approved by: Dan McDonald <[email protected]>
References:
https://www.illumos.org/issues/3603
https://www.illumos.org/issues/3604
https://www.illumos.org/issues/3871
illumos/illumos-gate@d04756377ddd1cf28ebcf652541094e17b03c889
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
Note that the patch from Illumos issue 3871 is not accepted into Illumos
at the time of this writing. It is something that I wrote when porting
this. Documentation is in the Illumos issue.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3588 provide zfs properties for logical (uncompressed) space
used and referenced
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Dan McDonald <[email protected]>
Reviewed by: Richard Elling <[email protected]>
Approved by: Richard Lowe <[email protected]>
References:
https://www.illumos.org/issues/3588
illumos/illumos-gate@77372cb0f35e8d3615ca2e16044f033397e88e21
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3578 transferring the freed map to the defer map should be constant time
3579 ztest trips assertion in metaslab_weight()
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: Dan Kimmel <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Reviewed by: Richard Elling <[email protected]>
Approved by: Dan McDonald <[email protected]>
References:
https://www.illumos.org/issues/3578
https://www.illumos.org/issues/3579
illumos/illumos-gate@9eb57f7f3fbb970d4b9b89dcd5ecf543fe2414d5
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3561 arc_meta_limit should be exposed via kstats
3116 zpool reguid may log negative guids to internal SPA history
Reviewed by: Matthew Ahrens <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Reviewed by: Gordon Ross <[email protected]>
Approved by: Garrett D'Amore <[email protected]>
References:
https://www.illumos.org/issues/3561
https://www.illumos.org/issues/3116
illumos/illumos-gate@20128a0826f9c53167caa9215c12f08beee48e30
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Porting Notes:
1. The spa change was accidentally included in the libzfs_core merge.
2. "Add missing arcstats" (1834f2d8b715d25bafbb0e4a099994f45c3211ae)
already implemented these kstats a few years ago.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3537 want pool io kstats
Reviewed by: George Wilson <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Eric Schrock <[email protected]>
Reviewed by: Sa?o Kiselkov <[email protected]>
Reviewed by: Garrett D'Amore <[email protected]>
Reviewed by: Brendan Gregg <[email protected]>
Approved by: Gordon Ross <[email protected]>
References:
http://www.illumos.org/issues/3537
illumos/illumos-gate@c3a6601
Ported by: Cyril Plisko <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Porting Notes:
1. The patch was restructured to take advantage of the existing
spa statistics infrastructure. To accomplish this the kstat
was moved in to spa->io_stats and the init/destroy code moved
to spa_stats.c.
2. The I/O kstat was simply named <pool> which conflicted with the
pool directory we had already created. Therefore it was renamed
to <pool>/io
3. An update handler was added to allow the kstat to be zeroed.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3522 zfs module should not allow uninitialized variables
Reviewed by: Sebastien Roy <[email protected]>
Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Matthew Ahrens <[email protected]>
Approved by: Garrett D'Amore <[email protected]>
References:
https://www.illumos.org/issues/3522
illumos/illumos-gate@d5285cae913f4e01ffa0e6693a6d8ef1fbea30ba
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Porting notes:
1. ZFSOnLinux had already addressed many of these issues because of
its use of -Wall. However, the manner in which they were addressed
differed. The illumos fixes replace the ones previously made in
ZFSOnLinux to reduce code differences.
2. Part of the upstream patch made a small change to arc.c that might
address zfsonlinux/zfs#1334.
3. The initialization of aclsize in zfs_log_create() differs because
vsecp is a NULL pointer on ZFSOnLinux.
4. The changes to zfs_register_callbacks() were dropped because it
has diverged and needs to be resynced.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Cstyle is the C source style checker used by Illumos. Since the
original ZFS source was written using these style guidelines they
must also be followed by ZoL for consistency.
The checker has been added to the scripts directory and may be
run on a per file basis. New patches should be careful to avoid
introducing new style warnings.
Additionally, the 'checkstyle' target has been added to the top
level Makefile and can be used to check the entire source tree.
While Zol has historically attempted to follow the SunOS style
guide the lack of a rigorous style checker has allowed various
warning to be introduced. Currently there are 2211 reported
style violations and we want to gradually eliminate these from
the tree.
Note the cstyle.1 man page is provided under man/man1/cstyle.1
but since it is a developer utility it is not installed along
with the other man pages.
