freebsd-nq/module/zfs/spa_misc.c

2921 lines
75 KiB
C
Raw Normal View History

2008-11-20 20:01:55 +00:00
/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or http://www.opensolaris.org/os/licensing.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
* Copyright (c) 2011, 2019 by Delphix. All rights reserved.
* Copyright 2015 Nexenta Systems, Inc. All rights reserved.
* Copyright (c) 2014 Spectra Logic Corporation, All rights reserved.
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
* Copyright 2013 Saso Kiselkov. All rights reserved.
* Copyright (c) 2017 Datto Inc.
* Copyright (c) 2017, Intel Corporation.
* Copyright (c) 2019, loli10K <ezomori.nozomu@gmail.com>. All rights reserved.
2008-11-20 20:01:55 +00:00
*/
#include <sys/zfs_context.h>
#include <sys/spa_impl.h>
#include <sys/zio.h>
#include <sys/zio_checksum.h>
#include <sys/zio_compress.h>
#include <sys/dmu.h>
#include <sys/dmu_tx.h>
#include <sys/zap.h>
#include <sys/zil.h>
#include <sys/vdev_impl.h>
OpenZFS 9102 - zfs should be able to initialize storage devices PROBLEM ======== The first access to a block incurs a performance penalty on some platforms (e.g. AWS's EBS, VMware VMDKs). Therefore we recommend that volumes are "thick provisioned", where supported by the platform (VMware). This can create a large delay in getting a new virtual machines up and running (or adding storage to an existing Engine). If the thick provision step is omitted, write performance will be suboptimal until all blocks on the LUN have been written. SOLUTION ========= This feature introduces a way to 'initialize' the disks at install or in the background to make sure we don't incur this first read penalty. When an entire LUN is added to ZFS, we make all space available immediately, and allow ZFS to find unallocated space and zero it out. This works with concurrent writes to arbitrary offsets, ensuring that we don't zero out something that has been (or is in the middle of being) written. This scheme can also be applied to existing pools (affecting only free regions on the vdev). Detailed design: - new subcommand:zpool initialize [-cs] <pool> [<vdev> ...] - start, suspend, or cancel initialization - Creates new open-context thread for each vdev - Thread iterates through all metaslabs in this vdev - Each metaslab: - select a metaslab - load the metaslab - mark the metaslab as being zeroed - walk all free ranges within that metaslab and translate them to ranges on the leaf vdev - issue a "zeroing" I/O on the leaf vdev that corresponds to a free range on the metaslab we're working on - continue until all free ranges for this metaslab have been "zeroed" - reset/unmark the metaslab being zeroed - if more metaslabs exist, then repeat above tasks. - if no more metaslabs, then we're done. - progress for the initialization is stored on-disk in the vdev’s leaf zap object. The following information is stored: - the last offset that has been initialized - the state of the initialization process (i.e. active, suspended, or canceled) - the start time for the initialization - progress is reported via the zpool status command and shows information for each of the vdevs that are initializing Porting notes: - Added zfs_initialize_value module parameter to set the pattern written by "zpool initialize". - Added zfs_vdev_{initializing,removal}_{min,max}_active module options. Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: John Wren Kennedy <john.kennedy@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: loli10K <ezomori.nozomu@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Signed-off-by: Tim Chase <tim@chase2k.com> Ported-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/9102 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/c3963210eb Closes #8230
2018-12-19 14:54:59 +00:00
#include <sys/vdev_initialize.h>
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
#include <sys/vdev_trim.h>
#include <sys/vdev_file.h>
SIMD implementation of vdev_raidz generate and reconstruct routines This is a new implementation of RAIDZ1/2/3 routines using x86_64 scalar, SSE, and AVX2 instruction sets. Included are 3 parity generation routines (P, PQ, and PQR) and 7 reconstruction routines, for all RAIDZ level. On module load, a quick benchmark of supported routines will select the fastest for each operation and they will be used at runtime. Original implementation is still present and can be selected via module parameter. Patch contains: - specialized gen/rec routines for all RAIDZ levels, - new scalar raidz implementation (unrolled), - two x86_64 SIMD implementations (SSE and AVX2 instructions sets), - fastest routines selected on module load (benchmark). - cmd/raidz_test - verify and benchmark all implementations - added raidz_test to the ZFS Test Suite New zfs module parameters: - zfs_vdev_raidz_impl (str): selects the implementation to use. On module load, the parameter will only accept first 3 options, and the other implementations can be set once module is finished loading. Possible values for this option are: "fastest" - use the fastest math available "original" - use the original raidz code "scalar" - new scalar impl "sse" - new SSE impl if available "avx2" - new AVX2 impl if available See contents of `/sys/module/zfs/parameters/zfs_vdev_raidz_impl` to get the list of supported values. If an implementation is not supported on the system, it will not be shown. Currently selected option is enclosed in `[]`. Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4328
2016-04-25 08:04:31 +00:00
#include <sys/vdev_raidz.h>
2008-11-20 20:01:55 +00:00
#include <sys/metaslab.h>
#include <sys/uberblock_impl.h>
#include <sys/txg.h>
#include <sys/avl.h>
#include <sys/unique.h>
#include <sys/dsl_pool.h>
#include <sys/dsl_dir.h>
#include <sys/dsl_prop.h>
Add linux events This topic branch leverages the Solaris style FMA call points in ZFS to create a user space visible event notification system under Linux. This new system is called zevent and it unifies all previous Solaris style ereports and sysevent notifications. Under this Linux specific scheme when a sysevent or ereport event occurs an nvlist describing the event is created which looks almost exactly like a Solaris ereport. These events are queued up in the kernel when they occur and conditionally logged to the console. It is then up to a user space application to consume the events and do whatever it likes with them. To make this possible the existing /dev/zfs ABI has been extended with two new ioctls which behave as follows. * ZFS_IOC_EVENTS_NEXT Get the next pending event. The kernel will keep track of the last event consumed by the file descriptor and provide the next one if available. If no new events are available the ioctl() will block waiting for the next event. This ioctl may also be called in a non-blocking mode by setting zc.zc_guid = ZEVENT_NONBLOCK. In the non-blocking case if no events are available ENOENT will be returned. It is possible that ESHUTDOWN will be returned if the ioctl() is called while module unloading is in progress. And finally ENOMEM may occur if the provided nvlist buffer is not large enough to contain the entire event. * ZFS_IOC_EVENTS_CLEAR Clear are events queued by the kernel. The kernel will keep a fairly large number of recent events queued, use this ioctl to clear the in kernel list. This will effect all user space processes consuming events. The zpool command has been extended to use this events ABI with the 'events' subcommand. You may run 'zpool events -v' to output a verbose log of all recent events. This is very similar to the Solaris 'fmdump -ev' command with the key difference being it also includes what would be considered sysevents under Solaris. You may also run in follow mode with the '-f' option. To clear the in kernel event queue use the '-c' option. $ sudo cmd/zpool/zpool events -fv TIME CLASS May 13 2010 16:31:15.777711000 ereport.fs.zfs.config.sync class = "ereport.fs.zfs.config.sync" ena = 0x40982b7897700001 detector = (embedded nvlist) version = 0x0 scheme = "zfs" pool = 0xed976600de75dfa6 (end detector) time = 0x4bec8bc3 0x2e5aed98 pool = "zpios" pool_guid = 0xed976600de75dfa6 pool_context = 0x0 While the 'zpool events' command is handy for interactive debugging it is not expected to be the primary consumer of zevents. This ABI was primarily added to facilitate the addition of a user space monitoring daemon. This daemon would consume all events posted by the kernel and based on the type of event perform an action. For most events simply forwarding them on to syslog is likely enough. But this interface also cleanly allows for more sophisticated actions to be taken such as generating an email for a failed drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:42:43 +00:00
#include <sys/fm/util.h>
#include <sys/dsl_scan.h>
2008-11-20 20:01:55 +00:00
#include <sys/fs/zfs.h>
#include <sys/metaslab_impl.h>
#include <sys/arc.h>
#include <sys/ddt.h>
Add visibility in to arc_read 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 <surya1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2013-09-06 23:09:05 +00:00
#include <sys/kstat.h>
2008-11-20 20:01:55 +00:00
#include "zfs_prop.h"
Reduce loaded range tree memory usage This patch implements a new tree structure for ZFS, and uses it to store range trees more efficiently. The new structure is approximately a B-tree, though there are some small differences from the usual characterizations. The tree has core nodes and leaf nodes; each contain data elements, which the elements in the core nodes acting as separators between its children. The difference between core and leaf nodes is that the core nodes have an array of children, while leaf nodes don't. Every node in the tree may be only partially full; in most cases, they are all at least 50% full (in terms of element count) except for the root node, which can be less full. Underfull nodes will steal from their neighbors or merge to remain full enough, while overfull nodes will split in two. The data elements are contained in tree-controlled buffers; they are copied into these on insertion, and overwritten on deletion. This means that the elements are not independently allocated, which reduces overhead, but also means they can't be shared between trees (and also that pointers to them are only valid until a side-effectful tree operation occurs). The overhead varies based on how dense the tree is, but is usually on the order of about 50% of the element size; the per-node overheads are very small, and so don't make a significant difference. The trees can accept arbitrary records; they accept a size and a comparator to allow them to be used for a variety of purposes. The new trees replace the AVL trees used in the range trees today. Currently, the range_seg_t structure contains three 8 byte integers of payload and two 24 byte avl_tree_node_ts to handle its storage in both an offset-sorted tree and a size-sorted tree (total size: 64 bytes). In the new model, the range seg structures are usually two 4 byte integers, but a separate one needs to exist for the size-sorted and offset-sorted tree. Between the raw size, the 50% overhead, and the double storage, the new btrees are expected to use 8*1.5*2 = 24 bytes per record, or 33.3% as much memory as the AVL trees (this is for the purposes of storing metaslab range trees; for other purposes, like scrubs, they use ~50% as much memory). We reduced the size of the payload in the range segments by teaching range trees about starting offsets and shifts; since metaslabs have a fixed starting offset, and they all operate in terms of disk sectors, we can store the ranges using 4-byte integers as long as the size of the metaslab divided by the sector size is less than 2^32. For 512-byte sectors, this is a 2^41 (or 2TB) metaslab, which with the default settings corresponds to a 256PB disk. 4k sector disks can handle metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not anticipate disks of this size in the near future, there should be almost no cases where metaslabs need 64-byte integers to store their ranges. We do still have the capability to store 64-byte integer ranges to account for cases where we are storing per-vdev (or per-dnode) trees, which could reasonably go above the limits discussed. We also do not store fill information in the compact version of the node, since it is only used for sorted scrub. We also optimized the metaslab loading process in various other ways to offset some inefficiencies in the btree model. While individual operations (find, insert, remove_from) are faster for the btree than they are for the avl tree, remove usually requires a find operation, while in the AVL tree model the element itself suffices. Some clever changes actually caused an overall speedup in metaslab loading; we use approximately 40% less cpu to load metaslabs in our tests on Illumos. Another memory and performance optimization was achieved by changing what is stored in the size-sorted trees. When a disk is heavily fragmented, the df algorithm used by default in ZFS will almost always find a number of small regions in its initial cursor-based search; it will usually only fall back to the size-sorted tree to find larger regions. If we increase the size of the cursor-based search slightly, and don't store segments that are smaller than a tunable size floor in the size-sorted tree, we can further cut memory usage down to below 20% of what the AVL trees store. This also results in further reductions in CPU time spent loading metaslabs. The 16KiB size floor was chosen because it results in substantial memory usage reduction while not usually resulting in situations where we can't find an appropriate chunk with the cursor and are forced to use an oversized chunk from the size-sorted tree. In addition, even if we do have to use an oversized chunk from the size-sorted tree, the chunk would be too small to use for ZIL allocations, so it isn't as big of a loss as it might otherwise be. And often, more small allocations will follow the initial one, and the cursor search will now find the remainder of the chunk we didn't use all of and use it for subsequent allocations. Practical testing has shown little or no change in fragmentation as a result of this change. If the size-sorted tree becomes empty while the offset sorted one still has entries, it will load all the entries from the offset sorted tree and disregard the size floor until it is unloaded again. This operation occurs rarely with the default setting, only on incredibly thoroughly fragmented pools. There are some other small changes to zdb to teach it to handle btrees, but nothing major. Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed by: Sebastien Roy seb@delphix.com Reviewed-by: Igor Kozhukhov <igor@dilos.org> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Closes #9181
2019-10-09 17:36:03 +00:00
#include <sys/btree.h>
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
#include <sys/zfeature.h>
#include <sys/qat.h>
2008-11-20 20:01:55 +00:00
/*
* SPA locking
*
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
* There are three basic locks for managing spa_t structures:
2008-11-20 20:01:55 +00:00
*
* spa_namespace_lock (global mutex)
*
* This lock must be acquired to do any of the following:
*
* - Lookup a spa_t by name
* - Add or remove a spa_t from the namespace
* - Increase spa_refcount from non-zero
* - Check if spa_refcount is zero
* - Rename a spa_t
* - add/remove/attach/detach devices
* - Held for the duration of create/destroy/import/export
*
* It does not need to handle recursion. A create or destroy may
* reference objects (files or zvols) in other pools, but by
* definition they must have an existing reference, and will never need
* to lookup a spa_t by name.
*
* spa_refcount (per-spa zfs_refcount_t protected by mutex)
2008-11-20 20:01:55 +00:00
*
* This reference count keep track of any active users of the spa_t. The
* spa_t cannot be destroyed or freed while this is non-zero. Internally,
* the refcount is never really 'zero' - opening a pool implicitly keeps
* some references in the DMU. Internally we check against spa_minref, but
2008-11-20 20:01:55 +00:00
* present the image of a zero/non-zero value to consumers.
*
* spa_config_lock[] (per-spa array of rwlocks)
2008-11-20 20:01:55 +00:00
*
* This protects the spa_t from config changes, and must be held in
* the following circumstances:
*
* - RW_READER to perform I/O to the spa
* - RW_WRITER to change the vdev config
*
* The locking order is fairly straightforward:
*
* spa_namespace_lock -> spa_refcount
*
* The namespace lock must be acquired to increase the refcount from 0
* or to check if it is zero.
*
* spa_refcount -> spa_config_lock[]
2008-11-20 20:01:55 +00:00
*
* There must be at least one valid reference on the spa_t to acquire
* the config lock.
*
* spa_namespace_lock -> spa_config_lock[]
2008-11-20 20:01:55 +00:00
*
* The namespace lock must always be taken before the config lock.
*
*
* The spa_namespace_lock can be acquired directly and is globally visible.
2008-11-20 20:01:55 +00:00
*
* The namespace is manipulated using the following functions, all of which
* require the spa_namespace_lock to be held.
2008-11-20 20:01:55 +00:00
*
* spa_lookup() Lookup a spa_t by name.
*
* spa_add() Create a new spa_t in the namespace.
*
* spa_remove() Remove a spa_t from the namespace. This also
* frees up any memory associated with the spa_t.
*
* spa_next() Returns the next spa_t in the system, or the
* first if NULL is passed.
*
* spa_evict_all() Shutdown and remove all spa_t structures in
* the system.
*
* spa_guid_exists() Determine whether a pool/device guid exists.
*
* The spa_refcount is manipulated using the following functions:
*
* spa_open_ref() Adds a reference to the given spa_t. Must be
* called with spa_namespace_lock held if the
* refcount is currently zero.
*
* spa_close() Remove a reference from the spa_t. This will
* not free the spa_t or remove it from the
* namespace. No locking is required.
*
* spa_refcount_zero() Returns true if the refcount is currently
* zero. Must be called with spa_namespace_lock
* held.
*
* The spa_config_lock[] is an array of rwlocks, ordered as follows:
* SCL_CONFIG > SCL_STATE > SCL_ALLOC > SCL_ZIO > SCL_FREE > SCL_VDEV.
* spa_config_lock[] is manipulated with spa_config_{enter,exit,held}().
*
* To read the configuration, it suffices to hold one of these locks as reader.
* To modify the configuration, you must hold all locks as writer. To modify
* vdev state without altering the vdev tree's topology (e.g. online/offline),
* you must hold SCL_STATE and SCL_ZIO as writer.
*
* We use these distinct config locks to avoid recursive lock entry.
* For example, spa_sync() (which holds SCL_CONFIG as reader) induces
* block allocations (SCL_ALLOC), which may require reading space maps
* from disk (dmu_read() -> zio_read() -> SCL_ZIO).
*
* The spa config locks cannot be normal rwlocks because we need the
* ability to hand off ownership. For example, SCL_ZIO is acquired
* by the issuing thread and later released by an interrupt thread.
* They do, however, obey the usual write-wanted semantics to prevent
* writer (i.e. system administrator) starvation.
*
* The lock acquisition rules are as follows:
*
* SCL_CONFIG
* Protects changes to the vdev tree topology, such as vdev
* add/remove/attach/detach. Protects the dirty config list
* (spa_config_dirty_list) and the set of spares and l2arc devices.
*
* SCL_STATE
* Protects changes to pool state and vdev state, such as vdev
* online/offline/fault/degrade/clear. Protects the dirty state list
* (spa_state_dirty_list) and global pool state (spa_state).
*
* SCL_ALLOC
* Protects changes to metaslab groups and classes.
* Held as reader by metaslab_alloc() and metaslab_claim().
*
* SCL_ZIO
* Held by bp-level zios (those which have no io_vd upon entry)
* to prevent changes to the vdev tree. The bp-level zio implicitly
* protects all of its vdev child zios, which do not hold SCL_ZIO.
*
* SCL_FREE
* Protects changes to metaslab groups and classes.
* Held as reader by metaslab_free(). SCL_FREE is distinct from
* SCL_ALLOC, and lower than SCL_ZIO, so that we can safely free
* blocks in zio_done() while another i/o that holds either
* SCL_ALLOC or SCL_ZIO is waiting for this i/o to complete.
*
* SCL_VDEV
* Held as reader to prevent changes to the vdev tree during trivial
* inquiries such as bp_get_dsize(). SCL_VDEV is distinct from the
* other locks, and lower than all of them, to ensure that it's safe
* to acquire regardless of caller context.
*
* In addition, the following rules apply:
*
* (a) spa_props_lock protects pool properties, spa_config and spa_config_list.
* The lock ordering is SCL_CONFIG > spa_props_lock.
*
* (b) I/O operations on leaf vdevs. For any zio operation that takes
* an explicit vdev_t argument -- such as zio_ioctl(), zio_read_phys(),
* or zio_write_phys() -- the caller must ensure that the config cannot
* cannot change in the interim, and that the vdev cannot be reopened.
* SCL_STATE as reader suffices for both.
2008-11-20 20:01:55 +00:00
*
* The vdev configuration is protected by spa_vdev_enter() / spa_vdev_exit().
*
* spa_vdev_enter() Acquire the namespace lock and the config lock
* for writing.
*
* spa_vdev_exit() Release the config lock, wait for all I/O
* to complete, sync the updated configs to the
* cache, and release the namespace lock.
*
* vdev state is protected by spa_vdev_state_enter() / spa_vdev_state_exit().
* Like spa_vdev_enter/exit, these are convenience wrappers -- the actual
* locking is, always, based on spa_namespace_lock and spa_config_lock[].
