summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Add nfs-kernel-server for DebianGrischa Zengel2016-02-251-2/+2
| | | | | | | | | | Debian based systems use nfs-kernel-server as the service name. List both nfs-server.service and nfs-kernel-server.service so this service will work on multiple distributions. Signed-off-by: Grischa Zengel <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4350
* Add l2arc_max_block_size tunableBrian Behlendorf2016-02-252-1/+47
| | | | | | | | | | | Set a limit for the largest compressed block which can be written to an L2ARC device. By default this limit is set to 16M so there is no change in behavior. Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Richard Elling <[email protected]> Signed-off-by: Tim Chase <[email protected]> Closes #4323
* Make zvol minor functionality more robustBoris Protopopov2016-02-241-16/+77
| | | | | | | | | | | | | | | | | | | | | | | | Close the race window in zvol_open() to prevent removal of zvol_state in the 'first open' code path. Move the call to check_disk_change() under zvol_state_lock to make sure the zvol_media_changed() and zvol_revalidate_disk() called by check_disk_change() are invoked with positive zv_open_count. Skip opened zvols when removing minors and set private_data to NULL for zvols that are not in use whose minors are being removed, to indicate to zvol_open() that the state is gone. Skip opened zvols when renaming minors to avoid modifying zv_name that might be in use, e.g. in zvol_ioctl(). Drop zvol_state_lock before calling add_disk() when creating minors to avoid deadlocks with zvol_open(). Wrap dmu_objset_find() with spl_fstran_mark()/unmark(). Signed-off-by: Boris Protopopov <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Richard Yao <[email protected]> Closes #4344
* Call dmu_read_uio_dbuf() in zvol_read()Richard Yao2016-02-181-1/+1
| | | | | | | | | | | | | | | | | The difference between `dmu_read_uio()` and `dmu_read_uio_dbuf()` is that the former takes a hold while the latter uses an existing hold. `zfs_read()` in the ZPL will use `dmu_read_uio_dbuf()` while our analogous `zvol_write()` will use `dmu_write_uio_dbuf()`, but for no apparent reason, we inherited a `zvol_read()` function from OpenSolaris that does `dmu_read_uio()`. illumos-gate also still uses `dmu_read_uio()` to this day. Lets switch to `dmu_read_uio_dbuf()`, which is more performant. Signed-off-by: Richard Yao <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Closes #4316
* Clean up zvol request processing to pass uio and fix porting regressionsRichard Yao2016-02-181-76/+71
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | In illumos-gate, `zvol_read` and `zvol_write` are both passed uio_t rather than bio_t. Since we are translating from bio to uio for both, we might as well unify the logic and have code more similar to its illumos counterpart. At the same time, we can fix some regressions that occurred versus the original code from illumos-gate. We refactor zvol_write to take uio and also correct the following problems: 1. We did `dnode_hold()` on each IO when we already had a hold. 2. We would attempt to send writes that exceeded `DMU_MAX_ACCESS` to the DMU. 3. We could call `zil_commit()` twice. In this case, this is because Linux uses the `->write` function to send flushes and can aggregate the flush with a write. If a synchronous write occurred with the flush, we effectively flushed twice when there is no need to do that. zvol_read also suffers from the first two problems. Other platforms suffer from the first, so we leave that for a second patch so that there is a discrete patch for them to cherry-pick. Signed-off-by: Richard Yao <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Closes #4316
* Make configure error clearer when failing to find SPLOlaf Faaland2016-02-171-3/+23
| | | | | | | Signed-off-by: Olaf Faaland <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Richard Yao <[email protected]> Closes #4251
* Remove wrong ASSERT in annotate_ecksumChunwei Chen2016-02-171-2/+2
| | | | | | | | | | When using large blocks like 1M, there will be more than UINT16_MAX qwords in one block, so this ASSERT would go off. Also, it is possible for the histogram to overflow. We cap them to UINT16_MAX to prevent this. Signed-off-by: Chunwei Chen <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4257