Signed-off-by: Brian Behlendorf <[email protected]>
|
|
|
|
|
|
| |
This is required to make Illumos 3962 merge.
Signed-off-by: Richard Yao <[email protected]>
|
|
|
|
|
|
|
|
| |
This resolves merge conflicts when merging Illumos #3588 and Illumos #4047.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
| |
Modifying the length of a string returned by strdup() is incorrect
because strfree() is allowed to use strlen() to determine which slab
cache was used to do the allocation.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
| |
The resolution of a merge conflict when merging Illumos #3464 caused us
to invert the order couple of function calls in zio_free_sync() versus
what they are in Illumos.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
| |
This was accidentally removed by overzealous commenting.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This change adds support for Posix ACLs by storing them as an xattr
which is common practice for many Linux file systems. Since the
Posix ACL is stored as an xattr it will not overwrite any existing
ZFS/NFSv4 ACLs which may have been set. The Posix ACL will also
be non-functional on other platforms although it may be visible
as an xattr if that platform understands SA based xattrs.
By default Posix ACLs are disabled but they may be enabled with
the new 'aclmode=noacl|posixacl' property. Set the property to
'posixacl' to enable them. If ZFS/NFSv4 ACL support is ever added
an appropriate acltype will be added.
This change passes the POSIX Test Suite cleanly with the exception
of xacl/00.t test 45 which is incorrect for Linux (Ext4 fails too).
http://www.tuxera.com/community/posix-test-suite/
Signed-off-by: Massimo Maggi <[email protected]>
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #170
|
|
|
|
|
|
|
|
|
|
|
|
| |
Extend the xattr property section of zfs(8) such that it covers
both styles of supported xattr. A short discussion of the benefits
and drawbacks of each type is presented to allow users to make an
informed choice.
Signed-off-by: Massimo Maggi <[email protected]>
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #170
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Attempting to remove an xattr from a file which does not contain
any directory based xattrs would result in the xattr directory
being created. This behavior is non-optimal because it results
in write operations to the pool in addition to the expected error
being returned.
To prevent this the CREATE_XATTR_DIR flag is only passed in
zpl_xattr_set_dir() when setting a non-NULL xattr value. In
addition, zpl_xattr_set() is updated similarly such that it will
return immediately if passed an xattr name which doesn't exist
and a NULL value.
Signed-off-by: Massimo Maggi <[email protected]>
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #170
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Added a simple sed script to do a search and replace on the Illumos
ZFS file names and replace them with the ZFS on Linux equivalent.
Example usage:
# Replace Illumos paths with Linux paths
$ ./scripts/zfs2zol-patch.sed arc.c.patch > arc.c.patch.linux
# Ensure the script worked as expected
$ diff arc.c.patch arc.c.patch.linux
# Apply the patch using Linux paths
$ patch -p1 < arc.c.patch.linux
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Prakash Surya <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1679
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This does the following:
1. It creates a uint8_t type value, which is initialized to DT_DIR on
dot directories and ZFS_DIRENT_TYPE(zap.za_first_integer) otherwise.
This resolves a regression where we return unintialized values as the
directory entry type on dot directories. This was accidentally
introduced by commit 8170d281263e52ff33d7fba93ab625196844df36.
2. It restructures zfs_readdir() code to use `uint64_t offset` like
Illumos instead of `loff_t *pos`. This resolves a regression where
negative ZAP cursors were treated as if they were dot directories.
3. It restructures the function to more closely match the structure of
zfs_readdir() on Illumos and removes the unused variable outcount, which
was only used on Illumos.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1750
|
|
|
|
|
|
|
|
| |
This works the same as the -p switch to "zfs get", displaying full
resolution values for appropriate attributes.
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1813
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This change introduces zpool_get_prop_literal. It's an expanded version
of zpool_get_prop taking one additional boolean parameter. With this
parameter set to B_FALSE it will behave identically to zpool_get_prop.
Setting it to B_TRUE will return full precision numbers for the
following properties:
ZPOOL_PROP_SIZE
ZPOOL_PROP_ALLOCATED
ZPOOL_PROP_FREE
ZPOOL_PROP_FREEING
ZPOOL_PROP_EXPANDSZ
ZPOOL_PROP_ASHIFT
Also introduced is a wrapper function for zpool_get_prop making it
use zpool_get_prop_literal in the background.