2008-11-20 20:01:55 +00:00
*/
static avl_tree_t spa_namespace_avl;
kmutex_t spa_namespace_lock;
static kcondvar_t spa_namespace_cv;
int spa_max_replication_override = SPA_DVAS_PER_BP;
static kmutex_t spa_spare_lock;
static avl_tree_t spa_spare_avl;
static kmutex_t spa_l2cache_lock;
static avl_tree_t spa_l2cache_avl;
kmem_cache_t *spa_buffer_pool;
spa_mode_t spa_mode_global = SPA_MODE_UNINIT;
2008-11-20 20:01:55 +00:00
Swap DTRACE_PROBE* with Linux tracepoints This patch leverages Linux tracepoints from within the ZFS on Linux code base. It also refactors the debug code to bring it back in sync with Illumos. The information exported via tracepoints can be used for a variety of reasons (e.g. debugging, tuning, general exploration/understanding, etc). It is advantageous to use Linux tracepoints as the mechanism to export this kind of information (as opposed to something else) for a number of reasons: * A number of external tools can make use of our tracepoints "automatically" (e.g. perf, systemtap) * Tracepoints are designed to be extremely cheap when disabled * It's one of the "accepted" ways to export this kind of information; many other kernel subsystems use tracepoints too. Unfortunately, though, there are a few caveats as well: * Linux tracepoints appear to only be available to GPL licensed modules due to the way certain kernel functions are exported. Thus, to actually make use of the tracepoints introduced by this patch, one might have to patch and re-compile the kernel; exporting the necessary functions to non-GPL modules. * Prior to upstream kernel version v3.14-rc6-30-g66cc69e, Linux tracepoints are not available for unsigned kernel modules (tracepoints will get disabled due to the module's 'F' taint). Thus, one either has to sign the zfs kernel module prior to loading it, or use a kernel versioned v3.14-rc6-30-g66cc69e or newer. Assuming the above two requirements are satisfied, lets look at an example of how this patch can be used and what information it exposes (all commands run as 'root'): # list all zfs tracepoints available $ ls /sys/kernel/debug/tracing/events/zfs enable filter zfs_arc__delete zfs_arc__evict zfs_arc__hit zfs_arc__miss zfs_l2arc__evict zfs_l2arc__hit zfs_l2arc__iodone zfs_l2arc__miss zfs_l2arc__read zfs_l2arc__write zfs_new_state__mfu zfs_new_state__mru # enable all zfs tracepoints, clear the tracepoint ring buffer $ echo 1 > /sys/kernel/debug/tracing/events/zfs/enable $ echo 0 > /sys/kernel/debug/tracing/trace # import zpool called 'tank', inspect tracepoint data (each line was # truncated, they're too long for a commit message otherwise) $ zpool import tank $ cat /sys/kernel/debug/tracing/trace | head -n35 # tracer: nop # # entries-in-buffer/entries-written: 1219/1219 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr... z_rd_int/0-30156 [003] .... 91344.200611: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201173: zfs_arc__miss: hdr... z_rd_int/1-30157 [003] .... 91344.201756: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201795: zfs_arc__miss: hdr... z_rd_int/2-30158 [003] .... 91344.202099: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202126: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202130: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202134: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202146: zfs_arc__miss: hdr... z_rd_int/3-30159 [003] .... 91344.202457: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202484: zfs_arc__miss: hdr... z_rd_int/4-30160 [003] .... 91344.202866: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202891: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203034: zfs_arc__miss: hdr... z_rd_iss/1-30149 [001] .... 91344.203749: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.203789: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203878: zfs_arc__miss: hdr... z_rd_iss/3-30151 [001] .... 91344.204315: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.204332: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204337: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204352: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204356: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204360: zfs_arc__hit: hdr ... To highlight the kind of detailed information that is being exported using this infrastructure, I've taken the first tracepoint line from the output above and reformatted it such that it fits in 80 columns: lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr { dva 0x1:0x40082 birth 15491 cksum0 0x163edbff3a flags 0x640 datacnt 1 type 1 size 2048 spa 3133524293419867460 state_type 0 access 0 mru_hits 0 mru_ghost_hits 0 mfu_hits 0 mfu_ghost_hits 0 l2_hits 0 refcount 1 } bp { dva0 0x1:0x40082 dva1 0x1:0x3000e5 dva2 0x1:0x5a006e cksum 0x163edbff3a:0x75af30b3dd6:0x1499263ff5f2b:0x288bd118815e00 lsize 2048 } zb { objset 0 object 0 level -1 blkid 0 } For the specific tracepoint shown here, 'zfs_arc__miss', data is exported detailing the arc_buf_hdr_t (hdr), blkptr_t (bp), and zbookmark_t (zb) that caused the ARC miss (down to the exact DVA!). This kind of precise and detailed information can be extremely valuable when trying to answer certain kinds of questions. For anybody unfamiliar but looking to build on this, I found the XFS source code along with the following three web links to be extremely helpful: * http://lwn.net/Articles/379903/ * http://lwn.net/Articles/381064/ * http://lwn.net/Articles/383362/ I should also node the more "boring" aspects of this patch: * The ZFS_LINUX_COMPILE_IFELSE autoconf macro was modified to support a sixth paramter. This parameter is used to populate the contents of the new conftest.h file. If no sixth parameter is provided, conftest.h will be empty. * The ZFS_LINUX_TRY_COMPILE_HEADER autoconf macro was introduced. This macro is nearly identical to the ZFS_LINUX_TRY_COMPILE macro, except it has support for a fifth option that is then passed as the sixth parameter to ZFS_LINUX_COMPILE_IFELSE. These autoconf changes were needed to test the availability of the Linux tracepoint macros. Due to the odd nature of the Linux tracepoint macro API, a separate ".h" must be created (the path and filename is used internally by the kernel's define_trace.h file). * The HAVE_DECLARE_EVENT_CLASS autoconf macro was introduced. This is to determine if we can safely enable the Linux tracepoint functionality. We need to selectively disable the tracepoint code due to the kernel exporting certain functions as GPL only. Without this check, the build process will fail at link time. In addition, the SET_ERROR macro was modified into a tracepoint as well. To do this, the 'sdt.h' file was moved into the 'include/sys' directory and now contains a userspace portion and a kernel space portion. The dprintf and zfs_dbgmsg* interfaces are now implemented as tracepoint as well. Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-06-13 17:54:48 +00:00
#ifdef ZFS_DEBUG
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
/*
* Everything except dprintf, set_error, spa, and indirect_remap is on
* by default in debug builds.
*/
int zfs_flags = ~(ZFS_DEBUG_DPRINTF | ZFS_DEBUG_SET_ERROR |
ZFS_DEBUG_INDIRECT_REMAP);
Swap DTRACE_PROBE* with Linux tracepoints This patch leverages Linux tracepoints from within the ZFS on Linux code base. It also refactors the debug code to bring it back in sync with Illumos. The information exported via tracepoints can be used for a variety of reasons (e.g. debugging, tuning, general exploration/understanding, etc). It is advantageous to use Linux tracepoints as the mechanism to export this kind of information (as opposed to something else) for a number of reasons: * A number of external tools can make use of our tracepoints "automatically" (e.g. perf, systemtap) * Tracepoints are designed to be extremely cheap when disabled * It's one of the "accepted" ways to export this kind of information; many other kernel subsystems use tracepoints too. Unfortunately, though, there are a few caveats as well: * Linux tracepoints appear to only be available to GPL licensed modules due to the way certain kernel functions are exported. Thus, to actually make use of the tracepoints introduced by this patch, one might have to patch and re-compile the kernel; exporting the necessary functions to non-GPL modules. * Prior to upstream kernel version v3.14-rc6-30-g66cc69e, Linux tracepoints are not available for unsigned kernel modules (tracepoints will get disabled due to the module's 'F' taint). Thus, one either has to sign the zfs kernel module prior to loading it, or use a kernel versioned v3.14-rc6-30-g66cc69e or newer. Assuming the above two requirements are satisfied, lets look at an example of how this patch can be used and what information it exposes (all commands run as 'root'): # list all zfs tracepoints available $ ls /sys/kernel/debug/tracing/events/zfs enable filter zfs_arc__delete zfs_arc__evict zfs_arc__hit zfs_arc__miss zfs_l2arc__evict zfs_l2arc__hit zfs_l2arc__iodone zfs_l2arc__miss zfs_l2arc__read zfs_l2arc__write zfs_new_state__mfu zfs_new_state__mru # enable all zfs tracepoints, clear the tracepoint ring buffer $ echo 1 > /sys/kernel/debug/tracing/events/zfs/enable $ echo 0 > /sys/kernel/debug/tracing/trace # import zpool called 'tank', inspect tracepoint data (each line was # truncated, they're too long for a commit message otherwise) $ zpool import tank $ cat /sys/kernel/debug/tracing/trace | head -n35 # tracer: nop # # entries-in-buffer/entries-written: 1219/1219 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr... z_rd_int/0-30156 [003] .... 91344.200611: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201173: zfs_arc__miss: hdr... z_rd_int/1-30157 [003] .... 91344.201756: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201795: zfs_arc__miss: hdr... z_rd_int/2-30158 [003] .... 91344.202099: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202126: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202130: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202134: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202146: zfs_arc__miss: hdr... z_rd_int/3-30159 [003] .... 91344.202457: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202484: zfs_arc__miss: hdr... z_rd_int/4-30160 [003] .... 91344.202866: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202891: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203034: zfs_arc__miss: hdr... z_rd_iss/1-30149 [001] .... 91344.203749: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.203789: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203878: zfs_arc__miss: hdr... z_rd_iss/3-30151 [001] .... 91344.204315: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.204332: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204337: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204352: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204356: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204360: zfs_arc__hit: hdr ... To highlight the kind of detailed information that is being exported using this infrastructure, I've taken the first tracepoint line from the output above and reformatted it such that it fits in 80 columns: lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr { dva 0x1:0x40082 birth 15491 cksum0 0x163edbff3a flags 0x640 datacnt 1 type 1 size 2048 spa 3133524293419867460 state_type 0 access 0 mru_hits 0 mru_ghost_hits 0 mfu_hits 0 mfu_ghost_hits 0 l2_hits 0 refcount 1 } bp { dva0 0x1:0x40082 dva1 0x1:0x3000e5 dva2 0x1:0x5a006e cksum 0x163edbff3a:0x75af30b3dd6:0x1499263ff5f2b:0x288bd118815e00 lsize 2048 } zb { objset 0 object 0 level -1 blkid 0 } For the specific tracepoint shown here, 'zfs_arc__miss', data is exported detailing the arc_buf_hdr_t (hdr), blkptr_t (bp), and zbookmark_t (zb) that caused the ARC miss (down to the exact DVA!). This kind of precise and detailed information can be extremely valuable when trying to answer certain kinds of questions. For anybody unfamiliar but looking to build on this, I found the XFS source code along with the following three web links to be extremely helpful: * http://lwn.net/Articles/379903/ * http://lwn.net/Articles/381064/ * http://lwn.net/Articles/383362/ I should also node the more "boring" aspects of this patch: * The ZFS_LINUX_COMPILE_IFELSE autoconf macro was modified to support a sixth paramter. This parameter is used to populate the contents of the new conftest.h file. If no sixth parameter is provided, conftest.h will be empty. * The ZFS_LINUX_TRY_COMPILE_HEADER autoconf macro was introduced. This macro is nearly identical to the ZFS_LINUX_TRY_COMPILE macro, except it has support for a fifth option that is then passed as the sixth parameter to ZFS_LINUX_COMPILE_IFELSE. These autoconf changes were needed to test the availability of the Linux tracepoint macros. Due to the odd nature of the Linux tracepoint macro API, a separate ".h" must be created (the path and filename is used internally by the kernel's define_trace.h file). * The HAVE_DECLARE_EVENT_CLASS autoconf macro was introduced. This is to determine if we can safely enable the Linux tracepoint functionality. We need to selectively disable the tracepoint code due to the kernel exporting certain functions as GPL only. Without this check, the build process will fail at link time. In addition, the SET_ERROR macro was modified into a tracepoint as well. To do this, the 'sdt.h' file was moved into the 'include/sys' directory and now contains a userspace portion and a kernel space portion. The dprintf and zfs_dbgmsg* interfaces are now implemented as tracepoint as well. Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-06-13 17:54:48 +00:00
#else
int zfs_flags = 0;
#endif
/*
* zfs_recover can be set to nonzero to attempt to recover from
* otherwise-fatal errors, typically caused by on-disk corruption. When
* set, calls to zfs_panic_recover() will turn into warning messages.
* This should only be used as a last resort, as it typically results
* in leaked space, or worse.
*/
int zfs_recover = B_FALSE;
/*
* If destroy encounters an EIO while reading metadata (e.g. indirect
* blocks), space referenced by the missing metadata can not be freed.
* Normally this causes the background destroy to become "stalled", as
* it is unable to make forward progress. While in this stalled state,
* all remaining space to free from the error-encountering filesystem is
* "temporarily leaked". Set this flag to cause it to ignore the EIO,
* permanently leak the space from indirect blocks that can not be read,
* and continue to free everything else that it can.
*
* The default, "stalling" behavior is useful if the storage partially
* fails (i.e. some but not all i/os fail), and then later recovers. In
* this case, we will be able to continue pool operations while it is
* partially failed, and when it recovers, we can continue to free the
* space, with no leaks. However, note that this case is actually
* fairly rare.
*
* Typically pools either (a) fail completely (but perhaps temporarily,
* e.g. a top-level vdev going offline), or (b) have localized,
* permanent errors (e.g. disk returns the wrong data due to bit flip or
* firmware bug). In case (a), this setting does not matter because the
* pool will be suspended and the sync thread will not be able to make
* forward progress regardless. In case (b), because the error is
* permanent, the best we can do is leak the minimum amount of space,
* which is what setting this flag will do. Therefore, it is reasonable
* for this flag to normally be set, but we chose the more conservative
* approach of not setting it, so that there is no possibility of
* leaking space in the "partial temporary" failure case.
*/
int zfs_free_leak_on_eio = B_FALSE;
/*
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@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
* Expiration time in milliseconds. This value has two meanings. First it is
* used to determine when the spa_deadman() logic should fire. By default the
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
* spa_deadman() will fire if spa_sync() has not completed in 600 seconds.
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@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
* Secondly, the value determines if an I/O is considered "hung". Any I/O that
* has not completed in zfs_deadman_synctime_ms is considered "hung" resulting
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
* in one of three behaviors controlled by zfs_deadman_failmode.
*/
unsigned long zfs_deadman_synctime_ms = 600000UL;
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
/*
* This value controls the maximum amount of time zio_wait() will block for an
* outstanding IO. By default this is 300 seconds at which point the "hung"
* behavior will be applied as described for zfs_deadman_synctime_ms.
*/
unsigned long zfs_deadman_ziotime_ms = 300000UL;
/*
* Check time in milliseconds. This defines the frequency at which we check
* for hung I/O.
*/
unsigned long zfs_deadman_checktime_ms = 60000UL;
/*
* By default the deadman is enabled.
*/
int zfs_deadman_enabled = 1;
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
/*
* Controls the behavior of the deadman when it detects a "hung" I/O.
* Valid values are zfs_deadman_failmode=<wait|continue|panic>.
*
* wait - Wait for the "hung" I/O (default)
* continue - Attempt to recover from a "hung" I/O
* panic - Panic the system
*/
char *zfs_deadman_failmode = "wait";
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@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
/*
* The worst case is single-sector max-parity RAID-Z blocks, in which
* case the space requirement is exactly (VDEV_RAIDZ_MAXPARITY + 1)
* times the size; so just assume that. Add to this the fact that
* we can have up to 3 DVAs per bp, and one more factor of 2 because
* the block may be dittoed with up to 3 DVAs by ddt_sync(). All together,
* the worst case is:
* (VDEV_RAIDZ_MAXPARITY + 1) * SPA_DVAS_PER_BP * 2 == 24
*/
int spa_asize_inflation = 24;
/*
* Normally, we don't allow the last 3.2% (1/(2^spa_slop_shift)) of space in
* the pool to be consumed. This ensures that we don't run the pool
* completely out of space, due to unaccounted changes (e.g. to the MOS).
* It also limits the worst-case time to allocate space. If we have
* less than this amount of free space, most ZPL operations (e.g. write,
* create) will return ENOSPC.
*
* Certain operations (e.g. file removal, most administrative actions) can
* use half the slop space. They will only return ENOSPC if less than half
* the slop space is free. Typically, once the pool has less than the slop
* space free, the user will use these operations to free up space in the pool.
* These are the operations that call dsl_pool_adjustedsize() with the netfree
* argument set to TRUE.
*
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
* Operations that are almost guaranteed to free up space in the absence of
* a pool checkpoint can use up to three quarters of the slop space
* (e.g zfs destroy).
*
* A very restricted set of operations are always permitted, regardless of
* the amount of free space. These are the operations that call
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
* dsl_sync_task(ZFS_SPACE_CHECK_NONE). If these operations result in a net
* increase in the amount of space used, it is possible to run the pool
* completely out of space, causing it to be permanently read-only.
*
* Note that on very small pools, the slop space will be larger than
* 3.2%, in an effort to have it be at least spa_min_slop (128MB),
* but we never allow it to be more than half the pool size.
*
* See also the comments in zfs_space_check_t.
*/
int spa_slop_shift = 5;
uint64_t spa_min_slop = 128 * 1024 * 1024;
OpenZFS 9112 - Improve allocation performance on high-end systems Overview ======== We parallelize the allocation process by creating the concept of "allocators". There are a certain number of allocators per metaslab group, defined by the value of a tunable at pool open time. Each allocator for a given metaslab group has up to 2 active metaslabs; one "primary", and one "secondary". The primary and secondary weight mean the same thing they did in in the pre-allocator world; primary metaslabs are used for most allocations, secondary metaslabs are used for ditto blocks being allocated in the same metaslab group. There is also the CLAIM weight, which has been separated out from the other weights, but that is less important to understanding the patch. The active metaslabs for each allocator are moved from their normal place in the metaslab tree for the group to the back of the tree. This way, they will not be selected for use by other allocators searching for new metaslabs unless all the passive metaslabs are unsuitable for allocations. If that does happen, the allocators will "steal" from each other to ensure that IOs don't fail until there is truly no space left to perform allocations. In addition, the alloc queue for each metaslab group has been broken into a separate queue for each allocator. We don't want to dramatically increase the number of inflight IOs on low-end systems, because it can significantly increase txg times. On the other hand, we want to ensure that there are enough IOs for each allocator to allow for good coalescing before sending the IOs to the disk. As a result, we take a compromise path; each allocator's alloc queue max depth starts at a certain value for every txg. Every time an IO completes, we increase the max depth. This should hopefully provide a good balance between the two failure modes, while not dramatically increasing complexity. We also parallelize the spa_alloc_tree and spa_alloc_lock, which cause very similar contention when selecting IOs to allocate. This parallelization uses the same allocator scheme as metaslab selection. Performance Results =================== Performance improvements from this change can vary significantly based on the number of CPUs in the system, whether or not the system has a NUMA architecture, the speed of the drives, the values for the various tunables, and the workload being performed. For an fio async sequential write workload on a 24 core NUMA system with 256 GB of RAM and 8 128 GB SSDs, there is a roughly 25% performance improvement. Future Work =========== Analysis of the performance of the system with this patch applied shows that a significant new bottleneck is the vdev disk queues, which also need to be parallelized. Prototyping of this change has occurred, and there was a performance improvement, but more work needs to be done before its stability has been verified and it is ready to be upstreamed. Authored by: Paul Dagnelie <pcd@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Alexander Motin <mav@FreeBSD.org> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Gordon Ross <gwr@nexenta.com> Ported-by: Paul Dagnelie <pcd@delphix.com> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Porting Notes: * Fix reservation test failures by increasing tolerance. OpenZFS-issue: https://illumos.org/issues/9112 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/3f3cc3c3 Closes #7682
2018-02-12 20:56:06 +00:00
int spa_allocators = 4;
/*PRINTFLIKE2*/
void
spa_load_failed(spa_t *spa, const char *fmt, ...)