* Prevent zpool_find_vdev() from truncating vdev pathBrian Behlendorf2016-02-081-2/+4
| | | | | | | | | | When extracting tokens from the string strtok(2) is allowed to modify the passed buffer. Therefore the zfs_strcmp_pathname() function must make a copy of the passed string before passing it to strtok(3). Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Don Brady <[email protected]> Closes #4312
* Illumos 5809 - Blowaway full receive in v1 pool causes kernel panicPaul Dagnelie2016-02-081-1/+2
| | | | | | | | | | | | | | 5809 Blowaway full receive in v1 pool causes kernel panic Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Alex Reece <[email protected]> Reviewed by: Will Andrews <[email protected]> Approved by: Gordon Ross <[email protected]> References: https://www.illumos.org/issues/5809 https://github.com/illumos/illumos-gate/commit/f40b29c Ported-by: Brian Behlendorf <[email protected]>
* Illumos 5767 - fix several problems with zfs test suiteJohn Wren Kennedy2016-02-081-10/+36
| | | | | | | | | | | | | | | | | | | | 5767 fix several problems with zfs test suite Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Christopher Siden <[email protected]> Approved by: Gordon Ross <[email protected]> References: https://www.illumos.org/issues/5767 https://github.com/illumos/illumos-gate/commit/52244c0 Porting Notes: - Only the updates to zpool_main.c were kept because the ZFS test suite is not currently part of the ZoL source tree. The test suite itself should be updated to include the latest versions of the tests once we're running it for every commit - Fixes `zpool list` output. Ported-by: Brian Behlendorf <[email protected]>
* Illumos 4448 - zfs diff misprints unicode charactersJoshua M. Clulow2016-02-051-6/+9
| | | | | | | | | | | | | | | | | | | 4448 zfs diff misprints unicode characters Reviewed by: Igor Kozhukhov <[email protected]> Reviewed by: Toomas Soome <[email protected]> Approved by: Matthew Ahrens <[email protected]> References: https://www.illumos.org/issues/4448 https://github.com/illumos/illumos-gate/commit/b211eb9 Porting Notes: - [lib/libzfs/libzfs_diff.c] - 38145d6 Ensure that zfs diff prints unicode safely. - 141b638 Change 3-digit octal escapes to 4-digit ones Ported-by: kernelOfTruth [email protected] Signed-off-by: Brian Behlendorf <[email protected]>
* Correctly parse -R flag argumentsTim Chase2016-02-051-1/+3
| | | | | | | | | | Currently, only the 'b' flag takes an argument which is an offset into the block at which a blkptr should be decoded. The index into the flag string needed to be updated after parsing an argument. Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4304
* Illumos 6537 - Panic on zpool scrub with DEBUG kernelGary Mills2016-02-051-2/+4
| | | | | | | | | | | | | | | 6537 Panic on zpool scrub with DEBUG kernel Reviewed by: Steve Gonczi <[email protected]> Reviewed by: Dan McDonald <[email protected]> Reviewed by: Igor Kozhukhov <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Approved by: Matthew Ahrens <[email protected]> References: https://www.illumos.org/issues/6537 https://github.com/illumos/illumos-gate/commit/8c04a1f Ported-by: Brian Behlendorf <[email protected]>
* Illumos 6096 - ZFS_SMB_ACL_RENAME needs to cleanup betterDan McDonald2016-02-051-0/+1
| | | | | | | | | | | | | | 6096 ZFS_SMB_ACL_RENAME needs to cleanup better Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Gordon Ross <[email protected]> Reviewed by: George Wilson <[email protected]> Approved by: Robert Mustacchi <[email protected]> References: https://www.illumos.org/issues/6096 https://github.com/illumos/illumos-gate/commit/8f5190a5 Ported-by: Brian Behlendorf <[email protected]>
* Illumos 6450 - scrub/resilver unnecessarily traverses snapshotsMatthew Ahrens2016-02-021-4/+50
| | | | | | | | | | | | | | | | 6450 scrub/resilver unnecessarily traverses snapshots created after the scrub started Reviewed by: George Wilson <[email protected]> Reviewed by: Prakash Surya <[email protected]> Reviewed by: Richard Elling <[email protected]> Approved by: Richard Lowe <[email protected]> References: https://www.illumos.org/issues/6450 https://github.com/illumos/illumos-gate/commit/38d6103 Ported-by: kernelOfTruth [email protected] Signed-off-by: Brian Behlendorf <[email protected]>