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1813
|
|
|
|
|
|
|
| |
in man page and in command help.
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1805
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This branch updates several of the zfs kstats to take advantage
of the improved raw kstat functionality. In addition, two new
kstats and a script called dbufstat.py are introduced.
Updated+New Kstats
* dbufs - Stats for all dbufs in the dbuf_hash
* <pool>/txgs - Stats for the last N txgs synced to disk
* <pool>/reads - Stats for rhe last N reads issues by the ARC
* <pool>/dmu_tx_assign - Histogram of tx assign times
Signed-off-by: Brian Behlendorf <[email protected]>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The dbufstat.py command was added to provide a conveniant way to
easily determine what ZFS is caching. The script consumes the
raw /proc/spl/kstat/zfs/dbufs kstat data can consolidates it in
to a more human readable form. This was designed primarily as
a tool to aid developers but it may also be useful for advanced
users who want more visibility in to what the ARC is caching.
When run without options dbufstat.py will default to showing a
list of all objects with at least one buffer present in the
cache. The total cache space consumed by that object will be
printed on the right along with the object type. Similar to the
arcstats.py command the -x option may used to display additional
fields.
Two other modes of operation are also supported by dbufstat.py
and the expectation is additional display modes may be added as
needed. The -t option will summerize the total number of bytes
cached for each object type, and the -b option will show every
dbuf currently cached.
The script was designed to be consistent with arcstat.py and
includes most of the same options and funcationality.
Signed-off-by: Prakash Surya <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently there is no mechanism to inspect which dbufs are being
cached by the system. There are some coarse counters in arcstats
by they only give a rough idea of what's being cached. This patch
aims to improve the current situation by adding a new dbufs kstat.
When read this new kstat will walk all cached dbufs linked in to
the dbuf_hash. For each dbuf it will dump detailed information
about the buffer. It will also dump additional information about
the referenced arc buffer and its related dnode. This provides a
more complete view in to exactly what is being cached.
With this generic infrastructure in place utilities can be written
to post-process the data to understand exactly how the caching is
working. For example, the data could be processed to show a list
of all cached dnodes and how much space they're consuming. Or a
similar list could be generated based on dnode type. Many other
ways to interpret the data exist based on what kinds of questions
you're trying to answer.
Signed-off-by: Brian Behlendorf <[email protected]>
Signed-off-by: Prakash Surya <[email protected]>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This change adds a new kstat to gain some visibility into the
amount of time spent in each call to dmu_tx_assign. A histogram
is exported via the new dmu_tx_assign file. The information
contained in this histogram is the frequency dmu_tx_assign
took to complete given an interval range.
Signed-off-by: Prakash Surya <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This change is an attempt to add visibility in to how txgs are being
formed on a system, in real time. To do this, a list was added to the
in memory SPA data structure for a pool, with each element on the list
corresponding to txg. These entries are then exported through the kstat
interface, which can then be interpreted in userspace.
For each txg, the following information is exported:
* Unique txg number (uint64_t)
* The time the txd was born (hrtime_t)
(*not* wall clock time; relative to the other entries on the list)
* The current txg state ((O)pen/(Q)uiescing/(S)yncing/(C)ommitted)
* The number of reserved bytes for the txg (uint64_t)
* The number of bytes read during the txg (uint64_t)
* The number of bytes written during the txg (uint64_t)
* The number of read operations during the txg (uint64_t)
* The number of write operations during the txg (uint64_t)
* The time the txg was closed (hrtime_t)
* The time the txg was quiesced (hrtime_t)
* The time the txg was synced (hrtime_t)
Note that while the raw kstat now stores relative hrtimes for the
open, quiesce, and sync times. Those relative times are used to
calculate how long each state took and these deltas and printed by
output handlers.
Signed-off-by: Brian Behlendorf <[email protected]>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This change is an attempt to add visibility into the arc_read calls
occurring on a system, in real time. To do this, a list was added to the
in memory SPA data structure for a pool, with each element on the list
corresponding to a call to arc_read. These entries are then exported
through the kstat interface, which can then be interpreted in userspace.