{
va_list adx;
char buf[256];
va_start(adx, fmt);
(void) vsnprintf(buf, sizeof (buf), fmt, adx);
va_end(adx);
OpenZFS 9075 - Improve ZFS pool import/load process and corrupted pool recovery Some work has been done lately to improve the debugability of the ZFS pool load (and import) process. This includes: 7638 Refactor spa_load_impl into several functions 8961 SPA load/import should tell us why it failed 7277 zdb should be able to print zfs_dbgmsg's To iterate on top of that, there's a few changes that were made to make the import process more resilient and crash free. One of the first tasks during the pool load process is to parse a config provided from userland that describes what devices the pool is composed of. A vdev tree is generated from that config, and then all the vdevs are opened. The Meta Object Set (MOS) of the pool is accessed, and several metadata objects that are necessary to load the pool are read. The exact configuration of the pool is also stored inside the MOS. Since the configuration provided from userland is external and might not accurately describe the vdev tree of the pool at the txg that is being loaded, it cannot be relied upon to safely operate the pool. For that reason, the configuration in the MOS is read early on. In the past, the two configurations were compared together and if there was a mismatch then the load process was aborted and an error was returned. The latter was a good way to ensure a pool does not get corrupted, however it made the pool load process needlessly fragile in cases where the vdev configuration changed or the userland configuration was outdated. Since the MOS is stored in 3 copies, the configuration provided by userland doesn't have to be perfect in order to read its contents. Hence, a new approach has been adopted: The pool is first opened with the untrusted userland configuration just so that the real configuration can be read from the MOS. The trusted MOS configuration is then used to generate a new vdev tree and the pool is re-opened. When the pool is opened with an untrusted configuration, writes are disabled to avoid accidentally damaging it. During reads, some sanity checks are performed on block pointers to see if each DVA points to a known vdev; when the configuration is untrusted, instead of panicking the system if those checks fail we simply avoid issuing reads to the invalid DVAs. This new two-step pool load process now allows rewinding pools accross vdev tree changes such as device replacement, addition, etc. Loading a pool from an external config file in a clustering environment also becomes much safer now since the pool will import even if the config is outdated and didn't, for instance, register a recent device addition. With this code in place, it became relatively easy to implement a long-sought-after feature: the ability to import a pool with missing top level (i.e. non-redundant) devices. Note that since this almost guarantees some loss of data, this feature is for now restricted to a read-only import. Porting notes (ZTS): * Fix 'make dist' target in zpool_import * The maximum path length allowed by tar is 99 characters. Several of the new test cases exceeded this limit resulting in them not being included in the tarball. Shorten the names slightly. * Set/get tunables using accessor functions. * Get last synced txg via the "zfs_txg_history" mechanism. * Clear zinject handlers in cleanup for import_cache_device_replaced and import_rewind_device_replaced in order that the zpool can be exported if there is an error. * Increase FILESIZE to 8G in zfs-test.sh to allow for a larger ext4 file system to be created on ZFS_DISK2. Also, there's no need to partition ZFS_DISK2 at all. The partitioning had already been disabled for multipath devices. Among other things, the partitioning steals some space from the ext4 file system, makes it difficult to accurately calculate the paramters to parted and can make some of the tests fail. * Increase FS_SIZE and FILE_SIZE in the zpool_import test configuration now that FILESIZE is larger. * Write more data in order that device evacuation take lonnger in a couple tests. * Use mkdir -p to avoid errors when the directory already exists. * Remove use of sudo in import_rewind_config_changed. Authored by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Andrew Stormont <andyjstormont@gmail.com> Approved by: Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9075 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/619c0123 Closes #7459
2016-07-22 14:39:36 +00:00
zfs_dbgmsg("spa_load(%s, config %s): FAILED: %s", spa->spa_name,
spa->spa_trust_config ? "trusted" : "untrusted", buf);
}
/*PRINTFLIKE2*/
void
spa_load_note(spa_t *spa, const char *fmt, ...)
{
va_list adx;
char buf[256];
va_start(adx, fmt);
(void) vsnprintf(buf, sizeof (buf), fmt, adx);
va_end(adx);
OpenZFS 9075 - Improve ZFS pool import/load process and corrupted pool recovery Some work has been done lately to improve the debugability of the ZFS pool load (and import) process. This includes: 7638 Refactor spa_load_impl into several functions 8961 SPA load/import should tell us why it failed 7277 zdb should be able to print zfs_dbgmsg's To iterate on top of that, there's a few changes that were made to make the import process more resilient and crash free. One of the first tasks during the pool load process is to parse a config provided from userland that describes what devices the pool is composed of. A vdev tree is generated from that config, and then all the vdevs are opened. The Meta Object Set (MOS) of the pool is accessed, and several metadata objects that are necessary to load the pool are read. The exact configuration of the pool is also stored inside the MOS. Since the configuration provided from userland is external and might not accurately describe the vdev tree of the pool at the txg that is being loaded, it cannot be relied upon to safely operate the pool. For that reason, the configuration in the MOS is read early on. In the past, the two configurations were compared together and if there was a mismatch then the load process was aborted and an error was returned. The latter was a good way to ensure a pool does not get corrupted, however it made the pool load process needlessly fragile in cases where the vdev configuration changed or the userland configuration was outdated. Since the MOS is stored in 3 copies, the configuration provided by userland doesn't have to be perfect in order to read its contents. Hence, a new approach has been adopted: The pool is first opened with the untrusted userland configuration just so that the real configuration can be read from the MOS. The trusted MOS configuration is then used to generate a new vdev tree and the pool is re-opened. When the pool is opened with an untrusted configuration, writes are disabled to avoid accidentally damaging it. During reads, some sanity checks are performed on block pointers to see if each DVA points to a known vdev; when the configuration is untrusted, instead of panicking the system if those checks fail we simply avoid issuing reads to the invalid DVAs. This new two-step pool load process now allows rewinding pools accross vdev tree changes such as device replacement, addition, etc. Loading a pool from an external config file in a clustering environment also becomes much safer now since the pool will import even if the config is outdated and didn't, for instance, register a recent device addition. With this code in place, it became relatively easy to implement a long-sought-after feature: the ability to import a pool with missing top level (i.e. non-redundant) devices. Note that since this almost guarantees some loss of data, this feature is for now restricted to a read-only import. Porting notes (ZTS): * Fix 'make dist' target in zpool_import * The maximum path length allowed by tar is 99 characters. Several of the new test cases exceeded this limit resulting in them not being included in the tarball. Shorten the names slightly. * Set/get tunables using accessor functions. * Get last synced txg via the "zfs_txg_history" mechanism. * Clear zinject handlers in cleanup for import_cache_device_replaced and import_rewind_device_replaced in order that the zpool can be exported if there is an error. * Increase FILESIZE to 8G in zfs-test.sh to allow for a larger ext4 file system to be created on ZFS_DISK2. Also, there's no need to partition ZFS_DISK2 at all. The partitioning had already been disabled for multipath devices. Among other things, the partitioning steals some space from the ext4 file system, makes it difficult to accurately calculate the paramters to parted and can make some of the tests fail. * Increase FS_SIZE and FILE_SIZE in the zpool_import test configuration now that FILESIZE is larger. * Write more data in order that device evacuation take lonnger in a couple tests. * Use mkdir -p to avoid errors when the directory already exists. * Remove use of sudo in import_rewind_config_changed. Authored by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Andrew Stormont <andyjstormont@gmail.com> Approved by: Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9075 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/619c0123 Closes #7459
2016-07-22 14:39:36 +00:00
zfs_dbgmsg("spa_load(%s, config %s): %s", spa->spa_name,
spa->spa_trust_config ? "trusted" : "untrusted", buf);
}
/*
* By default dedup and user data indirects land in the special class
*/
int zfs_ddt_data_is_special = B_TRUE;
int zfs_user_indirect_is_special = B_TRUE;
/*
* The percentage of special class final space reserved for metadata only.
* Once we allocate 100 - zfs_special_class_metadata_reserve_pct we only
* let metadata into the class.
*/
int zfs_special_class_metadata_reserve_pct = 25;
2008-11-20 20:01:55 +00:00
/*
* ==========================================================================
* SPA config locking
* ==========================================================================
*/
static void
spa_config_lock_init(spa_t *spa)
{
for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
mutex_init(&scl->scl_lock, NULL, MUTEX_DEFAULT, NULL);
cv_init(&scl->scl_cv, NULL, CV_DEFAULT, NULL);
zfs_refcount_create_untracked(&scl->scl_count);
scl->scl_writer = NULL;
scl->scl_write_wanted = 0;
}
2008-11-20 20:01:55 +00:00
}
static void
spa_config_lock_destroy(spa_t *spa)
{
for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
mutex_destroy(&scl->scl_lock);
cv_destroy(&scl->scl_cv);
zfs_refcount_destroy(&scl->scl_count);
ASSERT(scl->scl_writer == NULL);
ASSERT(scl->scl_write_wanted == 0);
}
}
int
spa_config_tryenter(spa_t *spa, int locks, void *tag, krw_t rw)
2008-11-20 20:01:55 +00:00
{
for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (!(locks & (1 << i)))
continue;
mutex_enter(&scl->scl_lock);
if (rw == RW_READER) {
if (scl->scl_writer || scl->scl_write_wanted) {
mutex_exit(&scl->scl_lock);
spa_config_exit(spa, locks & ((1 << i) - 1),
tag);
return (0);
}
} else {
ASSERT(scl->scl_writer != curthread);
if (!zfs_refcount_is_zero(&scl->scl_count)) {
mutex_exit(&scl->scl_lock);
spa_config_exit(spa, locks & ((1 << i) - 1),
tag);
return (0);
}
scl->scl_writer = curthread;
}
(void) zfs_refcount_add(&scl->scl_count, tag);
mutex_exit(&scl->scl_lock);
}
return (1);
2008-11-20 20:01:55 +00:00
}
void
spa_config_enter(spa_t *spa, int locks, const void *tag, krw_t rw)
2008-11-20 20:01:55 +00:00
{
2009-08-18 18:43:27 +00:00
int wlocks_held = 0;
ASSERT3U(SCL_LOCKS, <, sizeof (wlocks_held) * NBBY);
for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
2009-08-18 18:43:27 +00:00
if (scl->scl_writer == curthread)
wlocks_held |= (1 << i);
if (!(locks & (1 << i)))
continue;
mutex_enter(&scl->scl_lock);
if (rw == RW_READER) {
while (scl->scl_writer || scl->scl_write_wanted) {
cv_wait(&scl->scl_cv, &scl->scl_lock);
}
} else {
ASSERT(scl->scl_writer != curthread);
while (!zfs_refcount_is_zero(&scl->scl_count)) {
scl->scl_write_wanted++;
cv_wait(&scl->scl_cv, &scl->scl_lock);
scl->scl_write_wanted--;
}
scl->scl_writer = curthread;
}
(void) zfs_refcount_add(&scl->scl_count, tag);
mutex_exit(&scl->scl_lock);
2008-11-20 20:01:55 +00:00
}
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
ASSERT3U(wlocks_held, <=, locks);
2008-11-20 20:01:55 +00:00
}
void
spa_config_exit(spa_t *spa, int locks, const void *tag)
2008-11-20 20:01:55 +00:00
{
for (int i = SCL_LOCKS - 1; i >= 0; i--) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (!(locks & (1 << i)))
continue;
mutex_enter(&scl->scl_lock);
ASSERT(!zfs_refcount_is_zero(&scl->scl_count));
if (zfs_refcount_remove(&scl->scl_count, tag) == 0) {
ASSERT(scl->scl_writer == NULL ||
scl->scl_writer == curthread);
scl->scl_writer = NULL; /* OK in either case */
cv_broadcast(&scl->scl_cv);
}
mutex_exit(&scl->scl_lock);
2008-11-20 20:01:55 +00:00
}
}
int
spa_config_held(spa_t *spa, int locks, krw_t rw)
2008-11-20 20:01:55 +00:00
{
int locks_held = 0;
2008-11-20 20:01:55 +00:00
for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (!(locks & (1 << i)))
continue;
if ((rw == RW_READER &&
!zfs_refcount_is_zero(&scl->scl_count)) ||
(rw == RW_WRITER && scl->scl_writer == curthread))
locks_held |= 1 << i;
}
return (locks_held);
2008-11-20 20:01:55 +00:00
}
/*
* ==========================================================================
* SPA namespace functions
* ==========================================================================
*/
/*
* Lookup the named spa_t in the AVL tree. The spa_namespace_lock must be held.
* Returns NULL if no matching spa_t is found.
*/
spa_t *
spa_lookup(const char *name)
{
static spa_t search; /* spa_t is large; don't allocate on stack */
spa_t *spa;
2008-11-20 20:01:55 +00:00
avl_index_t where;
char *cp;
ASSERT(MUTEX_HELD(&spa_namespace_lock));
(void) strlcpy(search.spa_name, name, sizeof (search.spa_name));
2008-11-20 20:01:55 +00:00
/*
* If it's a full dataset name, figure out the pool name and
* just use that.
*/
cp = strpbrk(search.spa_name, "/@#");
if (cp != NULL)
2008-11-20 20:01:55 +00:00
*cp = '\0';
spa = avl_find(&spa_namespace_avl, &search, &where);
return (spa);
}
/*
* Fires when spa_sync has not completed within zfs_deadman_synctime_ms.
* If the zfs_deadman_enabled flag is set then it inspects all vdev queues
* looking for potentially hung I/Os.
*/
void
spa_deadman(void *arg)
{
spa_t *spa = arg;
/* Disable the deadman if the pool is suspended. */
if (spa_suspended(spa))
return;
zfs_dbgmsg("slow spa_sync: started %llu seconds ago, calls %llu",
(gethrtime() - spa->spa_sync_starttime) / NANOSEC,
++spa->spa_deadman_calls);
if (zfs_deadman_enabled)
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
vdev_deadman(spa->spa_root_vdev, FTAG);
spa->spa_deadman_tqid = taskq_dispatch_delay(system_delay_taskq,
spa_deadman, spa, TQ_SLEEP, ddi_get_lbolt() +
MSEC_TO_TICK(zfs_deadman_checktime_ms));
}
static int
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
spa_log_sm_sort_by_txg(const void *va, const void *vb)
{
const spa_log_sm_t *a = va;
const spa_log_sm_t *b = vb;
Reduce loaded range tree memory usage This patch implements a new tree structure for ZFS, and uses it to store range trees more efficiently. The new structure is approximately a B-tree, though there are some small differences from the usual characterizations. The tree has core nodes and leaf nodes; each contain data elements, which the elements in the core nodes acting as separators between its children. The difference between core and leaf nodes is that the core nodes have an array of children, while leaf nodes don't. Every node in the tree may be only partially full; in most cases, they are all at least 50% full (in terms of element count) except for the root node, which can be less full. Underfull nodes will steal from their neighbors or merge to remain full enough, while overfull nodes will split in two. The data elements are contained in tree-controlled buffers; they are copied into these on insertion, and overwritten on deletion. This means that the elements are not independently allocated, which reduces overhead, but also means they can't be shared between trees (and also that pointers to them are only valid until a side-effectful tree operation occurs). The overhead varies based on how dense the tree is, but is usually on the order of about 50% of the element size; the per-node overheads are very small, and so don't make a significant difference. The trees can accept arbitrary records; they accept a size and a comparator to allow them to be used for a variety of purposes. The new trees replace the AVL trees used in the range trees today. Currently, the range_seg_t structure contains three 8 byte integers of payload and two 24 byte avl_tree_node_ts to handle its storage in both an offset-sorted tree and a size-sorted tree (total size: 64 bytes). In the new model, the range seg structures are usually two 4 byte integers, but a separate one needs to exist for the size-sorted and offset-sorted tree. Between the raw size, the 50% overhead, and the double storage, the new btrees are expected to use 8*1.5*2 = 24 bytes per record, or 33.3% as much memory as the AVL trees (this is for the purposes of storing metaslab range trees; for other purposes, like scrubs, they use ~50% as much memory). We reduced the size of the payload in the range segments by teaching range trees about starting offsets and shifts; since metaslabs have a fixed starting offset, and they all operate in terms of disk sectors, we can store the ranges using 4-byte integers as long as the size of the metaslab divided by the sector size is less than 2^32. For 512-byte sectors, this is a 2^41 (or 2TB) metaslab, which with the default settings corresponds to a 256PB disk. 4k sector disks can handle metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not anticipate disks of this size in the near future, there should be almost no cases where metaslabs need 64-byte integers to store their ranges. We do still have the capability to store 64-byte integer ranges to account for cases where we are storing per-vdev (or per-dnode) trees, which could reasonably go above the limits discussed. We also do not store fill information in the compact version of the node, since it is only used for sorted scrub. We also optimized the metaslab loading process in various other ways to offset some inefficiencies in the btree model. While individual operations (find, insert, remove_from) are faster for the btree than they are for the avl tree, remove usually requires a find operation, while in the AVL tree model the element itself suffices. Some clever changes actually caused an overall speedup in metaslab loading; we use approximately 40% less cpu to load metaslabs in our tests on Illumos. Another memory and performance optimization was achieved by changing what is stored in the size-sorted trees. When a disk is heavily fragmented, the df algorithm used by default in ZFS will almost always find a number of small regions in its initial cursor-based search; it will usually only fall back to the size-sorted tree to find larger regions. If we increase the size of the cursor-based search slightly, and don't store segments that are smaller than a tunable size floor in the size-sorted tree, we can further cut memory usage down to below 20% of what the AVL trees store. This also results in further reductions in CPU time spent loading metaslabs. The 16KiB size floor was chosen because it results in substantial memory usage reduction while not usually resulting in situations where we can't find an appropriate chunk with the cursor and are forced to use an oversized chunk from the size-sorted tree. In addition, even if we do have to use an oversized chunk from the size-sorted tree, the chunk would be too small to use for ZIL allocations, so it isn't as big of a loss as it might otherwise be. And often, more small allocations will follow the initial one, and the cursor search will now find the remainder of the chunk we didn't use all of and use it for subsequent allocations. Practical testing has shown little or no change in fragmentation as a result of this change. If the size-sorted tree becomes empty while the offset sorted one still has entries, it will load all the entries from the offset sorted tree and disregard the size floor until it is unloaded again. This operation occurs rarely with the default setting, only on incredibly thoroughly fragmented pools. There are some other small changes to zdb to teach it to handle btrees, but nothing major. Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed by: Sebastien Roy seb@delphix.com Reviewed-by: Igor Kozhukhov <igor@dilos.org> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Closes #9181
2019-10-09 17:36:03 +00:00
return (TREE_CMP(a->sls_txg, b->sls_txg));
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
}
2008-11-20 20:01:55 +00:00
/*
* Create an uninitialized spa_t with the given name. Requires
* spa_namespace_lock. The caller must ensure that the spa_t doesn't already
* exist by calling spa_lookup() first.