* Handling negative dentries in a CI file system.Richard Sharpe2016-02-021-2/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | For a Case Insensitive file system we must avoid creating negative entries in the dentry cache. We must also pass the FIGNORECASE into zfs_lookup so that special files are handled correctly. We must also prevent negative dentries from being created when files are unlinked. Tested by running fsstress from LTP (10 loops, 10 processes, 10,000 ops.) Also tested with printks (now removed) to ensure that lookups come to zpl_lookup when negative should not exist. Tests: 1. ls Some-file.txt; touch some-file.txt; ls Some-file.txt and ensure no errors. 2. touch Some-file.txt; rm some-file.txt; ls Some-file.txt and ensure that the last ls shows log messages showing the lookup went all the way to zpl_lookup. Thanks to tuxoko for helping me get this correct. Signed-off-by: Richard Sharpe <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4243
* Illumos 6527 - Possible access beyond end of string in zpool commentJorgen Lundman2016-01-281-1/+0
| | | | | | | | | | | | | | | 6527 Possible access beyond end of string in zpool comment Reviewed by: George Wilson <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Dan McDonald <[email protected]> Approved by: Gordon Ross <[email protected]> References: https://www.illumos.org/issues/6527 https://github.com/illumos/illumos-gate/commit/2bd7a8d Ported-by: Brian Behlendorf <[email protected]> Signed-off-by: Chunwei Chen <[email protected]>
* Illumos 6495 - Fix mutex leak in dmu_objset_find_dpBrian Behlendorf2016-01-281-0/+3
| | | | | | | | | | | | | | 6495 Fix mutex leak in dmu_objset_find_dp Reviewed by: George Wilson <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Approved by: Albert Lee <[email protected]> References: https://www.illumos.org/issues/6495 https://github.com/illumos/illumos-gate/commit/2bad225 Ported-by: Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Chunwei Chen <[email protected]>
* Illumos 6414 - vdev_config_sync could be simplerBrian Behlendorf2016-01-283-16/+16
| | | | | | | | | | | | | | 6414 vdev_config_sync could be simpler Reviewed by: George Wilson <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Approved by: Robert Mustacchi <[email protected]> References: https://www.illumos.org/issues/6414 https://github.com/illumos/illumos-gate/commit/eb5bb58 Ported-by: Brian Behlendorf <[email protected]> Signed-off-by: Chunwei Chen <[email protected]>
* llumos 6334 - Cannot unlink files when over quotaSimon Klinkert2016-01-261-5/+2
| | | | | | | | | | | | | 6334 Cannot unlink files when over quota Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Toomas Soome <[email protected]> Approved by: Dan McDonald <[email protected]> References: https://www.illumos.org/issues/6334 https://github.com/illumos/illumos-gate/commit/6575bca Ported-by: Brian Behlendorf <[email protected]>
* Reintroduce zfs_remove() synchronous deleteskernelOfTruth2016-01-262-12/+81
| | | | | | | | | | | | | | | | | | | | | | | | Reintroduce a slightly adapted version of the Illumos logic for synchronous unlinks. The basic idea here is that only files smaller than zfs_delete_blocks (20480) blocks should be deleted synchronously. Unlinking larger files should be handled asynchronously to minimize impact to the caller. To accomplish this iput() which is responsible for calling zfs_znode_delete() on Linux is only called in the delete_now path. Otherwise zfs_async_iput() is used which allows the last reference to be dropped by a taskq thread effectively making the removal asynchronous. Porting notes: - Add zfs_delete_blocks module option for performance analysis. The default value is DMU_MAX_DELETEBLKCNT which is the same as upstream. Reducing this value means that smaller files will be unlinked asynchronously like large files. - All occurrences of zfsvfs changes to zsb. Ported-by: KernelOfTruth [email protected] Signed-off-by: Brian Behlendorf <[email protected]>
* Log zvol truncate/discard operationsDan McDonald2016-01-261-8/+59