For each arc_read call, the following information is exported:
* A unique identifier (uint64_t)
* The time the entry was added to the list (hrtime_t)
(*not* wall clock time; relative to the other entries on the list)
* The objset ID (uint64_t)
* The object number (uint64_t)
* The indirection level (uint64_t)
* The block ID (uint64_t)
* The name of the function originating the arc_read call (char[24])
* The arc_flags from the arc_read call (uint32_t)
* The PID of the reading thread (pid_t)
* The command or name of thread originating read (char[16])
From this exported information one can see, in real time, exactly what
is being read, what function is generating the read, and whether or not
the read was found to be already cached.
There is still some work to be done, but this should serve as a good
starting point.
Specifically, dbuf_read's are not accounted for in the currently
exported information. Thus, a follow up patch should probably be added
to export these calls that never call into arc_read (they only hit the
dbuf hash table). In addition, it might be nice to create a utility
similar to "arcstat.py" to digest the exported information and display
it in a more readable format. Or perhaps, log the information and allow
for it to be "replayed" at a later time.
Signed-off-by: Prakash Surya <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
|
| |
| |
| |
| | |
This reverts commit e95853a331529a6cb96fdf10476c53441e59f4e1.
|
|/
|
|
|
|
| |
This reverts commit 92334b14ec378b1693573b52c09816bbade9cf3e.
Signed-off-by: Brian Behlendorf <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When creating a new pool, or adding/replacing a disk in an existing
pool, partition tables will be automatically created on the devices.
Under normal circumstances it will take less than a second for udev
to create the expected device files under /dev/. However, it has
been observed that if the system is doing heavy IO concurrently udev
may take far longer. If you also throw in some cheap dodgy hardware
it may take even longer.
To prevent zpool commands from failing due to this the default wait
time for udev is being increased to 30 seconds. This will have no
impact on normal usage, the increase timeout should only be noticed
if your udev rules are incorrectly configured.
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1646
|
|
|
|
|
|
|
|
|
|
| |
Linus Torvalds merged LZ4 into Linux 3.11. This causes a conflict
whenever CONFIG_LZ4_DECOMPRESS=y or CONFIG_LZ4_COMPRESS=y are set in the
kernel's .config. We rename the symbols to avoid the conflict.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1789
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Document the "-D" and "-T" options and the optional interval
and count or "zpool status".
Also for zpool's man page, use a consistent order for the
various "-T" options to match the program's help output.
Document the effect of additional "-D" options for zdb.
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1786
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The semantics introduced by the restructured sync task of illumos
3464 require this lock when calling dmu_snapshot_list_next().
The pool is locked/unlocked for each iteration to reduce the
chance of long-running locks.
This was accidentally missed when doing the original port because
ZoL's control directory code is Linux-specific and is in a
different file than in illumos.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #1785
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
3552 condensing one space map burns 3 seconds of CPU in spa_sync()
thread (fix race condition)
References:
https://www.illumos.org/issues/3552
illumos/illumos-gate@03f8c366886542ed249a15d755ae78ea4e775d9d
Ported-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Porting notes:
This fixes an upstream regression that was introduced in commit
zfsonlinux/zfs@e51be06697762215dc3b679f8668987034a5a048, which
ported the Illumos 3552 changes. This fix was added to upstream
rather quickly, but at the time of the port, no one spotted it and
the race was rare enough that it passed our regression tests. I
discovered this when comparing our metaslab.c to the illumos
metaslab.c.
Without this change it is possible for metaslab_group_alloc() to
consume a large amount of cpu time. Since this occurs under a
mutex in a rcu critical section the kernel will log this to the
console as a self-detected cpu stall as follows:
INFO: rcu_sched self-detected stall on CPU { 0}
(t=60000 jiffies g=11431890 c=11431889 q=18271)
Closes #1687
Closes #1720
Closes #1731
Closes #1747
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The GNU libtool documentation states to start with a version of 0:0:0,
rather than 1:1:0. Illumos uses the name libzfs_core.so.1, so to be
consistent, we should go with 1:0:0.
http://www.gnu.org/software/libtool/manual/libtool.html#Updating-version-info
The GNU libtool documentation also provides guidence on how the version
information should be incremented. Doing this does a SONAME bump of the
libzfs and libzpool libraries. This is particularly important on Gentoo
because a SONAME bump enables portage to retain the older libraries
until any packages that link to them are rebuilt. The main example of
this is GRUB2's grub2-mkconfig, which will break unless it is rebuilt
against the new libraries.
Signed-off-by: Richard Yao <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Issue #1751
|