*/
spa_t *
spa_add(const char *name, nvlist_t *config, const char *altroot)
2008-11-20 20:01:55 +00:00
{
spa_t *spa;
spa_config_dirent_t *dp;
2008-11-20 20:01:55 +00:00
ASSERT(MUTEX_HELD(&spa_namespace_lock));
spa = kmem_zalloc(sizeof (spa_t), KM_SLEEP);
2008-11-20 20:01:55 +00:00
mutex_init(&spa->spa_async_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_errlist_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_errlog_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_evicting_os_lock, NULL, MUTEX_DEFAULT, NULL);
2008-11-20 20:01:55 +00:00
mutex_init(&spa->spa_history_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_proc_lock, NULL, MUTEX_DEFAULT, NULL);
2008-11-20 20:01:55 +00:00
mutex_init(&spa->spa_props_lock, NULL, MUTEX_DEFAULT, NULL);
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
mutex_init(&spa->spa_cksum_tmpls_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_scrub_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_suspend_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_vdev_top_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_feat_stats_lock, NULL, MUTEX_DEFAULT, NULL);
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
mutex_init(&spa->spa_flushed_ms_lock, NULL, MUTEX_DEFAULT, NULL);
Add subcommand to wait for background zfs activity to complete Currently the best way to wait for the completion of a long-running operation in a pool, like a scrub or device removal, is to poll 'zpool status' and parse its output, which is neither efficient nor convenient. This change adds a 'wait' subcommand to the zpool command. When invoked, 'zpool wait' will block until a specified type of background activity completes. Currently, this subcommand can wait for any of the following: - Scrubs or resilvers to complete - Devices to initialized - Devices to be replaced - Devices to be removed - Checkpoints to be discarded - Background freeing to complete For example, a scrub that is in progress could be waited for by running zpool wait -t scrub <pool> This also adds a -w flag to the attach, checkpoint, initialize, replace, remove, and scrub subcommands. When used, this flag makes the operations kicked off by these subcommands synchronous instead of asynchronous. This functionality is implemented using a new ioctl. The type of activity to wait for is provided as input to the ioctl, and the ioctl blocks until all activity of that type has completed. An ioctl was used over other methods of kernel-userspace communiction primarily for the sake of portability. Porting Notes: This is ported from Delphix OS change DLPX-44432. The following changes were made while porting: - Added ZoL-style ioctl input declaration. - Reorganized error handling in zpool_initialize in libzfs to integrate better with changes made for TRIM support. - Fixed check for whether a checkpoint discard is in progress. Previously it also waited if the pool had a checkpoint, instead of just if a checkpoint was being discarded. - Exposed zfs_initialize_chunk_size as a ZoL-style tunable. - Updated more existing tests to make use of new 'zpool wait' functionality, tests that don't exist in Delphix OS. - Used existing ZoL tunable zfs_scan_suspend_progress, together with zinject, in place of a new tunable zfs_scan_max_blks_per_txg. - Added support for a non-integral interval argument to zpool wait. Future work: ZoL has support for trimming devices, which Delphix OS does not. In the future, 'zpool wait' could be extended to add the ability to wait for trim operations to complete. Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: John Gallagher <john.gallagher@delphix.com> Closes #9162
2019-09-14 01:09:06 +00:00
mutex_init(&spa->spa_activities_lock, NULL, MUTEX_DEFAULT, NULL);
2008-11-20 20:01:55 +00:00
cv_init(&spa->spa_async_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_evicting_os_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_proc_cv, NULL, CV_DEFAULT, NULL);
2008-11-20 20:01:55 +00:00
cv_init(&spa->spa_scrub_io_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_suspend_cv, NULL, CV_DEFAULT, NULL);
Add subcommand to wait for background zfs activity to complete Currently the best way to wait for the completion of a long-running operation in a pool, like a scrub or device removal, is to poll 'zpool status' and parse its output, which is neither efficient nor convenient. This change adds a 'wait' subcommand to the zpool command. When invoked, 'zpool wait' will block until a specified type of background activity completes. Currently, this subcommand can wait for any of the following: - Scrubs or resilvers to complete - Devices to initialized - Devices to be replaced - Devices to be removed - Checkpoints to be discarded - Background freeing to complete For example, a scrub that is in progress could be waited for by running zpool wait -t scrub <pool> This also adds a -w flag to the attach, checkpoint, initialize, replace, remove, and scrub subcommands. When used, this flag makes the operations kicked off by these subcommands synchronous instead of asynchronous. This functionality is implemented using a new ioctl. The type of activity to wait for is provided as input to the ioctl, and the ioctl blocks until all activity of that type has completed. An ioctl was used over other methods of kernel-userspace communiction primarily for the sake of portability. Porting Notes: This is ported from Delphix OS change DLPX-44432. The following changes were made while porting: - Added ZoL-style ioctl input declaration. - Reorganized error handling in zpool_initialize in libzfs to integrate better with changes made for TRIM support. - Fixed check for whether a checkpoint discard is in progress. Previously it also waited if the pool had a checkpoint, instead of just if a checkpoint was being discarded. - Exposed zfs_initialize_chunk_size as a ZoL-style tunable. - Updated more existing tests to make use of new 'zpool wait' functionality, tests that don't exist in Delphix OS. - Used existing ZoL tunable zfs_scan_suspend_progress, together with zinject, in place of a new tunable zfs_scan_max_blks_per_txg. - Added support for a non-integral interval argument to zpool wait. Future work: ZoL has support for trimming devices, which Delphix OS does not. In the future, 'zpool wait' could be extended to add the ability to wait for trim operations to complete. Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: John Gallagher <john.gallagher@delphix.com> Closes #9162
2019-09-14 01:09:06 +00:00
cv_init(&spa->spa_activities_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_waiters_cv, NULL, CV_DEFAULT, NULL);
2008-11-20 20:01:55 +00:00
for (int t = 0; t < TXG_SIZE; t++)
bplist_create(&spa->spa_free_bplist[t]);
(void) strlcpy(spa->spa_name, name, sizeof (spa->spa_name));
2008-11-20 20:01:55 +00:00
spa->spa_state = POOL_STATE_UNINITIALIZED;
spa->spa_freeze_txg = UINT64_MAX;
spa->spa_final_txg = UINT64_MAX;
spa->spa_load_max_txg = UINT64_MAX;
spa->spa_proc = &p0;
spa->spa_proc_state = SPA_PROC_NONE;
OpenZFS 9075 - Improve ZFS pool import/load process and corrupted pool recovery Some work has been done lately to improve the debugability of the ZFS pool load (and import) process. This includes: 7638 Refactor spa_load_impl into several functions 8961 SPA load/import should tell us why it failed 7277 zdb should be able to print zfs_dbgmsg's To iterate on top of that, there's a few changes that were made to make the import process more resilient and crash free. One of the first tasks during the pool load process is to parse a config provided from userland that describes what devices the pool is composed of. A vdev tree is generated from that config, and then all the vdevs are opened. The Meta Object Set (MOS) of the pool is accessed, and several metadata objects that are necessary to load the pool are read. The exact configuration of the pool is also stored inside the MOS. Since the configuration provided from userland is external and might not accurately describe the vdev tree of the pool at the txg that is being loaded, it cannot be relied upon to safely operate the pool. For that reason, the configuration in the MOS is read early on. In the past, the two configurations were compared together and if there was a mismatch then the load process was aborted and an error was returned. The latter was a good way to ensure a pool does not get corrupted, however it made the pool load process needlessly fragile in cases where the vdev configuration changed or the userland configuration was outdated. Since the MOS is stored in 3 copies, the configuration provided by userland doesn't have to be perfect in order to read its contents. Hence, a new approach has been adopted: The pool is first opened with the untrusted userland configuration just so that the real configuration can be read from the MOS. The trusted MOS configuration is then used to generate a new vdev tree and the pool is re-opened. When the pool is opened with an untrusted configuration, writes are disabled to avoid accidentally damaging it. During reads, some sanity checks are performed on block pointers to see if each DVA points to a known vdev; when the configuration is untrusted, instead of panicking the system if those checks fail we simply avoid issuing reads to the invalid DVAs. This new two-step pool load process now allows rewinding pools accross vdev tree changes such as device replacement, addition, etc. Loading a pool from an external config file in a clustering environment also becomes much safer now since the pool will import even if the config is outdated and didn't, for instance, register a recent device addition. With this code in place, it became relatively easy to implement a long-sought-after feature: the ability to import a pool with missing top level (i.e. non-redundant) devices. Note that since this almost guarantees some loss of data, this feature is for now restricted to a read-only import. Porting notes (ZTS): * Fix 'make dist' target in zpool_import * The maximum path length allowed by tar is 99 characters. Several of the new test cases exceeded this limit resulting in them not being included in the tarball. Shorten the names slightly. * Set/get tunables using accessor functions. * Get last synced txg via the "zfs_txg_history" mechanism. * Clear zinject handlers in cleanup for import_cache_device_replaced and import_rewind_device_replaced in order that the zpool can be exported if there is an error. * Increase FILESIZE to 8G in zfs-test.sh to allow for a larger ext4 file system to be created on ZFS_DISK2. Also, there's no need to partition ZFS_DISK2 at all. The partitioning had already been disabled for multipath devices. Among other things, the partitioning steals some space from the ext4 file system, makes it difficult to accurately calculate the paramters to parted and can make some of the tests fail. * Increase FS_SIZE and FILE_SIZE in the zpool_import test configuration now that FILESIZE is larger. * Write more data in order that device evacuation take lonnger in a couple tests. * Use mkdir -p to avoid errors when the directory already exists. * Remove use of sudo in import_rewind_config_changed. Authored by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Andrew Stormont <andyjstormont@gmail.com> Approved by: Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9075 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/619c0123 Closes #7459
2016-07-22 14:39:36 +00:00
spa->spa_trust_config = B_TRUE;
spa->spa_hostid = zone_get_hostid(NULL);
2008-11-20 20:01:55 +00:00
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@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
spa->spa_deadman_synctime = MSEC2NSEC(zfs_deadman_synctime_ms);
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
spa->spa_deadman_ziotime = MSEC2NSEC(zfs_deadman_ziotime_ms);
spa_set_deadman_failmode(spa, zfs_deadman_failmode);
zfs_refcount_create(&spa->spa_refcount);
spa_config_lock_init(spa);
Add visibility in to arc_read 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 <surya1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2013-09-06 23:09:05 +00:00
spa_stats_init(spa);
2008-11-20 20:01:55 +00:00
avl_add(&spa_namespace_avl, spa);
/*
* Set the alternate root, if there is one.
*/
if (altroot)
2008-11-20 20:01:55 +00:00
spa->spa_root = spa_strdup(altroot);
OpenZFS 9112 - Improve allocation performance on high-end systems Overview ======== We parallelize the allocation process by creating the concept of "allocators". There are a certain number of allocators per metaslab group, defined by the value of a tunable at pool open time. Each allocator for a given metaslab group has up to 2 active metaslabs; one "primary", and one "secondary". The primary and secondary weight mean the same thing they did in in the pre-allocator world; primary metaslabs are used for most allocations, secondary metaslabs are used for ditto blocks being allocated in the same metaslab group. There is also the CLAIM weight, which has been separated out from the other weights, but that is less important to understanding the patch. The active metaslabs for each allocator are moved from their normal place in the metaslab tree for the group to the back of the tree. This way, they will not be selected for use by other allocators searching for new metaslabs unless all the passive metaslabs are unsuitable for allocations. If that does happen, the allocators will "steal" from each other to ensure that IOs don't fail until there is truly no space left to perform allocations. In addition, the alloc queue for each metaslab group has been broken into a separate queue for each allocator. We don't want to dramatically increase the number of inflight IOs on low-end systems, because it can significantly increase txg times. On the other hand, we want to ensure that there are enough IOs for each allocator to allow for good coalescing before sending the IOs to the disk. As a result, we take a compromise path; each allocator's alloc queue max depth starts at a certain value for every txg. Every time an IO completes, we increase the max depth. This should hopefully provide a good balance between the two failure modes, while not dramatically increasing complexity. We also parallelize the spa_alloc_tree and spa_alloc_lock, which cause very similar contention when selecting IOs to allocate. This parallelization uses the same allocator scheme as metaslab selection. Performance Results =================== Performance improvements from this change can vary significantly based on the number of CPUs in the system, whether or not the system has a NUMA architecture, the speed of the drives, the values for the various tunables, and the workload being performed. For an fio async sequential write workload on a 24 core NUMA system with 256 GB of RAM and 8 128 GB SSDs, there is a roughly 25% performance improvement. Future Work =========== Analysis of the performance of the system with this patch applied shows that a significant new bottleneck is the vdev disk queues, which also need to be parallelized. Prototyping of this change has occurred, and there was a performance improvement, but more work needs to be done before its stability has been verified and it is ready to be upstreamed. Authored by: Paul Dagnelie <pcd@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Alexander Motin <mav@FreeBSD.org> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Gordon Ross <gwr@nexenta.com> Ported-by: Paul Dagnelie <pcd@delphix.com> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Porting Notes: * Fix reservation test failures by increasing tolerance. OpenZFS-issue: https://illumos.org/issues/9112 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/3f3cc3c3 Closes #7682
2018-02-12 20:56:06 +00:00
spa->spa_alloc_count = spa_allocators;
spa->spa_alloc_locks = kmem_zalloc(spa->spa_alloc_count *
sizeof (kmutex_t), KM_SLEEP);
spa->spa_alloc_trees = kmem_zalloc(spa->spa_alloc_count *
sizeof (avl_tree_t), KM_SLEEP);
for (int i = 0; i < spa->spa_alloc_count; i++) {
mutex_init(&spa->spa_alloc_locks[i], NULL, MUTEX_DEFAULT, NULL);
avl_create(&spa->spa_alloc_trees[i], zio_bookmark_compare,
sizeof (zio_t), offsetof(zio_t, io_alloc_node));
}
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
avl_create(&spa->spa_metaslabs_by_flushed, metaslab_sort_by_flushed,
sizeof (metaslab_t), offsetof(metaslab_t, ms_spa_txg_node));
avl_create(&spa->spa_sm_logs_by_txg, spa_log_sm_sort_by_txg,
sizeof (spa_log_sm_t), offsetof(spa_log_sm_t, sls_node));
list_create(&spa->spa_log_summary, sizeof (log_summary_entry_t),
offsetof(log_summary_entry_t, lse_node));
OpenZFS 7090 - zfs should throttle allocations OpenZFS 7090 - zfs should throttle allocations Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: Alex Reece <alex@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Paul Dagnelie <paul.dagnelie@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Approved by: Matthew Ahrens <mahrens@delphix.com> Ported-by: Don Brady <don.brady@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> When write I/Os are issued, they are issued in block order but the ZIO pipeline will drive them asynchronously through the allocation stage which can result in blocks being allocated out-of-order. It would be nice to preserve as much of the logical order as possible. In addition, the allocations are equally scattered across all top-level VDEVs but not all top-level VDEVs are created equally. The pipeline should be able to detect devices that are more capable of handling allocations and should allocate more blocks to those devices. This allows for dynamic allocation distribution when devices are imbalanced as fuller devices will tend to be slower than empty devices. The change includes a new pool-wide allocation queue which would throttle and order allocations in the ZIO pipeline. The queue would be ordered by issued time and offset and would provide an initial amount of allocation of work to each top-level vdev. The allocation logic utilizes a reservation system to reserve allocations that will be performed by the allocator. Once an allocation is successfully completed it's scheduled on a given top-level vdev. Each top-level vdev maintains a maximum number of allocations that it can handle (mg_alloc_queue_depth). The pool-wide reserved allocations (top-levels * mg_alloc_queue_depth) are distributed across the top-level vdevs metaslab groups and round robin across all eligible metaslab groups to distribute the work. As top-levels complete their work, they receive additional work from the pool-wide allocation queue until the allocation queue is emptied. OpenZFS-issue: https://www.illumos.org/issues/7090 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/4756c3d7 Closes #5258 Porting Notes: - Maintained minimal stack in zio_done - Preserve linux-specific io sizes in zio_write_compress - Added module params and documentation - Updated to use optimize AVL cmp macros
2016-10-14 00:59:18 +00:00
/*
* Every pool starts with the default cachefile
*/
list_create(&spa->spa_config_list, sizeof (spa_config_dirent_t),
offsetof(spa_config_dirent_t, scd_link));
dp = kmem_zalloc(sizeof (spa_config_dirent_t), KM_SLEEP);
dp->scd_path = altroot ? NULL : spa_strdup(spa_config_path);
list_insert_head(&spa->spa_config_list, dp);
VERIFY(nvlist_alloc(&spa->spa_load_info, NV_UNIQUE_NAME,
KM_SLEEP) == 0);
if (config != NULL) {
nvlist_t *features;
if (nvlist_lookup_nvlist(config, ZPOOL_CONFIG_FEATURES_FOR_READ,
&features) == 0) {
VERIFY(nvlist_dup(features, &spa->spa_label_features,
0) == 0);
}
VERIFY(nvlist_dup(config, &spa->spa_config, 0) == 0);
}
if (spa->spa_label_features == NULL) {
VERIFY(nvlist_alloc(&spa->spa_label_features, NV_UNIQUE_NAME,
KM_SLEEP) == 0);
}
spa->spa_min_ashift = INT_MAX;
spa->spa_max_ashift = 0;
/* Reset cached value */
spa->spa_dedup_dspace = ~0ULL;
/*
* As a pool is being created, treat all features as disabled by
* setting SPA_FEATURE_DISABLED for all entries in the feature
* refcount cache.
*/
for (int i = 0; i < SPA_FEATURES; i++) {
spa->spa_feat_refcount_cache[i] = SPA_FEATURE_DISABLED;
}
MMP writes rotate over leaves Instead of choosing a leaf vdev quasi-randomly, by starting at the root vdev and randomly choosing children, rotate over leaves to issue MMP writes. This fixes an issue in a pool whose top-level vdevs have different numbers of leaves. The issue is that the frequency at which individual leaves are chosen for MMP writes is based not on the total number of leaves but based on how many siblings the leaves have. For example, in a pool like this: root-vdev +------+---------------+ vdev1 vdev2 | | | +------+-----+-----+----+ disk1 disk2 disk3 disk4 disk5 disk6 vdev1 and vdev2 will each be chosen 50% of the time. Every time vdev1 is chosen, disk1 will be chosen. However, every time vdev2 is chosen, disk2 is chosen 20% of the time. As a result, disk1 will be sent 5x as many MMP writes as disk2. This may create wear issues in the case of SSDs. It also reduces the effectiveness of MMP as it depends on the writes being evenly distributed for the case where some devices fail or are partitioned. The new code maintains a list of leaf vdevs in the pool. MMP records the last leaf used for an MMP write in mmp->mmp_last_leaf. To choose the next leaf, MMP starts at mmp->mmp_last_leaf and traverses the list, continuing from the head if the tail is reached. It stops when a suitable leaf is found or all leaves have been examined. Added a test to verify MMP write distribution is even. Reviewed-by: Tom Caputi <tcaputi@datto.com> Reviewed-by: Kash Pande <kash@tripleback.net> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: loli10K <ezomori.nozomu@gmail.com> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #7953
2019-03-12 17:37:06 +00:00
list_create(&spa->spa_leaf_list, sizeof (vdev_t),
offsetof(vdev_t, vdev_leaf_node));
2008-11-20 20:01:55 +00:00
return (spa);
}
/*
* Removes a spa_t from the namespace, freeing up any memory used. Requires
* spa_namespace_lock. This is called only after the spa_t has been closed and
* deactivated.