| | | | | | | | | | | | | | As the comments in zvol_discard() suggested, the discard operation could be logged to the zil. This is a port of the relevant code from Nexenta as it was added in "701 UNMAP support for COMSTAR" and has been attributed to the author of that commit. References: https://github.com/Nexenta/illumos-nexenta/commit/b77b923 https://github.com/zfsonlinux/zfs/blob/089fa91b/module/zfs/zvol.c#L637 Ported-by: Tim Chase <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]>
* mount.zfs: use getopt_long instead of getopt to guarantee permutation of argv.Christian Neukirchen2016-01-251-1/+2
| | | | | | | | | | | | | | | | mount.zfs is called by convention (and util-linux) with arguments last, i.e. % mount.zfs <dataset> <mountpoint> -o <options> This is not a problem on glibc since GNU getopt(3) will reorder the arguments. However, alternative libc such as musl libc (or glibc with $POSIXLY_CORRECT set) will not permute argv and fail to parse the -o <options>. Use getopt_long so musl will permute arguments. Signed-off-by: Christian Neukirchen <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4222
* Illumos 6451 - ztest fails due to checksum errorsMatthew Ahrens2016-01-251-1/+26
| | | | | | | | | | | | | | 6451 ztest fails due to checksum errors Reviewed by: George Wilson <[email protected]> Reviewed by: Prakash Surya <[email protected]> Reviewed by: Jorgen Lundman <[email protected]> Approved by: Dan McDonald <[email protected]> References: https://www.illumos.org/issues/6451 https://github.com/illumos/illumos-gate/commit/f9eb9fd Ported-by: Brian Behlendorf <[email protected]>
* Illumos 6251 - add tunable to disable free_bpobj processingGeorge Wilson2016-01-252-2/+22
| | | | | | | | | | | | | | | | | | | | | 6251 - add tunable to disable free_bpobj processing Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Prakash Surya <[email protected]> Reviewed by: Simon Klinkert <[email protected]> Reviewed by: Richard Elling <[email protected]> Reviewed by: Albert Lee <[email protected]> Reviewed by: Xin Li <[email protected]> Approved by: Garrett D'Amore <[email protected]> References: https://www.illumos.org/issues/6251 https://github.com/illumos/illumos-gate/commit/139510f Porting notes: - Added as module option declaration. - Added to zfs-module-parameters.5 man page. Ported-by: Signed-off-by: Brian Behlendorf <[email protected]>
* Set arc_c_min properly in userland buildsTim Chase2016-01-251-4/+4
| | | | | | | | | | Since it's set to arc_c_max / 2, it must be set after arc_c_max is set. Also added protection against it falling below 2 * maxblocksize in userland builds. Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4268
* Prevent arc_c collapseTim Chase2016-01-251-8/+8
| | | | | | | | | | | | | | | Adjusting arc_c directly is racy because it can happen in the context of multiple threads. It should always be >= 2 * maxblocksize. Set it to a known valid value rather than adjusting it directly. In addition refactor arc_shrink() to a simpler structure, protect against underflow in the calculation of the new arc_c value. Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Reverts: 935434ef Closes: #3904 Closes: #4161
* kobj_read_file: Return -1 on vn_rdwr() errorRichard Yao2016-01-231-2/+3
| | | | | | | | | | | | LLVM's static analyzer showed that we could subtract using an uninitialized value on an error from vn_rdwr(). The correct behavior is to return -1 on an error, so lets do that instead. Signed-off-by: Richard Yao <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4104
* Illumos 1778 - Assertion failed: rn->rn_nozpool == B_FALSEAndrew Stormont2016-01-221-3/+1
| | | | | | | | | | | | | | 1778 Assertion failed: rn->rn_nozpool == B_FALSE Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: George Wilson <[email protected]> Reviewed by: Richard Elling <[email protected]> Approved by: Gordon Ross <[email protected]> References: https://www.illumos.org/issues/1778 https://github.com/illumos/illumos-gate/commit/bd0f709 Ported-by: Brian Behlendorf <[email protected]>
* Illumos 5518 - Memory leaks in libzfs import implementationMarcel Telka2016-01-222-33/+43
| | | | | | | | | | | | | | | | | | 5518 Memory leaks in libzfs import implementation Reviewed by: Dan Fields <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Serghei Samsi <[email protected]> Approved by: Dan McDonald <[email protected]> References: https://www.illumos.org/issues/5518 https://github.com/illumos/illumos-gate/commit/078266a Porting notes: - One hunk of this change was already applied independently in commit 4def05f. Ported-by: Brian Behlendorf <[email protected]>