*/
void
spa_remove(spa_t *spa)
{
spa_config_dirent_t *dp;
2008-11-20 20:01:55 +00:00
ASSERT(MUTEX_HELD(&spa_namespace_lock));
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
ASSERT(spa_state(spa) == POOL_STATE_UNINITIALIZED);
ASSERT3U(zfs_refcount_count(&spa->spa_refcount), ==, 0);
Add subcommand to wait for background zfs activity to complete Currently the best way to wait for the completion of a long-running operation in a pool, like a scrub or device removal, is to poll 'zpool status' and parse its output, which is neither efficient nor convenient. This change adds a 'wait' subcommand to the zpool command. When invoked, 'zpool wait' will block until a specified type of background activity completes. Currently, this subcommand can wait for any of the following: - Scrubs or resilvers to complete - Devices to initialized - Devices to be replaced - Devices to be removed - Checkpoints to be discarded - Background freeing to complete For example, a scrub that is in progress could be waited for by running zpool wait -t scrub <pool> This also adds a -w flag to the attach, checkpoint, initialize, replace, remove, and scrub subcommands. When used, this flag makes the operations kicked off by these subcommands synchronous instead of asynchronous. This functionality is implemented using a new ioctl. The type of activity to wait for is provided as input to the ioctl, and the ioctl blocks until all activity of that type has completed. An ioctl was used over other methods of kernel-userspace communiction primarily for the sake of portability. Porting Notes: This is ported from Delphix OS change DLPX-44432. The following changes were made while porting: - Added ZoL-style ioctl input declaration. - Reorganized error handling in zpool_initialize in libzfs to integrate better with changes made for TRIM support. - Fixed check for whether a checkpoint discard is in progress. Previously it also waited if the pool had a checkpoint, instead of just if a checkpoint was being discarded. - Exposed zfs_initialize_chunk_size as a ZoL-style tunable. - Updated more existing tests to make use of new 'zpool wait' functionality, tests that don't exist in Delphix OS. - Used existing ZoL tunable zfs_scan_suspend_progress, together with zinject, in place of a new tunable zfs_scan_max_blks_per_txg. - Added support for a non-integral interval argument to zpool wait. Future work: ZoL has support for trimming devices, which Delphix OS does not. In the future, 'zpool wait' could be extended to add the ability to wait for trim operations to complete. Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: John Gallagher <john.gallagher@delphix.com> Closes #9162
2019-09-14 01:09:06 +00:00
ASSERT0(spa->spa_waiters);
2008-11-20 20:01:55 +00:00
nvlist_free(spa->spa_config_splitting);
2008-11-20 20:01:55 +00:00
avl_remove(&spa_namespace_avl, spa);
cv_broadcast(&spa_namespace_cv);
if (spa->spa_root)
2008-11-20 20:01:55 +00:00
spa_strfree(spa->spa_root);
while ((dp = list_head(&spa->spa_config_list)) != NULL) {
list_remove(&spa->spa_config_list, dp);
if (dp->scd_path != NULL)
spa_strfree(dp->scd_path);
kmem_free(dp, sizeof (spa_config_dirent_t));
}
2008-11-20 20:01:55 +00:00
OpenZFS 9112 - Improve allocation performance on high-end systems Overview ======== We parallelize the allocation process by creating the concept of "allocators". There are a certain number of allocators per metaslab group, defined by the value of a tunable at pool open time. Each allocator for a given metaslab group has up to 2 active metaslabs; one "primary", and one "secondary". The primary and secondary weight mean the same thing they did in in the pre-allocator world; primary metaslabs are used for most allocations, secondary metaslabs are used for ditto blocks being allocated in the same metaslab group. There is also the CLAIM weight, which has been separated out from the other weights, but that is less important to understanding the patch. The active metaslabs for each allocator are moved from their normal place in the metaslab tree for the group to the back of the tree. This way, they will not be selected for use by other allocators searching for new metaslabs unless all the passive metaslabs are unsuitable for allocations. If that does happen, the allocators will "steal" from each other to ensure that IOs don't fail until there is truly no space left to perform allocations. In addition, the alloc queue for each metaslab group has been broken into a separate queue for each allocator. We don't want to dramatically increase the number of inflight IOs on low-end systems, because it can significantly increase txg times. On the other hand, we want to ensure that there are enough IOs for each allocator to allow for good coalescing before sending the IOs to the disk. As a result, we take a compromise path; each allocator's alloc queue max depth starts at a certain value for every txg. Every time an IO completes, we increase the max depth. This should hopefully provide a good balance between the two failure modes, while not dramatically increasing complexity. We also parallelize the spa_alloc_tree and spa_alloc_lock, which cause very similar contention when selecting IOs to allocate. This parallelization uses the same allocator scheme as metaslab selection. Performance Results =================== Performance improvements from this change can vary significantly based on the number of CPUs in the system, whether or not the system has a NUMA architecture, the speed of the drives, the values for the various tunables, and the workload being performed. For an fio async sequential write workload on a 24 core NUMA system with 256 GB of RAM and 8 128 GB SSDs, there is a roughly 25% performance improvement. Future Work =========== Analysis of the performance of the system with this patch applied shows that a significant new bottleneck is the vdev disk queues, which also need to be parallelized. Prototyping of this change has occurred, and there was a performance improvement, but more work needs to be done before its stability has been verified and it is ready to be upstreamed. Authored by: Paul Dagnelie <pcd@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Alexander Motin <mav@FreeBSD.org> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Gordon Ross <gwr@nexenta.com> Ported-by: Paul Dagnelie <pcd@delphix.com> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Porting Notes: * Fix reservation test failures by increasing tolerance. OpenZFS-issue: https://illumos.org/issues/9112 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/3f3cc3c3 Closes #7682
2018-02-12 20:56:06 +00:00
for (int i = 0; i < spa->spa_alloc_count; i++) {
avl_destroy(&spa->spa_alloc_trees[i]);
mutex_destroy(&spa->spa_alloc_locks[i]);
}
kmem_free(spa->spa_alloc_locks, spa->spa_alloc_count *
sizeof (kmutex_t));
kmem_free(spa->spa_alloc_trees, spa->spa_alloc_count *
sizeof (avl_tree_t));
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
avl_destroy(&spa->spa_metaslabs_by_flushed);
avl_destroy(&spa->spa_sm_logs_by_txg);
list_destroy(&spa->spa_log_summary);
list_destroy(&spa->spa_config_list);
MMP writes rotate over leaves Instead of choosing a leaf vdev quasi-randomly, by starting at the root vdev and randomly choosing children, rotate over leaves to issue MMP writes. This fixes an issue in a pool whose top-level vdevs have different numbers of leaves. The issue is that the frequency at which individual leaves are chosen for MMP writes is based not on the total number of leaves but based on how many siblings the leaves have. For example, in a pool like this: root-vdev +------+---------------+ vdev1 vdev2 | | | +------+-----+-----+----+ disk1 disk2 disk3 disk4 disk5 disk6 vdev1 and vdev2 will each be chosen 50% of the time. Every time vdev1 is chosen, disk1 will be chosen. However, every time vdev2 is chosen, disk2 is chosen 20% of the time. As a result, disk1 will be sent 5x as many MMP writes as disk2. This may create wear issues in the case of SSDs. It also reduces the effectiveness of MMP as it depends on the writes being evenly distributed for the case where some devices fail or are partitioned. The new code maintains a list of leaf vdevs in the pool. MMP records the last leaf used for an MMP write in mmp->mmp_last_leaf. To choose the next leaf, MMP starts at mmp->mmp_last_leaf and traverses the list, continuing from the head if the tail is reached. It stops when a suitable leaf is found or all leaves have been examined. Added a test to verify MMP write distribution is even. Reviewed-by: Tom Caputi <tcaputi@datto.com> Reviewed-by: Kash Pande <kash@tripleback.net> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: loli10K <ezomori.nozomu@gmail.com> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #7953
2019-03-12 17:37:06 +00:00
list_destroy(&spa->spa_leaf_list);
2008-11-20 20:01:55 +00:00
nvlist_free(spa->spa_label_features);
nvlist_free(spa->spa_load_info);
nvlist_free(spa->spa_feat_stats);
2008-11-20 20:01:55 +00:00
spa_config_set(spa, NULL);
zfs_refcount_destroy(&spa->spa_refcount);
2008-11-20 20:01:55 +00:00
Add visibility in to arc_read 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 <surya1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2013-09-06 23:09:05 +00:00
spa_stats_destroy(spa);
spa_config_lock_destroy(spa);
2008-11-20 20:01:55 +00:00
for (int t = 0; t < TXG_SIZE; t++)
bplist_destroy(&spa->spa_free_bplist[t]);
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
zio_checksum_templates_free(spa);
2008-11-20 20:01:55 +00:00
cv_destroy(&spa->spa_async_cv);
cv_destroy(&spa->spa_evicting_os_cv);
cv_destroy(&spa->spa_proc_cv);
2008-11-20 20:01:55 +00:00
cv_destroy(&spa->spa_scrub_io_cv);
cv_destroy(&spa->spa_suspend_cv);
Add subcommand to wait for background zfs activity to complete Currently the best way to wait for the completion of a long-running operation in a pool, like a scrub or device removal, is to poll 'zpool status' and parse its output, which is neither efficient nor convenient. This change adds a 'wait' subcommand to the zpool command. When invoked, 'zpool wait' will block until a specified type of background activity completes. Currently, this subcommand can wait for any of the following: - Scrubs or resilvers to complete - Devices to initialized - Devices to be replaced - Devices to be removed - Checkpoints to be discarded - Background freeing to complete For example, a scrub that is in progress could be waited for by running zpool wait -t scrub <pool> This also adds a -w flag to the attach, checkpoint, initialize, replace, remove, and scrub subcommands. When used, this flag makes the operations kicked off by these subcommands synchronous instead of asynchronous. This functionality is implemented using a new ioctl. The type of activity to wait for is provided as input to the ioctl, and the ioctl blocks until all activity of that type has completed. An ioctl was used over other methods of kernel-userspace communiction primarily for the sake of portability. Porting Notes: This is ported from Delphix OS change DLPX-44432. The following changes were made while porting: - Added ZoL-style ioctl input declaration. - Reorganized error handling in zpool_initialize in libzfs to integrate better with changes made for TRIM support. - Fixed check for whether a checkpoint discard is in progress. Previously it also waited if the pool had a checkpoint, instead of just if a checkpoint was being discarded. - Exposed zfs_initialize_chunk_size as a ZoL-style tunable. - Updated more existing tests to make use of new 'zpool wait' functionality, tests that don't exist in Delphix OS. - Used existing ZoL tunable zfs_scan_suspend_progress, together with zinject, in place of a new tunable zfs_scan_max_blks_per_txg. - Added support for a non-integral interval argument to zpool wait. Future work: ZoL has support for trimming devices, which Delphix OS does not. In the future, 'zpool wait' could be extended to add the ability to wait for trim operations to complete. Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: John Gallagher <john.gallagher@delphix.com> Closes #9162
2019-09-14 01:09:06 +00:00
cv_destroy(&spa->spa_activities_cv);
cv_destroy(&spa->spa_waiters_cv);
2008-11-20 20:01:55 +00:00
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
mutex_destroy(&spa->spa_flushed_ms_lock);
2008-11-20 20:01:55 +00:00
mutex_destroy(&spa->spa_async_lock);
mutex_destroy(&spa->spa_errlist_lock);
mutex_destroy(&spa->spa_errlog_lock);
mutex_destroy(&spa->spa_evicting_os_lock);
2008-11-20 20:01:55 +00:00
mutex_destroy(&spa->spa_history_lock);
mutex_destroy(&spa->spa_proc_lock);
2008-11-20 20:01:55 +00:00
mutex_destroy(&spa->spa_props_lock);
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
mutex_destroy(&spa->spa_cksum_tmpls_lock);
mutex_destroy(&spa->spa_scrub_lock);
mutex_destroy(&spa->spa_suspend_lock);
mutex_destroy(&spa->spa_vdev_top_lock);
mutex_destroy(&spa->spa_feat_stats_lock);
Add subcommand to wait for background zfs activity to complete Currently the best way to wait for the completion of a long-running operation in a pool, like a scrub or device removal, is to poll 'zpool status' and parse its output, which is neither efficient nor convenient. This change adds a 'wait' subcommand to the zpool command. When invoked, 'zpool wait' will block until a specified type of background activity completes. Currently, this subcommand can wait for any of the following: - Scrubs or resilvers to complete - Devices to initialized - Devices to be replaced - Devices to be removed - Checkpoints to be discarded - Background freeing to complete For example, a scrub that is in progress could be waited for by running zpool wait -t scrub <pool> This also adds a -w flag to the attach, checkpoint, initialize, replace, remove, and scrub subcommands. When used, this flag makes the operations kicked off by these subcommands synchronous instead of asynchronous. This functionality is implemented using a new ioctl. The type of activity to wait for is provided as input to the ioctl, and the ioctl blocks until all activity of that type has completed. An ioctl was used over other methods of kernel-userspace communiction primarily for the sake of portability. Porting Notes: This is ported from Delphix OS change DLPX-44432. The following changes were made while porting: - Added ZoL-style ioctl input declaration. - Reorganized error handling in zpool_initialize in libzfs to integrate better with changes made for TRIM support. - Fixed check for whether a checkpoint discard is in progress. Previously it also waited if the pool had a checkpoint, instead of just if a checkpoint was being discarded. - Exposed zfs_initialize_chunk_size as a ZoL-style tunable. - Updated more existing tests to make use of new 'zpool wait' functionality, tests that don't exist in Delphix OS. - Used existing ZoL tunable zfs_scan_suspend_progress, together with zinject, in place of a new tunable zfs_scan_max_blks_per_txg. - Added support for a non-integral interval argument to zpool wait. Future work: ZoL has support for trimming devices, which Delphix OS does not. In the future, 'zpool wait' could be extended to add the ability to wait for trim operations to complete. Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: John Gallagher <john.gallagher@delphix.com> Closes #9162
2019-09-14 01:09:06 +00:00
mutex_destroy(&spa->spa_activities_lock);
2008-11-20 20:01:55 +00:00
kmem_free(spa, sizeof (spa_t));
}
/*
* Given a pool, return the next pool in the namespace, or NULL if there is
* none. If 'prev' is NULL, return the first pool.
*/
spa_t *
spa_next(spa_t *prev)
{
ASSERT(MUTEX_HELD(&spa_namespace_lock));
if (prev)
return (AVL_NEXT(&spa_namespace_avl, prev));
else
return (avl_first(&spa_namespace_avl));
}
/*
* ==========================================================================
* SPA refcount functions
* ==========================================================================
*/
/*
* Add a reference to the given spa_t. Must have at least one reference, or
* have the namespace lock held.
*/
void
spa_open_ref(spa_t *spa, void *tag)
{
ASSERT(zfs_refcount_count(&spa->spa_refcount) >= spa->spa_minref ||
2008-11-20 20:01:55 +00:00
MUTEX_HELD(&spa_namespace_lock));
(void) zfs_refcount_add(&spa->spa_refcount, tag);
2008-11-20 20:01:55 +00:00
}
/*
* Remove a reference to the given spa_t. Must have at least one reference, or
* have the namespace lock held.
*/
void
spa_close(spa_t *spa, void *tag)
{
ASSERT(zfs_refcount_count(&spa->spa_refcount) > spa->spa_minref ||
2008-11-20 20:01:55 +00:00
MUTEX_HELD(&spa_namespace_lock));
(void) zfs_refcount_remove(&spa->spa_refcount, tag);
2008-11-20 20:01:55 +00:00
}
/*
* Remove a reference to the given spa_t held by a dsl dir that is
* being asynchronously released. Async releases occur from a taskq
* performing eviction of dsl datasets and dirs. The namespace lock
* isn't held and the hold by the object being evicted may contribute to
* spa_minref (e.g. dataset or directory released during pool export),
* so the asserts in spa_close() do not apply.
*/
void
spa_async_close(spa_t *spa, void *tag)
{
(void) zfs_refcount_remove(&spa->spa_refcount, tag);
}
2008-11-20 20:01:55 +00:00
/*
* Check to see if the spa refcount is zero. Must be called with
* spa_namespace_lock held. We really compare against spa_minref, which is the
2008-11-20 20:01:55 +00:00
* number of references acquired when opening a pool
*/
boolean_t
spa_refcount_zero(spa_t *spa)
{
ASSERT(MUTEX_HELD(&spa_namespace_lock));
return (zfs_refcount_count(&spa->spa_refcount) == spa->spa_minref);
2008-11-20 20:01:55 +00:00
}
/*
* ==========================================================================
* SPA spare and l2cache tracking
* ==========================================================================
*/
/*
* Hot spares and cache devices are tracked using the same code below,
* for 'auxiliary' devices.
*/
typedef struct spa_aux {
uint64_t aux_guid;
uint64_t aux_pool;
avl_node_t aux_avl;
int aux_count;
} spa_aux_t;
Performance optimization of AVL tree comparator functions perf: 2.75x faster ddt_entry_compare() First 256bits of ddt_key_t is a block checksum, which are expected to be close to random data. Hence, on average, comparison only needs to look at first few bytes of the keys. To reduce number of conditional jump instructions, the result is computed as: sign(memcmp(k1, k2)). Sign of an integer 'a' can be obtained as: `(0 < a) - (a < 0)` := {-1, 0, 1} , which is computed efficiently. Synthetic performance evaluation of original and new algorithm over 1G random keys on 2.6GHz Intel(R) Xeon(R) CPU E5-2660 v3: old 6.85789 s new 2.49089 s perf: 2.8x faster vdev_queue_offset_compare() and vdev_queue_timestamp_compare() Compute the result directly instead of using conditionals perf: zfs_range_compare() Speedup between 1.1x - 2.5x, depending on compiler version and optimization level. perf: spa_error_entry_compare() `bcmp()` is not suitable for comparator use. Use `memcmp()` instead. perf: 2.8x faster metaslab_compare() and metaslab_rangesize_compare() perf: 2.8x faster zil_bp_compare() perf: 2.8x faster mze_compare() perf: faster dbuf_compare() perf: faster compares in spa_misc perf: 2.8x faster layout_hash_compare() perf: 2.8x faster space_reftree_compare() perf: libzfs: faster avl tree comparators perf: guid_compare() perf: dsl_deadlist_compare() perf: perm_set_compare() perf: 2x faster range_tree_seg_compare() perf: faster unique_compare() perf: faster vdev_cache _compare() perf: faster vdev_uberblock_compare() perf: faster fuid _compare() perf: faster zfs_znode_hold_compare() Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Richard Elling <richard.elling@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5033
2016-08-27 18:12:53 +00:00
static inline int
2008-11-20 20:01:55 +00:00
spa_aux_compare(const void *a, const void *b)
{
Performance optimization of AVL tree comparator functions perf: 2.75x faster ddt_entry_compare() First 256bits of ddt_key_t is a block checksum, which are expected to be close to random data. Hence, on average, comparison only needs to look at first few bytes of the keys. To reduce number of conditional jump instructions, the result is computed as: sign(memcmp(k1, k2)). Sign of an integer 'a' can be obtained as: `(0 < a) - (a < 0)` := {-1, 0, 1} , which is computed efficiently. Synthetic performance evaluation of original and new algorithm over 1G random keys on 2.6GHz Intel(R) Xeon(R) CPU E5-2660 v3: old 6.85789 s new 2.49089 s perf: 2.8x faster vdev_queue_offset_compare() and vdev_queue_timestamp_compare() Compute the result directly instead of using conditionals perf: zfs_range_compare() Speedup between 1.1x - 2.5x, depending on compiler version and optimization level. perf: spa_error_entry_compare() `bcmp()` is not suitable for comparator use. Use `memcmp()` instead. perf: 2.8x faster metaslab_compare() and metaslab_rangesize_compare() perf: 2.8x faster zil_bp_compare() perf: 2.8x faster mze_compare() perf: faster dbuf_compare() perf: faster compares in spa_misc perf: 2.8x faster layout_hash_compare() perf: 2.8x faster space_reftree_compare() perf: libzfs: faster avl tree comparators perf: guid_compare() perf: dsl_deadlist_compare() perf: perm_set_compare() perf: 2x faster range_tree_seg_compare() perf: faster unique_compare() perf: faster vdev_cache _compare() perf: faster vdev_uberblock_compare() perf: faster fuid _compare() perf: faster zfs_znode_hold_compare() Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Richard Elling <richard.elling@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5033
2016-08-27 18:12:53 +00:00
const spa_aux_t *sa = (const spa_aux_t *)a;
const spa_aux_t *sb = (const spa_aux_t *)b;
2008-11-20 20:01:55 +00:00
Reduce loaded range tree memory usage This patch implements a new tree structure for ZFS, and uses it to store range trees more efficiently. The new structure is approximately a B-tree, though there are some small differences from the usual characterizations. The tree has core nodes and leaf nodes; each contain data elements, which the elements in the core nodes acting as separators between its children. The difference between core and leaf nodes is that the core nodes have an array of children, while leaf nodes don't. Every node in the tree may be only partially full; in most cases, they are all at least 50% full (in terms of element count) except for the root node, which can be less full. Underfull nodes will steal from their neighbors or merge to remain full enough, while overfull nodes will split in two. The data elements are contained in tree-controlled buffers; they are copied into these on insertion, and overwritten on deletion. This means that the elements are not independently allocated, which reduces overhead, but also means they can't be shared between trees (and also that pointers to them are only valid until a side-effectful tree operation occurs). The overhead varies based on how dense the tree is, but is usually on the order of about 50% of the element size; the per-node overheads are very small, and so don't make a significant difference. The trees can accept arbitrary records; they accept a size and a comparator to allow them to be used for a variety of purposes. The new trees replace the AVL trees used in the range trees today. Currently, the range_seg_t structure contains three 8 byte integers of payload and two 24 byte avl_tree_node_ts to handle its storage in both an offset-sorted tree and a size-sorted tree (total size: 64 bytes). In the new model, the range seg structures are usually two 4 byte integers, but a separate one needs to exist for the size-sorted and offset-sorted tree. Between the raw size, the 50% overhead, and the double storage, the new btrees are expected to use 8*1.5*2 = 24 bytes per record, or 33.3% as much memory as the AVL trees (this is for the purposes of storing metaslab range trees; for other purposes, like scrubs, they use ~50% as much memory). We reduced the size of the payload in the range segments by teaching range trees about starting offsets and shifts; since metaslabs have a fixed starting offset, and they all operate in terms of disk sectors, we can store the ranges using 4-byte integers as long as the size of the metaslab divided by the sector size is less than 2^32. For 512-byte sectors, this is a 2^41 (or 2TB) metaslab, which with the default settings corresponds to a 256PB disk. 4k sector disks can handle metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not anticipate disks of this size in the near future, there should be almost no cases where metaslabs need 64-byte integers to store their ranges. We do still have the capability to store 64-byte integer ranges to account for cases where we are storing per-vdev (or per-dnode) trees, which could reasonably go above the limits discussed. We also do not store fill information in the compact version of the node, since it is only used for sorted scrub. We also optimized the metaslab loading process in various other ways to offset some inefficiencies in the btree model. While individual operations (find, insert, remove_from) are faster for the btree than they are for the avl tree, remove usually requires a find operation, while in the AVL tree model the element itself suffices. Some clever changes actually caused an overall speedup in metaslab loading; we use approximately 40% less cpu to load metaslabs in our tests on Illumos. Another memory and performance optimization was achieved by changing what is stored in the size-sorted trees. When a disk is heavily fragmented, the df algorithm used by default in ZFS will almost always find a number of small regions in its initial cursor-based search; it will usually only fall back to the size-sorted tree to find larger regions. If we increase the size of the cursor-based search slightly, and don't store segments that are smaller than a tunable size floor in the size-sorted tree, we can further cut memory usage down to below 20% of what the AVL trees store. This also results in further reductions in CPU time spent loading metaslabs. The 16KiB size floor was chosen because it results in substantial memory usage reduction while not usually resulting in situations where we can't find an appropriate chunk with the cursor and are forced to use an oversized chunk from the size-sorted tree. In addition, even if we do have to use an oversized chunk from the size-sorted tree, the chunk would be too small to use for ZIL allocations, so it isn't as big of a loss as it might otherwise be. And often, more small allocations will follow the initial one, and the cursor search will now find the remainder of the chunk we didn't use all of and use it for subsequent allocations. Practical testing has shown little or no change in fragmentation as a result of this change. If the size-sorted tree becomes empty while the offset sorted one still has entries, it will load all the entries from the offset sorted tree and disregard the size floor until it is unloaded again. This operation occurs rarely with the default setting, only on incredibly thoroughly fragmented pools. There are some other small changes to zdb to teach it to handle btrees, but nothing major. Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed by: Sebastien Roy seb@delphix.com Reviewed-by: Igor Kozhukhov <igor@dilos.org> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Closes #9181
2019-10-09 17:36:03 +00:00
return (TREE_CMP(sa->aux_guid, sb->aux_guid));
2008-11-20 20:01:55 +00:00
}
static void
2008-11-20 20:01:55 +00:00
spa_aux_add(vdev_t *vd, avl_tree_t *avl)
{
avl_index_t where;
spa_aux_t search;
spa_aux_t *aux;
search.aux_guid = vd->vdev_guid;
if ((aux = avl_find(avl, &search, &where)) != NULL) {
aux->aux_count++;
} else {
aux = kmem_zalloc(sizeof (spa_aux_t), KM_SLEEP);
2008-11-20 20:01:55 +00:00
aux->aux_guid = vd->vdev_guid;
aux->aux_count = 1;
avl_insert(avl, aux, where);
}
}
static void
2008-11-20 20:01:55 +00:00
spa_aux_remove(vdev_t *vd, avl_tree_t *avl)
{
spa_aux_t search;
spa_aux_t *aux;
avl_index_t where;
search.aux_guid = vd->vdev_guid;
aux = avl_find(avl, &search, &where);
ASSERT(aux != NULL);
if (--aux->aux_count == 0) {
avl_remove(avl, aux);
kmem_free(aux, sizeof (spa_aux_t));
} else if (aux->aux_pool == spa_guid(vd->vdev_spa)) {
aux->aux_pool = 0ULL;
}
}
static boolean_t
spa_aux_exists(uint64_t guid, uint64_t *pool, int *refcnt, avl_tree_t *avl)
2008-11-20 20:01:55 +00:00
{
spa_aux_t search, *found;
search.aux_guid = guid;
found = avl_find(avl, &search, NULL);
2008-11-20 20:01:55 +00:00
if (pool) {
if (found)
*pool = found->aux_pool;
else
*pool = 0ULL;
}
if (refcnt) {
if (found)
*refcnt = found->aux_count;
else
*refcnt = 0;
}
2008-11-20 20:01:55 +00:00
return (found != NULL);
}
static void
2008-11-20 20:01:55 +00:00
spa_aux_activate(vdev_t *vd, avl_tree_t *avl)
{
spa_aux_t search, *found;
avl_index_t where;
search.aux_guid = vd->vdev_guid;
found = avl_find(avl, &search, &where);
ASSERT(found != NULL);
ASSERT(found->aux_pool == 0ULL);
found->aux_pool = spa_guid(vd->vdev_spa);
}
/*
* Spares are tracked globally due to the following constraints:
*
* - A spare may be part of multiple pools.
* - A spare may be added to a pool even if it's actively in use within
* another pool.
* - A spare in use in any pool can only be the source of a replacement if
* the target is a spare in the same pool.
*
* We keep track of all spares on the system through the use of a reference
* counted AVL tree. When a vdev is added as a spare, or used as a replacement
* spare, then we bump the reference count in the AVL tree. In addition, we set
* the 'vdev_isspare' member to indicate that the device is a spare (active or
* inactive). When a spare is made active (used to replace a device in the
* pool), we also keep track of which pool its been made a part of.
*
* The 'spa_spare_lock' protects the AVL tree. These functions are normally
* called under the spa_namespace lock as part of vdev reconfiguration. The
* separate spare lock exists for the status query path, which does not need to
* be completely consistent with respect to other vdev configuration changes.
*/
static int
spa_spare_compare(const void *a, const void *b)
{
return (spa_aux_compare(a, b));
}
void
spa_spare_add(vdev_t *vd)
{
mutex_enter(&spa_spare_lock);
ASSERT(!vd->vdev_isspare);
spa_aux_add(vd, &spa_spare_avl);
vd->vdev_isspare = B_TRUE;
mutex_exit(&spa_spare_lock);
}
void
spa_spare_remove(vdev_t *vd)
{
mutex_enter(&spa_spare_lock);
ASSERT(vd->vdev_isspare);
spa_aux_remove(vd, &spa_spare_avl);
vd->vdev_isspare = B_FALSE;
mutex_exit(&spa_spare_lock);
}
boolean_t
spa_spare_exists(uint64_t guid, uint64_t *pool, int *refcnt)
2008-11-20 20:01:55 +00:00
{
boolean_t found;
mutex_enter(&spa_spare_lock);
found = spa_aux_exists(guid, pool, refcnt, &spa_spare_avl);
2008-11-20 20:01:55 +00:00
mutex_exit(&spa_spare_lock);
return (found);
}
void
spa_spare_activate(vdev_t *vd)
{
mutex_enter(&spa_spare_lock);
ASSERT(vd->vdev_isspare);
spa_aux_activate(vd, &spa_spare_avl);
mutex_exit(&spa_spare_lock);
}
/*
* Level 2 ARC devices are tracked globally for the same reasons as spares.
* Cache devices currently only support one pool per cache device, and so
* for these devices the aux reference count is currently unused beyond 1.
*/
static int
spa_l2cache_compare(const void *a, const void *b)
{
return (spa_aux_compare(a, b));
}
void
spa_l2cache_add(vdev_t *vd)
{
mutex_enter(&spa_l2cache_lock);
ASSERT(!vd->vdev_isl2cache);
spa_aux_add(vd, &spa_l2cache_avl);
vd->vdev_isl2cache = B_TRUE;
mutex_exit(&spa_l2cache_lock);
}
void
spa_l2cache_remove(vdev_t *vd)
{
mutex_enter(&spa_l2cache_lock);
ASSERT(vd->vdev_isl2cache);
spa_aux_remove(vd, &spa_l2cache_avl);
vd->vdev_isl2cache = B_FALSE;
mutex_exit(&spa_l2cache_lock);
}
boolean_t
spa_l2cache_exists(uint64_t guid, uint64_t *pool)
{
boolean_t found;
mutex_enter(&spa_l2cache_lock);
found = spa_aux_exists(guid, pool, NULL, &spa_l2cache_avl);
2008-11-20 20:01:55 +00:00
mutex_exit(&spa_l2cache_lock);
return (found);
}
void
spa_l2cache_activate(vdev_t *vd)
{
mutex_enter(&spa_l2cache_lock);
ASSERT(vd->vdev_isl2cache);
spa_aux_activate(vd, &spa_l2cache_avl);
mutex_exit(&spa_l2cache_lock);
}
/*
* ==========================================================================
* SPA vdev locking
* ==========================================================================
*/
/*
* Lock the given spa_t for the purpose of adding or removing a vdev.
* Grabs the global spa_namespace_lock plus the spa config lock for writing.
* It returns the next transaction group for the spa_t.
*/
uint64_t
spa_vdev_enter(spa_t *spa)
{
mutex_enter(&spa->spa_vdev_top_lock);
2008-11-20 20:01:55 +00:00
mutex_enter(&spa_namespace_lock);
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
vdev_autotrim_stop_all(spa);
return (spa_vdev_config_enter(spa));
}
Add device rebuild feature The device_rebuild feature enables sequential reconstruction when resilvering. Mirror vdevs can be rebuilt in LBA order which may more quickly restore redundancy depending on the pools average block size, overall fragmentation and the performance characteristics of the devices. However, block checksums cannot be verified as part of the rebuild thus a scrub is automatically started after the sequential resilver completes. The new '-s' option has been added to the `zpool attach` and `zpool replace` command to request sequential reconstruction instead of healing reconstruction when resilvering. zpool attach -s <pool> <existing vdev> <new vdev> zpool replace -s <pool> <old vdev> <new vdev> The `zpool status` output has been updated to report the progress of sequential resilvering in the same way as healing resilvering. The one notable difference is that multiple sequential resilvers may be in progress as long as they're operating on different top-level vdevs. The `zpool wait -t resilver` command was extended to wait on sequential resilvers. From this perspective they are no different than healing resilvers. Sequential resilvers cannot be supported for RAIDZ, but are compatible with the dRAID feature being developed. As part of this change the resilver_restart_* tests were moved in to the functional/replacement directory. Additionally, the replacement tests were renamed and extended to verify both resilvering and rebuilding. Original-patch-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Tony Hutter <hutter2@llnl.gov> Reviewed-by: John Poduska <jpoduska@datto.com> Co-authored-by: Mark Maybee <mmaybee@cray.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #10349
2020-07-03 18:05:50 +00:00
/*
* The same as spa_vdev_enter() above but additionally takes the guid of
* the vdev being detached. When there is a rebuild in process it will be
* suspended while the vdev tree is modified then resumed by spa_vdev_exit().
* The rebuild is canceled if only a single child remains after the detach.
*/
uint64_t
spa_vdev_detach_enter(spa_t *spa, uint64_t guid)
{
mutex_enter(&spa->spa_vdev_top_lock);
mutex_enter(&spa_namespace_lock);
vdev_autotrim_stop_all(spa);
if (guid != 0) {
vdev_t *vd = spa_lookup_by_guid(spa, guid, B_FALSE);
if (vd) {
vdev_rebuild_stop_wait(vd->vdev_top);
}
}
return (spa_vdev_config_enter(spa));
}
/*
* Internal implementation for spa_vdev_enter(). Used when a vdev
* operation requires multiple syncs (i.e. removing a device) while
* keeping the spa_namespace_lock held.
*/
uint64_t
spa_vdev_config_enter(spa_t *spa)
{
ASSERT(MUTEX_HELD(&spa_namespace_lock));
2008-11-20 20:01:55 +00:00
spa_config_enter(spa, SCL_ALL, spa, RW_WRITER);
2008-11-20 20:01:55 +00:00
return (spa_last_synced_txg(spa) + 1);
}
/*
* Used in combination with spa_vdev_config_enter() to allow the syncing
* of multiple transactions without releasing the spa_namespace_lock.
2008-11-20 20:01:55 +00:00
*/
void
spa_vdev_config_exit(spa_t *spa, vdev_t *vd, uint64_t txg, int error, char *tag)
2008-11-20 20:01:55 +00:00
{
ASSERT(MUTEX_HELD(&spa_namespace_lock));
2008-11-20 20:01:55 +00:00
int config_changed = B_FALSE;
ASSERT(txg > spa_last_synced_txg(spa));
spa->spa_pending_vdev = NULL;
2008-11-20 20:01:55 +00:00
/*
* Reassess the DTLs.
*/
Add device rebuild feature The device_rebuild feature enables sequential reconstruction when resilvering. Mirror vdevs can be rebuilt in LBA order which may more quickly restore redundancy depending on the pools average block size, overall fragmentation and the performance characteristics of the devices. However, block checksums cannot be verified as part of the rebuild thus a scrub is automatically started after the sequential resilver completes. The new '-s' option has been added to the `zpool attach` and `zpool replace` command to request sequential reconstruction instead of healing reconstruction when resilvering. zpool attach -s <pool> <existing vdev> <new vdev> zpool replace -s <pool> <old vdev> <new vdev> The `zpool status` output has been updated to report the progress of sequential resilvering in the same way as healing resilvering. The one notable difference is that multiple sequential resilvers may be in progress as long as they're operating on different top-level vdevs. The `zpool wait -t resilver` command was extended to wait on sequential resilvers. From this perspective they are no different than healing resilvers. Sequential resilvers cannot be supported for RAIDZ, but are compatible with the dRAID feature being developed. As part of this change the resilver_restart_* tests were moved in to the functional/replacement directory. Additionally, the replacement tests were renamed and extended to verify both resilvering and rebuilding. Original-patch-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Tony Hutter <hutter2@llnl.gov> Reviewed-by: John Poduska <jpoduska@datto.com> Co-authored-by: Mark Maybee <mmaybee@cray.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #10349
2020-07-03 18:05:50 +00:00
vdev_dtl_reassess(spa->spa_root_vdev, 0, 0, B_FALSE, B_FALSE);
2008-11-20 20:01:55 +00:00
if (error == 0 && !list_is_empty(&spa->spa_config_dirty_list)) {
2008-11-20 20:01:55 +00:00
config_changed = B_TRUE;
spa->spa_config_generation++;
2008-11-20 20:01:55 +00:00
}
/*
* Verify the metaslab classes.
*/
ASSERT(metaslab_class_validate(spa_normal_class(spa)) == 0);
ASSERT(metaslab_class_validate(spa_log_class(spa)) == 0);
ASSERT(metaslab_class_validate(spa_special_class(spa)) == 0);
ASSERT(metaslab_class_validate(spa_dedup_class(spa)) == 0);
spa_config_exit(spa, SCL_ALL, spa);
2008-11-20 20:01:55 +00:00
/*
* Panic the system if the specified tag requires it. This
* is useful for ensuring that configurations are updated
* transactionally.
*/
if (zio_injection_enabled)
zio_handle_panic_injection(spa, tag, 0);
2008-11-20 20:01:55 +00:00
/*
* Note: this txg_wait_synced() is important because it ensures
* that there won't be more than one config change per txg.