* Illumos 6815179, 6844191Brian Behlendorf2016-01-223-64/+294
| | | | | | | | | | | | | | | | | | | | | | | | | 6815179 zpool import with a large number of LUNs is too slow 6844191 zpool import, scanning of disks should be multi-threaded References: https://github.com/illumos/illumos-gate/commit/4f67d75 Porting notes: - This change was originally never ported to Linux due to it dependence on the thread pool interface. This patch solves that issue by switching the code to use the existing taskq implementation which provides the same basic functionality. However, in order for this to work properly thread_init() and thread_fini() must be called around to taskq consumer to perform the needed thread initialization. - The check_one_slice, nozpool_all_slices, and check_slices functions have been disabled for Linux. They are difficult, but possible, to implement for Linux due to how partitions are get names. Since this is only an optimization this code can be added at a latter date. Signed-off-by: Brian Behlendorf <[email protected]>
* Illumos 4950 - files sometimes can't be removed from a full filesystemMatthew Ahrens2016-01-217-7/+57
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <[email protected]> Reviewed by: George Wilson <[email protected]> Reviewed by: Sebastien Roy <[email protected]> Reviewed by: Boris Protopopov <[email protected]> Approved by: Dan McDonald <[email protected]> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]>
* Create zfs-kmod-debuginfo rpm with redhat spec fileOlaf Faaland2016-01-212-1/+4
| | | | | | | | | | Correct the redhat specfile so that working debuginfo rpms are created for the kernel modules. The generic specfile already does the right thing. Signed-off-by: Olaf Faaland <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4224
* Close possible zfs_znode_held() raceBrian Behlendorf2016-01-201-4/+3
| | | | | | | | | | | | Check if the lock is held while holding the z_hold_locks() lock. This prevents a possible use-after-free bug for callers which are not holding the lock. There currently are no such callers so this can't cause a problem today but it has been fixed regardless. Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Closes #4244 Issue #4124
* Linux 4.5 compat: pfn_t typedefBrian Behlendorf2016-01-201-1/+0
| | | | | | | | | | | | The pfn_t typedef was inherited from Illumos but never directly used by any libspl consumers. This doesn't cause any issues in user space but for consistency with the kernel build it has been removed. See torvalds/linux/commit/34c0fd54. Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Issue #4228
* Linux 4.5 compat: xattr list handlerBrian Behlendorf2016-01-203-282/+518
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The registered xattr .list handler was simplified in the 4.5 kernel to only perform a permission check. Given a dentry for the file it must return a boolean indicating if the name is visible. This differs slightly from the previous APIs which also required the function to copy the name in to the provided list and return its size. That is now all the responsibility of the caller. This should be straight forward change to make to ZoL since we've always required the caller to make the copy. However, this was slightly complicated by the need to support 3 older APIs. Yes, between 2.6.32 and 4.5 there are 4 versions of this interface! Therefore, while the functional change in this patch is small it includes significant cleanup to make the code understandable and maintainable. These changes include: - Improved configure checks for .list, .get, and .set interfaces. - Interfaces checked from newest to oldest. - Strict checking for each possible known interface. - Configure fails when no known interface is available. - HAVE_*_XATTR_LIST renamed HAVE_XATTR_LIST_* for consistency with similar iops and fops configure checks. - POSIX_ACL_XATTR_{DEFAULT|ACCESS} were removed forcing callers to move to their replacements, XATTR_NAME_POSIX_ACL_{DEFAULT|ACCESS}. Compatibility wrapper were added for old kernels. - ZPL_XATTR_LIST_WRAPPER added which behaves the same as the existing ZPL_XATTR_{GET|SET} WRAPPERs. Only the inode is guaranteed to be a valid pointer, passing NULL for the 'list' and 'name' variables is allowed and must be checked for. All .list functions were updated to use the wrapper to aid readability. - zpl_xattr_filldir() updated to use the .list function for its permission check which is consistent with the updated Linux 4.5 interface. If a .list function is registered it should return 0 to indicate a name should be skipped, if there is no registered function the name will be added. - Additional documentation from xattr(7) describing the correct behavior for each namespace was added before the relevant handlers. Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Issue #4228