* This allows us to use the txg as the generation number.
*/
if (error == 0)
txg_wait_synced(spa->spa_dsl_pool, txg);
if (vd != NULL) {
Illumos #4101, #4102, #4103, #4105, #4106 4101 metaslab_debug should allow for fine-grained control 4102 space_maps should store more information about themselves 4103 space map object blocksize should be increased 4105 removing a mirrored log device results in a leaked object 4106 asynchronously load metaslab Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Sebastien Roy <seb@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> Prior to this patch, space_maps were preferred solely based on the amount of free space left in each. Unfortunately, this heuristic didn't contain any information about the make-up of that free space, which meant we could keep preferring and loading a highly fragmented space map that wouldn't actually have enough contiguous space to satisfy the allocation; then unloading that space_map and repeating the process. This change modifies the space_map's to store additional information about the contiguous space in the space_map, so that we can use this information to make a better decision about which space_map to load. This requires reallocating all space_map objects to increase their bonus buffer size sizes enough to fit the new metadata. The above feature can be enabled via a new feature flag introduced by this change: com.delphix:spacemap_histogram In addition to the above, this patch allows the space_map block size to be increase. Currently the block size is set to be 4K in size, which has certain implications including the following: * 4K sector devices will not see any compression benefit * large space_maps require more metadata on-disk * large space_maps require more time to load (typically random reads) Now the space_map block size can adjust as needed up to the maximum size set via the space_map_max_blksz variable. A bug was fixed which resulted in potentially leaking an object when removing a mirrored log device. The previous logic for vdev_remove() did not deal with removing top-level vdevs that are interior vdevs (i.e. mirror) correctly. The problem would occur when removing a mirrored log device, and result in the DTL space map object being leaked; because top-level vdevs don't have DTL space map objects associated with them. References: https://www.illumos.org/issues/4101 https://www.illumos.org/issues/4102 https://www.illumos.org/issues/4103 https://www.illumos.org/issues/4105 https://www.illumos.org/issues/4106 https://github.com/illumos/illumos-gate/commit/0713e23 Porting notes: A handful of kmem_alloc() calls were converted to kmem_zalloc(). Also, the KM_PUSHPAGE and TQ_PUSHPAGE flags were used as necessary. Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2488
2013-10-01 21:25:53 +00:00
ASSERT(!vd->vdev_detached || vd->vdev_dtl_sm == NULL);
OpenZFS 9102 - zfs should be able to initialize storage devices PROBLEM ======== The first access to a block incurs a performance penalty on some platforms (e.g. AWS's EBS, VMware VMDKs). Therefore we recommend that volumes are "thick provisioned", where supported by the platform (VMware). This can create a large delay in getting a new virtual machines up and running (or adding storage to an existing Engine). If the thick provision step is omitted, write performance will be suboptimal until all blocks on the LUN have been written. SOLUTION ========= This feature introduces a way to 'initialize' the disks at install or in the background to make sure we don't incur this first read penalty. When an entire LUN is added to ZFS, we make all space available immediately, and allow ZFS to find unallocated space and zero it out. This works with concurrent writes to arbitrary offsets, ensuring that we don't zero out something that has been (or is in the middle of being) written. This scheme can also be applied to existing pools (affecting only free regions on the vdev). Detailed design: - new subcommand:zpool initialize [-cs] <pool> [<vdev> ...] - start, suspend, or cancel initialization - Creates new open-context thread for each vdev - Thread iterates through all metaslabs in this vdev - Each metaslab: - select a metaslab - load the metaslab - mark the metaslab as being zeroed - walk all free ranges within that metaslab and translate them to ranges on the leaf vdev - issue a "zeroing" I/O on the leaf vdev that corresponds to a free range on the metaslab we're working on - continue until all free ranges for this metaslab have been "zeroed" - reset/unmark the metaslab being zeroed - if more metaslabs exist, then repeat above tasks. - if no more metaslabs, then we're done. - progress for the initialization is stored on-disk in the vdev’s leaf zap object. The following information is stored: - the last offset that has been initialized - the state of the initialization process (i.e. active, suspended, or canceled) - the start time for the initialization - progress is reported via the zpool status command and shows information for each of the vdevs that are initializing Porting notes: - Added zfs_initialize_value module parameter to set the pattern written by "zpool initialize". - Added zfs_vdev_{initializing,removal}_{min,max}_active module options. Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: John Wren Kennedy <john.kennedy@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: loli10K <ezomori.nozomu@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Signed-off-by: Tim Chase <tim@chase2k.com> Ported-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/9102 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/c3963210eb Closes #8230
2018-12-19 14:54:59 +00:00
if (vd->vdev_ops->vdev_op_leaf) {
mutex_enter(&vd->vdev_initialize_lock);
zfs initialize performance enhancements PROBLEM ======== When invoking "zpool initialize" on a pool the command will create a thread to initialize each disk. Unfortunately, it does this serially across many transaction groups which can result in commands taking a long time to return to the user and may appear hung. The same thing is true when trying to suspend/cancel the operation. SOLUTION ========= This change refactors the way we invoke the initialize interface to ensure we can start or stop the intialization in just a few transaction groups. When stopping or cancelling a vdev initialization perform it in two phases. First signal each vdev initialization thread that it should exit, then after all threads have been signaled wait for them to exit. On a pool with 40 leaf vdevs this reduces the vdev initialize stop/cancel time from ~10 minutes to under a second. The reason for this is spa_vdev_initialize() no longer needs to wait on multiple full TXGs per leaf vdev being stopped. This commit additionally adds some missing checks for the passed "initialize_vdevs" input nvlist. The contents of the user provided input "initialize_vdevs" nvlist must be validated to ensure all values are uint64s. This is done in zfs_ioc_pool_initialize() in order to keep all of these checks in a single location. Updated the innvl and outnvl comments to match the formatting used for all other new sytle ioctls. Reviewed by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: loli10K <ezomori.nozomu@gmail.com> Reviewed-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: George Wilson <george.wilson@delphix.com> Closes #8230
2018-12-19 16:20:39 +00:00
vdev_initialize_stop(vd, VDEV_INITIALIZE_CANCELED,
NULL);
OpenZFS 9102 - zfs should be able to initialize storage devices PROBLEM ======== The first access to a block incurs a performance penalty on some platforms (e.g. AWS's EBS, VMware VMDKs). Therefore we recommend that volumes are "thick provisioned", where supported by the platform (VMware). This can create a large delay in getting a new virtual machines up and running (or adding storage to an existing Engine). If the thick provision step is omitted, write performance will be suboptimal until all blocks on the LUN have been written. SOLUTION ========= This feature introduces a way to 'initialize' the disks at install or in the background to make sure we don't incur this first read penalty. When an entire LUN is added to ZFS, we make all space available immediately, and allow ZFS to find unallocated space and zero it out. This works with concurrent writes to arbitrary offsets, ensuring that we don't zero out something that has been (or is in the middle of being) written. This scheme can also be applied to existing pools (affecting only free regions on the vdev). Detailed design: - new subcommand:zpool initialize [-cs] <pool> [<vdev> ...] - start, suspend, or cancel initialization - Creates new open-context thread for each vdev - Thread iterates through all metaslabs in this vdev - Each metaslab: - select a metaslab - load the metaslab - mark the metaslab as being zeroed - walk all free ranges within that metaslab and translate them to ranges on the leaf vdev - issue a "zeroing" I/O on the leaf vdev that corresponds to a free range on the metaslab we're working on - continue until all free ranges for this metaslab have been "zeroed" - reset/unmark the metaslab being zeroed - if more metaslabs exist, then repeat above tasks. - if no more metaslabs, then we're done. - progress for the initialization is stored on-disk in the vdev’s leaf zap object. The following information is stored: - the last offset that has been initialized - the state of the initialization process (i.e. active, suspended, or canceled) - the start time for the initialization - progress is reported via the zpool status command and shows information for each of the vdevs that are initializing Porting notes: - Added zfs_initialize_value module parameter to set the pattern written by "zpool initialize". - Added zfs_vdev_{initializing,removal}_{min,max}_active module options. Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: John Wren Kennedy <john.kennedy@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: loli10K <ezomori.nozomu@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Signed-off-by: Tim Chase <tim@chase2k.com> Ported-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/9102 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/c3963210eb Closes #8230
2018-12-19 14:54:59 +00:00
mutex_exit(&vd->vdev_initialize_lock);
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
mutex_enter(&vd->vdev_trim_lock);
vdev_trim_stop(vd, VDEV_TRIM_CANCELED, NULL);
mutex_exit(&vd->vdev_trim_lock);
OpenZFS 9102 - zfs should be able to initialize storage devices PROBLEM ======== The first access to a block incurs a performance penalty on some platforms (e.g. AWS's EBS, VMware VMDKs). Therefore we recommend that volumes are "thick provisioned", where supported by the platform (VMware). This can create a large delay in getting a new virtual machines up and running (or adding storage to an existing Engine). If the thick provision step is omitted, write performance will be suboptimal until all blocks on the LUN have been written. SOLUTION ========= This feature introduces a way to 'initialize' the disks at install or in the background to make sure we don't incur this first read penalty. When an entire LUN is added to ZFS, we make all space available immediately, and allow ZFS to find unallocated space and zero it out. This works with concurrent writes to arbitrary offsets, ensuring that we don't zero out something that has been (or is in the middle of being) written. This scheme can also be applied to existing pools (affecting only free regions on the vdev). Detailed design: - new subcommand:zpool initialize [-cs] <pool> [<vdev> ...] - start, suspend, or cancel initialization - Creates new open-context thread for each vdev - Thread iterates through all metaslabs in this vdev - Each metaslab: - select a metaslab - load the metaslab - mark the metaslab as being zeroed - walk all free ranges within that metaslab and translate them to ranges on the leaf vdev - issue a "zeroing" I/O on the leaf vdev that corresponds to a free range on the metaslab we're working on - continue until all free ranges for this metaslab have been "zeroed" - reset/unmark the metaslab being zeroed - if more metaslabs exist, then repeat above tasks. - if no more metaslabs, then we're done. - progress for the initialization is stored on-disk in the vdev’s leaf zap object. The following information is stored: - the last offset that has been initialized - the state of the initialization process (i.e. active, suspended, or canceled) - the start time for the initialization - progress is reported via the zpool status command and shows information for each of the vdevs that are initializing Porting notes: - Added zfs_initialize_value module parameter to set the pattern written by "zpool initialize". - Added zfs_vdev_{initializing,removal}_{min,max}_active module options. Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: John Wren Kennedy <john.kennedy@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: loli10K <ezomori.nozomu@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Signed-off-by: Tim Chase <tim@chase2k.com> Ported-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/9102 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/c3963210eb Closes #8230
2018-12-19 14:54:59 +00:00
}
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
/*
* The vdev may be both a leaf and top-level device.
*/
vdev_autotrim_stop_wait(vd);
2009-01-15 21:59:39 +00:00
spa_config_enter(spa, SCL_ALL, spa, RW_WRITER);
2008-11-20 20:01:55 +00:00
vdev_free(vd);
2009-01-15 21:59:39 +00:00
spa_config_exit(spa, SCL_ALL, spa);
2008-11-20 20:01:55 +00:00
}
/*
* If the config changed, update the config cache.
*/
if (config_changed)
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
spa_write_cachefile(spa, B_FALSE, B_TRUE);
}
2008-11-20 20:01:55 +00:00
/*
* Unlock the spa_t after adding or removing a vdev. Besides undoing the
* locking of spa_vdev_enter(), we also want make sure the transactions have
* synced to disk, and then update the global configuration cache with the new
* information.
*/
int
spa_vdev_exit(spa_t *spa, vdev_t *vd, uint64_t txg, int error)
{
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
vdev_autotrim_restart(spa);
Add device rebuild feature The device_rebuild feature enables sequential reconstruction when resilvering. Mirror vdevs can be rebuilt in LBA order which may more quickly restore redundancy depending on the pools average block size, overall fragmentation and the performance characteristics of the devices. However, block checksums cannot be verified as part of the rebuild thus a scrub is automatically started after the sequential resilver completes. The new '-s' option has been added to the `zpool attach` and `zpool replace` command to request sequential reconstruction instead of healing reconstruction when resilvering. zpool attach -s <pool> <existing vdev> <new vdev> zpool replace -s <pool> <old vdev> <new vdev> The `zpool status` output has been updated to report the progress of sequential resilvering in the same way as healing resilvering. The one notable difference is that multiple sequential resilvers may be in progress as long as they're operating on different top-level vdevs. The `zpool wait -t resilver` command was extended to wait on sequential resilvers. From this perspective they are no different than healing resilvers. Sequential resilvers cannot be supported for RAIDZ, but are compatible with the dRAID feature being developed. As part of this change the resilver_restart_* tests were moved in to the functional/replacement directory. Additionally, the replacement tests were renamed and extended to verify both resilvering and rebuilding. Original-patch-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Tony Hutter <hutter2@llnl.gov> Reviewed-by: John Poduska <jpoduska@datto.com> Co-authored-by: Mark Maybee <mmaybee@cray.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #10349
2020-07-03 18:05:50 +00:00
vdev_rebuild_restart(spa);
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
spa_vdev_config_exit(spa, vd, txg, error, FTAG);
2008-11-20 20:01:55 +00:00
mutex_exit(&spa_namespace_lock);
mutex_exit(&spa->spa_vdev_top_lock);
2008-11-20 20:01:55 +00:00
return (error);
}
/*
* Lock the given spa_t for the purpose of changing vdev state.
*/
void
spa_vdev_state_enter(spa_t *spa, int oplocks)
{
int locks = SCL_STATE_ALL | oplocks;
/*
* Root pools may need to read of the underlying devfs filesystem
* when opening up a vdev. Unfortunately if we're holding the
* SCL_ZIO lock it will result in a deadlock when we try to issue
* the read from the root filesystem. Instead we "prefetch"
* the associated vnodes that we need prior to opening the
* underlying devices and cache them so that we can prevent
* any I/O when we are doing the actual open.
*/
if (spa_is_root(spa)) {
int low = locks & ~(SCL_ZIO - 1);
int high = locks & ~low;
spa_config_enter(spa, high, spa, RW_WRITER);
vdev_hold(spa->spa_root_vdev);
spa_config_enter(spa, low, spa, RW_WRITER);
} else {
spa_config_enter(spa, locks, spa, RW_WRITER);
}
spa->spa_vdev_locks = locks;
}
int
spa_vdev_state_exit(spa_t *spa, vdev_t *vd, int error)
{
boolean_t config_changed = B_FALSE;
vdev_t *vdev_top;
if (vd == NULL || vd == spa->spa_root_vdev) {
vdev_top = spa->spa_root_vdev;
} else {
vdev_top = vd->vdev_top;
}
if (vd != NULL || error == 0)
Add device rebuild feature The device_rebuild feature enables sequential reconstruction when resilvering. Mirror vdevs can be rebuilt in LBA order which may more quickly restore redundancy depending on the pools average block size, overall fragmentation and the performance characteristics of the devices. However, block checksums cannot be verified as part of the rebuild thus a scrub is automatically started after the sequential resilver completes. The new '-s' option has been added to the `zpool attach` and `zpool replace` command to request sequential reconstruction instead of healing reconstruction when resilvering. zpool attach -s <pool> <existing vdev> <new vdev> zpool replace -s <pool> <old vdev> <new vdev> The `zpool status` output has been updated to report the progress of sequential resilvering in the same way as healing resilvering. The one notable difference is that multiple sequential resilvers may be in progress as long as they're operating on different top-level vdevs. The `zpool wait -t resilver` command was extended to wait on sequential resilvers. From this perspective they are no different than healing resilvers. Sequential resilvers cannot be supported for RAIDZ, but are compatible with the dRAID feature being developed. As part of this change the resilver_restart_* tests were moved in to the functional/replacement directory. Additionally, the replacement tests were renamed and extended to verify both resilvering and rebuilding. Original-patch-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Tony Hutter <hutter2@llnl.gov> Reviewed-by: John Poduska <jpoduska@datto.com> Co-authored-by: Mark Maybee <mmaybee@cray.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #10349
2020-07-03 18:05:50 +00:00
vdev_dtl_reassess(vdev_top, 0, 0, B_FALSE, B_FALSE);
if (vd != NULL) {
if (vd != spa->spa_root_vdev)
vdev_state_dirty(vdev_top);
config_changed = B_TRUE;
spa->spa_config_generation++;
}
if (spa_is_root(spa))
vdev_rele(spa->spa_root_vdev);
ASSERT3U(spa->spa_vdev_locks, >=, SCL_STATE_ALL);
spa_config_exit(spa, spa->spa_vdev_locks, spa);
2009-01-15 21:59:39 +00:00
/*
* If anything changed, wait for it to sync. This ensures that,
* from the system administrator's perspective, zpool(1M) commands
* are synchronous. This is important for things like zpool offline:
* when the command completes, you expect no further I/O from ZFS.
*/
if (vd != NULL)
txg_wait_synced(spa->spa_dsl_pool, 0);
/*
* If the config changed, update the config cache.
*/
if (config_changed) {
mutex_enter(&spa_namespace_lock);
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
spa_write_cachefile(spa, B_FALSE, B_TRUE);
mutex_exit(&spa_namespace_lock);
}
return (error);
}
2008-11-20 20:01:55 +00:00
/*
* ==========================================================================
* Miscellaneous functions
* ==========================================================================
*/
void
spa_activate_mos_feature(spa_t *spa, const char *feature, dmu_tx_t *tx)
{
if (!nvlist_exists(spa->spa_label_features, feature)) {
fnvlist_add_boolean(spa->spa_label_features, feature);
/*
* When we are creating the pool (tx_txg==TXG_INITIAL), we can't
* dirty the vdev config because lock SCL_CONFIG is not held.
* Thankfully, in this case we don't need to dirty the config
* because it will be written out anyway when we finish
* creating the pool.
*/
if (tx->tx_txg != TXG_INITIAL)
vdev_config_dirty(spa->spa_root_vdev);
}
}
void
spa_deactivate_mos_feature(spa_t *spa, const char *feature)
{
if (nvlist_remove_all(spa->spa_label_features, feature) == 0)
vdev_config_dirty(spa->spa_root_vdev);
}
2008-11-20 20:01:55 +00:00
/*
* Return the spa_t associated with given pool_guid, if it exists. If
* device_guid is non-zero, determine whether the pool exists *and* contains
* a device with the specified device_guid.
2008-11-20 20:01:55 +00:00
*/
spa_t *
spa_by_guid(uint64_t pool_guid, uint64_t device_guid)
2008-11-20 20:01:55 +00:00
{
spa_t *spa;
avl_tree_t *t = &spa_namespace_avl;
ASSERT(MUTEX_HELD(&spa_namespace_lock));
for (spa = avl_first(t); spa != NULL; spa = AVL_NEXT(t, spa)) {
if (spa->spa_state == POOL_STATE_UNINITIALIZED)
continue;
if (spa->spa_root_vdev == NULL)
continue;
if (spa_guid(spa) == pool_guid) {
if (device_guid == 0)
break;
if (vdev_lookup_by_guid(spa->spa_root_vdev,
device_guid) != NULL)
break;
/*
* Check any devices we may be in the process of adding.