* Linux 4.5 compat: get_link() / put_link()Brian Behlendorf2016-01-206-81/+253
| | | | | | | | | | | | | | | | | | | | The follow_link() interface was retired in favor of get_link(). In the process of phasing in get_link() the Linux kernel went through two different versions. The first of which depended on put_link() and the final version on a delayed done function. - Improved configure checks for .follow_link, .get_link, .put_link. - Interfaces checked from newest to oldest. - Strict checking for each possible known interface. - Configure fails when no known interface is available. - Both versions .get_link are detected and supported as well two previous versions of .follow_link. Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Tim Chase <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Issue #4228
* Illumos 5045 - use atomic_{inc,dec}_* instead of atomic_add_*Josef 'Jeff' Sipek2016-01-157-73/+70
| | | | | | | | | | | | | | | | | | 5045 use atomic_{inc,dec}_* instead of atomic_add_* Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Garrett D'Amore <[email protected]> Approved by: Robert Mustacchi <[email protected]> References: https://www.illumos.org/issues/5045 https://github.com/illumos/illumos-gate/commit/1a5e258 Porting notes: - All changes to non-ZFS files dropped. - Changes to zfs_vfsops.c dropped because they were Illumos specific. Ported-by: Brian Behlendorf <[email protected]> Closes #4220
* Illumos 4953, 4954, 4955Matthew Ahrens2016-01-152-15/+15
| | | | | | | | | | | | | | | | | | | | | | | 4953 zfs rename <snapshot> need not involve libshare 4954 "zfs create" need not involve libshare if we are not sharing 4955 libshare's get_zfs_dataset need not sort the datasets Reviewed by: George Wilson <[email protected]> Reviewed by: Sebastien Roy <[email protected]> Reviewed by: Dan McDonald <[email protected]> Reviewed by: Gordon Ross <[email protected]> Approved by: Garrett D'Amore <[email protected]> References: https://www.illumos.org/issues/4953 https://www.illumos.org/issues/4954 https://www.illumos.org/issues/4955 https://github.com/illumos/illumos-gate/commit/33cde0d Porting notes: - Dropped qsort libshare_zfs.c hunk, no equivalent ZoL code. Ported-by: Brian Behlendorf <[email protected]> Closes #4219
* Illumos 4039 - zfs_rename()/zfs_link() needs stronger test for XDEVMarcel Telka2016-01-151-5/+14
| | | | | | | | | | | | | | | | | | | | | 4039 zfs_rename()/zfs_link() needs stronger test for XDEV Reviewed by: Gordon Ross <[email protected]> Reviewed by: Kevin Crowe <[email protected]> Reviewed by: Saso Kiselkov <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Approved by: Dan McDonald <[email protected]> References: https://www.illumos.org/issues/4039 https://github.com/illumos/illumos-gate/commit/18e6497 Porting notes: - This check was updated in Linux in a similar fashion early on in the port. Therefore, this patch just reorders the function and updates the comment so it flows the same way as the upstream code. Ported-by: Brian Behlendorf <[email protected]> Closes #4218
* Illumos 6298 - zfs_create_008_neg and zpool_create_023_negJoe Stein2016-01-154-24/+59
| | | | | | | | | | | | | | | 6298 zfs_create_008_neg and zpool_create_023_neg need to be updated for large block support Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: John Kennedy <[email protected]> Approved by: Robert Mustacchi <[email protected]> References: https://www.illumos.org/issues/6298 https://github.com/illumos/illumos-gate/commit/e9316f7 Ported-by: Brian Behlendorf <[email protected]> Closes #4217
* Illumos 3557, 3558, 3559, 3560George Wilson2016-01-155-36/+27
| | | | | | | | | | | | | | | | | | | | | | | | | 3557 dumpvp_size is not updated correctly when a dump zvol's size is changed 3558 setting the volsize on a dump device does not return back ENOSPC 3559 setting a volsize larger than the space available sometimes succeeds 3560 dumpadm should be able to remove a dump device Reviewed by: Adam Leventhal <[email protected]> Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Christopher Siden <[email protected]> Approved by: Albert Lee <[email protected]> References: https://www.illumos.org/issues/3559 https://github.com/illumos/illumos-gate/commit/c61ea56 Porting notes: - Internal zvol.c changes not applied due to implementation differences. The external interface and behavior was already consistent with the latest upstream code. - Retired 2.6.28 HAVE_CHECK_DISK_SIZE_CHANGE configure check. All supported kernels (2.6.32 and newer) provide this interface. Ported-by: Brian Behlendorf <[email protected]> Closes #4217
* Prevent duplicated xattr between SA and dirChunwei Chen2016-01-151-5/+68
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When replacing an xattr would cause overflowing in SA, we would fallback to xattr dir. However, current implementation don't clear the one in SA, so we would end up with duplicated SA. For example, running the following script on an xattr=sa filesystem would cause duplicated "user.1". -- dup_xattr.sh begin -- randbase64() { dd if=/dev/urandom bs=1 count=$1 2>/dev/null | openssl enc -a -A } file=$1 touch $file setfattr -h -n user.1 -v `randbase64 5000` $file setfattr -h -n user.2 -v `randbase64 20000` $file setfattr -h -n user.3 -v `randbase64 20000` $file setfattr -h -n user.1 -v `randbase64 20000` $file getfattr -m. -d $file -- dup_xattr.sh end -- Also, when a filesystem is switch from xattr=sa to xattr=on, it will never modify those in SA. This would cause strange behavior like, you cannot delete an xattr, or setxattr would cause duplicate and the result would not match when you getxattr. For example, the following shell sequence. -- shell begin -- $ sudo zfs set xattr=sa pp/fs0 $ touch zzz $ setfattr -n user.test -v asdf zzz $ sudo zfs set xattr=on pp/fs0 $ setfattr -x user.test zzz setfattr: zzz: No such attribute $ getfattr -d zzz user.test="asdf" $ setfattr -n user.test -v zxcv zzz $ getfattr -d zzz user.test="asdf" user.test="asdf" -- shell end -- We fix this behavior, by first finding where the xattr resides before setxattr. Then, after we successfully updated the xattr in one location, we will clear the other location. Note that, because update and clear are not in single tx, we could still end up with duplicated xattr. But by doing setxattr again, it can be fixed. Signed-off-by: Chunwei Chen <[email protected]> Closes #3472 Closes #4153
* SET_ERROR should print stringsRichard Yao2016-01-153-19/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When debugging with tracepoints, we see string pointers: zfs 3017 [006] 8878.728915: zfs:zfs_set__error: ffffffffa0eec3fc:3013:ffffffffa0ebcd60(): error 0x2 ffffffffa0e1ca43 spa_open_common (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffffa0e1cbe3 spa_open (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffffa0e6f6ef zfs_ioc_stable (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffffa0e6f2a9 zfsdev_ioctl (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffff811909dd do_vfs_ioctl ([kernel.kallsyms]) ffffffff81190c41 sys_ioctl ([kernel.kallsyms]) ffffffff8156e2e9 system_call_fastpath ([kernel.kallsyms]) 7ff7d8be69c7 __GI___ioctl (/lib64/libc-2.19.so) 7ff7d90cac53 lzc_ioctl.constprop.3 (/lib64/libzfs_core.so.1.0.0) 636f695f637a6c00 [unknown] ([unknown]) Printing the actual strings is more convenient: zfs 3461 [001] 10599.847692: zfs:zfs_set__error: spa.c:3013:spa_open_common(): error 0x2 ffffffffa116ba43 spa_open_common (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffffa116bbe3 spa_open (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffffa11be8df zfs_ioc_stable (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffffa11be499 zfsdev_ioctl (/lib/modules/3.12.21-gentoo-r1/extra/zfs/zfs.ko) ffffffff811909dd do_vfs_ioctl ([kernel.kallsyms]) ffffffff81190c41 sys_ioctl ([kernel.kallsyms]) ffffffff8156e2e9 system_call_fastpath ([kernel.kallsyms]) 7f11b843c9c7 __GI___ioctl (/lib64/libc-2.19.so) 7f11b8920c53 lzc_ioctl.constprop.3 (/lib64/libzfs_core.so.1.0.0) 636f695f637a6c00 [unknown] ([unknown]) A few other tracepoints have strings as well, so switch to printing the actual string values at the same time. Signed-off-by: Richard Yao <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #4212
* Remove fastwrite mutexRichard Yao2016-01-152-10/+0
| | | | | | | | | | | | | | | | The fast write mutex is intended to protect accounting, but it is redundant because all accounting is performed through atomic operations. It also serializes all metaslab IO behind a mutex, which introduces a theoretical scaling regression that the Illumos developers did not like when we showed this to them. Removing it makes the selection of the metaslab_group lock free as it is on Illumos. The selection is not quite the same without the lock because the loop races with IO completions, but any imbalances caused by this are likely to be corrected by subsequent metaslab group selections. Signed-off-by: Richard Yao <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #3643
* Fix zsb->z_hold_mtx deadlockBrian Behlendorf2016-01-154-64/+218
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <[email protected]> Issue #4106
* Add zfs_object_mutex_size module optionBrian Behlendorf2016-01-154-12/+26
| | | | | | | | | | | Add a zfs_object_mutex_size module option to facilitate resizing the the per-dataset znode mutex array. Increasing this value may help make the deadlock described in #4106 less common, but this is not a proper fix. This patch is primarily to aid debugging and analysis. Signed-off-by: Brian Behlendorf <[email protected]> Signed-off-by: Tim Chase <[email protected]> Issue #4106
* Illumos 3465, 3466, 3467, 3468, 3470, 3473Brian Behlendorf2016-01-131-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | 3465 ::walk ... | ::<dcmd> misinterprets input as symbol names 3466 ::tsd should handle missing/NULL values better 3467 mdb_ctf_vread() could be more useful 3468 mdb enhancements for zfs development 3470 ::whatis does not print callers from KMF_LITE 3473 mdb_get_module() returns wrong module Reviewed by: Adam Leventhal <[email protected]> Reviewed by: Eric Schrock <[email protected]> Reviewed by: Dan Kimmel <[email protected]> Reviewed by: Robert Mustacchi <[email protected]> Approved by: Dan McDonald <[email protected]> References: https://www.illumos.org/issues/3468 https://github.com/illumos/illumos-gate/commit/28e4da2 Porting notes: - The only portion of this patch which applies to ZoL is a small change to types used in the refcount structure. Ported-by: Brian Behlendorf <[email protected]> Closes #4216
* Increase default user space stack sizeBrian Behlendorf2016-01-133-6/+5
| | | | | | | | | | | | | | | | | | | | Under RHEL6/CentOS6 the default stack size must be increased to 32K to prevent overflowing the stack when running ztest. This isn't an issue for other distributions due to either the version of pthreads or perhaps the compiler. Doubling the stack size resolves the issue safely for all distribution and leaves us some headroom. $ sudo -E ztest -V -T 300 -f /var/tmp/ 5 vdevs, 7 datasets, 23 threads, 300 seconds... loading space map for vdev 0 of 1, metaslab 0 of 30 ... ... loading space map for vdev 0 of 1, metaslab 14 of 30 ... child died with signal 11 Exited ztest with error 3 Signed-off-by: Brian Behlendorf <[email protected]> Closes #4215
* Illumos 3749 - zfs event processing should work on R/O root filesystemsWill Andrews2016-01-124-20/+77
| | | | | | | | | | | | | | | | | | | | | | | | | 3749 zfs event processing should work on R/O root filesystems Reviewed by: Matthew Ahrens <[email protected]> Reviewed by: Eric Schrock <[email protected]> Approved by: Christopher Siden <[email protected]> References: https://www.illumos.org/issues/3749 https://github.com/illumos/illumos-gate/commit/3cb69f7 Porting notes: - [include/sys/spa_impl.h] - ffe9d38 Add generic errata infrastructure - 1421c89 Add visibility in to arc_read - [include/sys/fm/fs/zfs.h] - 2668527 Add linux events - 6283f55 Support custom build directories and move includes - [module/zfs/spa_config.c] - Updated spa_config_sync() to match illumos with the exception of a Linux specific block. Ported-by: kernelOfTruth [email protected] Signed-off-by: Brian Behlendorf <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-17 22:15:42 +0000