*/
if (spa->spa_pending_vdev) {
if (vdev_lookup_by_guid(spa->spa_pending_vdev,
device_guid) != NULL)
break;
}
}
}
return (spa);
}
/*
* Determine whether a pool with the given pool_guid exists.
*/
boolean_t
spa_guid_exists(uint64_t pool_guid, uint64_t device_guid)
{
return (spa_by_guid(pool_guid, device_guid) != NULL);
2008-11-20 20:01:55 +00:00
}
char *
spa_strdup(const char *s)
{
size_t len;
char *new;
len = strlen(s);
new = kmem_alloc(len + 1, KM_SLEEP);
2008-11-20 20:01:55 +00:00
bcopy(s, new, len);
new[len] = '\0';
return (new);
}
void
spa_strfree(char *s)
{
kmem_free(s, strlen(s) + 1);
}
uint64_t
spa_get_random(uint64_t range)
{
uint64_t r;
ASSERT(range != 0);
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 03:20:35 +00:00
if (range == 1)
return (0);
2008-11-20 20:01:55 +00:00
(void) random_get_pseudo_bytes((void *)&r, sizeof (uint64_t));
return (r % range);
}
uint64_t
spa_generate_guid(spa_t *spa)
2008-11-20 20:01:55 +00:00
{
uint64_t guid = spa_get_random(-1ULL);
2008-11-20 20:01:55 +00:00
if (spa != NULL) {
while (guid == 0 || spa_guid_exists(spa_guid(spa), guid))
guid = spa_get_random(-1ULL);
} else {
while (guid == 0 || spa_guid_exists(guid, 0))
guid = spa_get_random(-1ULL);
2008-11-20 20:01:55 +00:00
}
return (guid);
}
void
snprintf_blkptr(char *buf, size_t buflen, const blkptr_t *bp)
{
char type[256];
char *checksum = NULL;
char *compress = NULL;
2008-11-20 20:01:55 +00:00
if (bp != NULL) {
if (BP_GET_TYPE(bp) & DMU_OT_NEWTYPE) {
dmu_object_byteswap_t bswap =
DMU_OT_BYTESWAP(BP_GET_TYPE(bp));
(void) snprintf(type, sizeof (type), "bswap %s %s",
DMU_OT_IS_METADATA(BP_GET_TYPE(bp)) ?
"metadata" : "data",
dmu_ot_byteswap[bswap].ob_name);
} else {
(void) strlcpy(type, dmu_ot[BP_GET_TYPE(bp)].ot_name,
sizeof (type));
}
if (!BP_IS_EMBEDDED(bp)) {
checksum =
zio_checksum_table[BP_GET_CHECKSUM(bp)].ci_name;
}
compress = zio_compress_table[BP_GET_COMPRESS(bp)].ci_name;
2008-11-20 20:01:55 +00:00
}
SNPRINTF_BLKPTR(snprintf, ' ', buf, buflen, bp, type, checksum,
compress);
2008-11-20 20:01:55 +00:00
}
void
spa_freeze(spa_t *spa)
{
uint64_t freeze_txg = 0;
spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
2008-11-20 20:01:55 +00:00
if (spa->spa_freeze_txg == UINT64_MAX) {
freeze_txg = spa_last_synced_txg(spa) + TXG_SIZE;
spa->spa_freeze_txg = freeze_txg;
}
spa_config_exit(spa, SCL_ALL, FTAG);
2008-11-20 20:01:55 +00:00
if (freeze_txg != 0)
txg_wait_synced(spa_get_dsl(spa), freeze_txg);
}
Swap DTRACE_PROBE* with Linux tracepoints This patch leverages Linux tracepoints from within the ZFS on Linux code base. It also refactors the debug code to bring it back in sync with Illumos. The information exported via tracepoints can be used for a variety of reasons (e.g. debugging, tuning, general exploration/understanding, etc). It is advantageous to use Linux tracepoints as the mechanism to export this kind of information (as opposed to something else) for a number of reasons: * A number of external tools can make use of our tracepoints "automatically" (e.g. perf, systemtap) * Tracepoints are designed to be extremely cheap when disabled * It's one of the "accepted" ways to export this kind of information; many other kernel subsystems use tracepoints too. Unfortunately, though, there are a few caveats as well: * Linux tracepoints appear to only be available to GPL licensed modules due to the way certain kernel functions are exported. Thus, to actually make use of the tracepoints introduced by this patch, one might have to patch and re-compile the kernel; exporting the necessary functions to non-GPL modules. * Prior to upstream kernel version v3.14-rc6-30-g66cc69e, Linux tracepoints are not available for unsigned kernel modules (tracepoints will get disabled due to the module's 'F' taint). Thus, one either has to sign the zfs kernel module prior to loading it, or use a kernel versioned v3.14-rc6-30-g66cc69e or newer. Assuming the above two requirements are satisfied, lets look at an example of how this patch can be used and what information it exposes (all commands run as 'root'): # list all zfs tracepoints available $ ls /sys/kernel/debug/tracing/events/zfs enable filter zfs_arc__delete zfs_arc__evict zfs_arc__hit zfs_arc__miss zfs_l2arc__evict zfs_l2arc__hit zfs_l2arc__iodone zfs_l2arc__miss zfs_l2arc__read zfs_l2arc__write zfs_new_state__mfu zfs_new_state__mru # enable all zfs tracepoints, clear the tracepoint ring buffer $ echo 1 > /sys/kernel/debug/tracing/events/zfs/enable $ echo 0 > /sys/kernel/debug/tracing/trace # import zpool called 'tank', inspect tracepoint data (each line was # truncated, they're too long for a commit message otherwise) $ zpool import tank $ cat /sys/kernel/debug/tracing/trace | head -n35 # tracer: nop # # entries-in-buffer/entries-written: 1219/1219 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr... z_rd_int/0-30156 [003] .... 91344.200611: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201173: zfs_arc__miss: hdr... z_rd_int/1-30157 [003] .... 91344.201756: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201795: zfs_arc__miss: hdr... z_rd_int/2-30158 [003] .... 91344.202099: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202126: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202130: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202134: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202146: zfs_arc__miss: hdr... z_rd_int/3-30159 [003] .... 91344.202457: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202484: zfs_arc__miss: hdr... z_rd_int/4-30160 [003] .... 91344.202866: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202891: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203034: zfs_arc__miss: hdr... z_rd_iss/1-30149 [001] .... 91344.203749: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.203789: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203878: zfs_arc__miss: hdr... z_rd_iss/3-30151 [001] .... 91344.204315: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.204332: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204337: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204352: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204356: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204360: zfs_arc__hit: hdr ... To highlight the kind of detailed information that is being exported using this infrastructure, I've taken the first tracepoint line from the output above and reformatted it such that it fits in 80 columns: lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr { dva 0x1:0x40082 birth 15491 cksum0 0x163edbff3a flags 0x640 datacnt 1 type 1 size 2048 spa 3133524293419867460 state_type 0 access 0 mru_hits 0 mru_ghost_hits 0 mfu_hits 0 mfu_ghost_hits 0 l2_hits 0 refcount 1 } bp { dva0 0x1:0x40082 dva1 0x1:0x3000e5 dva2 0x1:0x5a006e cksum 0x163edbff3a:0x75af30b3dd6:0x1499263ff5f2b:0x288bd118815e00 lsize 2048 } zb { objset 0 object 0 level -1 blkid 0 } For the specific tracepoint shown here, 'zfs_arc__miss', data is exported detailing the arc_buf_hdr_t (hdr), blkptr_t (bp), and zbookmark_t (zb) that caused the ARC miss (down to the exact DVA!). This kind of precise and detailed information can be extremely valuable when trying to answer certain kinds of questions. For anybody unfamiliar but looking to build on this, I found the XFS source code along with the following three web links to be extremely helpful: * http://lwn.net/Articles/379903/ * http://lwn.net/Articles/381064/ * http://lwn.net/Articles/383362/ I should also node the more "boring" aspects of this patch: * The ZFS_LINUX_COMPILE_IFELSE autoconf macro was modified to support a sixth paramter. This parameter is used to populate the contents of the new conftest.h file. If no sixth parameter is provided, conftest.h will be empty. * The ZFS_LINUX_TRY_COMPILE_HEADER autoconf macro was introduced. This macro is nearly identical to the ZFS_LINUX_TRY_COMPILE macro, except it has support for a fifth option that is then passed as the sixth parameter to ZFS_LINUX_COMPILE_IFELSE. These autoconf changes were needed to test the availability of the Linux tracepoint macros. Due to the odd nature of the Linux tracepoint macro API, a separate ".h" must be created (the path and filename is used internally by the kernel's define_trace.h file). * The HAVE_DECLARE_EVENT_CLASS autoconf macro was introduced. This is to determine if we can safely enable the Linux tracepoint functionality. We need to selectively disable the tracepoint code due to the kernel exporting certain functions as GPL only. Without this check, the build process will fail at link time. In addition, the SET_ERROR macro was modified into a tracepoint as well. To do this, the 'sdt.h' file was moved into the 'include/sys' directory and now contains a userspace portion and a kernel space portion. The dprintf and zfs_dbgmsg* interfaces are now implemented as tracepoint as well. Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-06-13 17:54:48 +00:00
void
zfs_panic_recover(const char *fmt, ...)
{
va_list adx;
va_start(adx, fmt);
vcmn_err(zfs_recover ? CE_WARN : CE_PANIC, fmt, adx);
va_end(adx);
}
/*
* This is a stripped-down version of strtoull, suitable only for converting
* lowercase hexadecimal numbers that don't overflow.
*/
uint64_t
zfs_strtonum(const char *str, char **nptr)
{
uint64_t val = 0;
char c;
int digit;
while ((c = *str) != '\0') {
if (c >= '0' && c <= '9')
digit = c - '0';
else if (c >= 'a' && c <= 'f')
digit = 10 + c - 'a';
else
break;
val *= 16;
val += digit;
str++;
}
if (nptr)
*nptr = (char *)str;
return (val);
}
void
spa_activate_allocation_classes(spa_t *spa, dmu_tx_t *tx)
{
/*
* We bump the feature refcount for each special vdev added to the pool
*/
ASSERT(spa_feature_is_enabled(spa, SPA_FEATURE_ALLOCATION_CLASSES));
spa_feature_incr(spa, SPA_FEATURE_ALLOCATION_CLASSES, tx);
}
2008-11-20 20:01:55 +00:00
/*
* ==========================================================================
* Accessor functions
* ==========================================================================
*/
boolean_t
spa_shutting_down(spa_t *spa)
2008-11-20 20:01:55 +00:00
{
return (spa->spa_async_suspended);
2008-11-20 20:01:55 +00:00
}
dsl_pool_t *
spa_get_dsl(spa_t *spa)
{
return (spa->spa_dsl_pool);
}
boolean_t
spa_is_initializing(spa_t *spa)
{
return (spa->spa_is_initializing);
}
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
boolean_t
spa_indirect_vdevs_loaded(spa_t *spa)
{
return (spa->spa_indirect_vdevs_loaded);
}
2008-11-20 20:01:55 +00:00
blkptr_t *
spa_get_rootblkptr(spa_t *spa)
{
return (&spa->spa_ubsync.ub_rootbp);
}
void
spa_set_rootblkptr(spa_t *spa, const blkptr_t *bp)
{
spa->spa_uberblock.ub_rootbp = *bp;
}
void
spa_altroot(spa_t *spa, char *buf, size_t buflen)
{
if (spa->spa_root == NULL)
buf[0] = '\0';
else
(void) strncpy(buf, spa->spa_root, buflen);
}
int
spa_sync_pass(spa_t *spa)
{
return (spa->spa_sync_pass);
}
char *
spa_name(spa_t *spa)
{
return (spa->spa_name);
}
uint64_t
spa_guid(spa_t *spa)
{
dsl_pool_t *dp = spa_get_dsl(spa);
uint64_t guid;
2008-11-20 20:01:55 +00:00
/*
* If we fail to parse the config during spa_load(), we can go through
* the error path (which posts an ereport) and end up here with no root
* vdev. We stash the original pool guid in 'spa_config_guid' to handle
2008-11-20 20:01:55 +00:00
* this case.
*/
if (spa->spa_root_vdev == NULL)
return (spa->spa_config_guid);
guid = spa->spa_last_synced_guid != 0 ?
spa->spa_last_synced_guid : spa->spa_root_vdev->vdev_guid;
/*
* Return the most recently synced out guid unless we're
* in syncing context.
*/
if (dp && dsl_pool_sync_context(dp))
2008-11-20 20:01:55 +00:00
return (spa->spa_root_vdev->vdev_guid);
else
return (guid);
}
uint64_t
spa_load_guid(spa_t *spa)
{
/*
* This is a GUID that exists solely as a reference for the
* purposes of the arc. It is generated at load time, and
* is never written to persistent storage.
*/
return (spa->spa_load_guid);
2008-11-20 20:01:55 +00:00
}
uint64_t
spa_last_synced_txg(spa_t *spa)
{
return (spa->spa_ubsync.ub_txg);
}
uint64_t
spa_first_txg(spa_t *spa)
{
return (spa->spa_first_txg);
}
uint64_t
spa_syncing_txg(spa_t *spa)
{
return (spa->spa_syncing_txg);
}
/*
* Return the last txg where data can be dirtied. The final txgs
* will be used to just clear out any deferred frees that remain.
*/
uint64_t
spa_final_dirty_txg(spa_t *spa)
{
return (spa->spa_final_txg - TXG_DEFER_SIZE);
}
pool_state_t
2008-11-20 20:01:55 +00:00
spa_state(spa_t *spa)
{
return (spa->spa_state);
}
spa_load_state_t
spa_load_state(spa_t *spa)
2008-11-20 20:01:55 +00:00
{
return (spa->spa_load_state);
2008-11-20 20:01:55 +00:00
}
uint64_t
spa_freeze_txg(spa_t *spa)
2008-11-20 20:01:55 +00:00
{
return (spa->spa_freeze_txg);
2008-11-20 20:01:55 +00:00
}
/*
* Return the inflated asize for a logical write in bytes. This is used by the
* DMU to calculate the space a logical write will require on disk.
* If lsize is smaller than the largest physical block size allocatable on this
* pool we use its value instead, since the write will end up using the whole
* block anyway.
*/
2008-11-20 20:01:55 +00:00
uint64_t
OpenZFS 7793 - ztest fails assertion in dmu_tx_willuse_space Reviewed by: Steve Gonczi <steve.gonczi@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com> Ported-by: Brian Behlendorf <behlendorf1@llnl.gov> Background information: This assertion about tx_space_* verifies that we are not dirtying more stuff than we thought we would. We “need” to know how much we will dirty so that we can check if we should fail this transaction with ENOSPC/EDQUOT, in dmu_tx_assign(). While the transaction is open (i.e. between dmu_tx_assign() and dmu_tx_commit() — typically less than a millisecond), we call dbuf_dirty() on the exact blocks that will be modified. Once this happens, the temporary accounting in tx_space_* is unnecessary, because we know exactly what blocks are newly dirtied; we call dnode_willuse_space() to track this more exact accounting. The fundamental problem causing this bug is that dmu_tx_hold_*() relies on the current state in the DMU (e.g. dn_nlevels) to predict how much will be dirtied by this transaction, but this state can change before we actually perform the transaction (i.e. call dbuf_dirty()). This bug will be fixed by removing the assertion that the tx_space_* accounting is perfectly accurate (i.e. we never dirty more than was predicted by dmu_tx_hold_*()). By removing the requirement that this accounting be perfectly accurate, we can also vastly simplify it, e.g. removing most of the logic in dmu_tx_count_*(). The new tx space accounting will be very approximate, and may be more or less than what is actually dirtied. It will still be used to determine if this transaction will put us over quota. Transactions that are marked by dmu_tx_mark_netfree() will be excepted from this check. We won’t make an attempt to determine how much space will be freed by the transaction — this was rarely accurate enough to determine if a transaction should be permitted when we are over quota, which is why dmu_tx_mark_netfree() was introduced in 2014. We also won’t attempt to give “credit” when overwriting existing blocks, if those blocks may be freed. This allows us to remove the do_free_accounting logic in dbuf_dirty(), and associated routines. This logic attempted to predict what will be on disk when this txg syncs, to know if the overwritten block will be freed (i.e. exists, and has no snapshots). OpenZFS-issue: https://www.illumos.org/issues/7793 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/3704e0a Upstream bugs: DLPX-32883a Closes #5804 Porting notes: - DNODE_SIZE replaced with DNODE_MIN_SIZE in dmu_tx_count_dnode(), Using the default dnode size would be slightly better. - DEBUG_DMU_TX wrappers and configure option removed. - Resolved _by_dnode() conflicts these changes have not yet been applied to OpenZFS.
2017-03-07 17:51:59 +00:00
spa_get_worst_case_asize(spa_t *spa, uint64_t lsize)
2008-11-20 20:01:55 +00:00
{
if (lsize == 0)
return (0); /* No inflation needed */
return (MAX(lsize, 1 << spa->spa_max_ashift) * spa_asize_inflation);
2008-11-20 20:01:55 +00:00
}
/*
* Return the amount of slop space in bytes. It is 1/32 of the pool (3.2%),
* or at least 128MB, unless that would cause it to be more than half the
* pool size.
*
* See the comment above spa_slop_shift for details.
*/
uint64_t
spa_get_slop_space(spa_t *spa)
{
uint64_t space = spa_get_dspace(spa);
return (MAX(space >> spa_slop_shift, MIN(space >> 1, spa_min_slop)));
}
2008-11-20 20:01:55 +00:00
uint64_t
spa_get_dspace(spa_t *spa)
{
return (spa->spa_dspace);
2008-11-20 20:01:55 +00:00
}
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
uint64_t
spa_get_checkpoint_space(spa_t *spa)
{
return (spa->spa_checkpoint_info.sci_dspace);
}
void
spa_update_dspace(spa_t *spa)
2008-11-20 20:01:55 +00:00
{
spa->spa_dspace = metaslab_class_get_dspace(spa_normal_class(spa)) +
ddt_get_dedup_dspace(spa);
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
if (spa->spa_vdev_removal != NULL) {
/*
* We can't allocate from the removing device, so
* subtract its size. This prevents the DMU/DSL from
* filling up the (now smaller) pool while we are in the
* middle of removing the device.
*
* Note that the DMU/DSL doesn't actually know or care
* how much space is allocated (it does its own tracking
* of how much space has been logically used). So it
* doesn't matter that the data we are moving may be
* allocated twice (on the old device and the new
* device).
*/
OpenZFS 9290 - device removal reduces redundancy of mirrors Mirrors are supposed to provide redundancy in the face of whole-disk failure and silent damage (e.g. some data on disk is not right, but ZFS hasn't detected the whole device as being broken). However, the current device removal implementation bypasses some of the mirror's redundancy. Note that in no case is incorrect data returned, but we might get a checksum error when we should have been able to find the right data. There are two underlying problems: 1. When we remove a mirror device, we only read one side of the mirror. Since we can't verify the checksum, this side may be silently bad, but the good data is on the other side of the mirror (which we didn't read). This can cause the removal to "bake in" the busted data – all copies of the data in the new location are the same, busted version, while we left the good version behind. The fix for this is to read and copy both sides of the mirror. If the old and new vdevs are mirrors, we will read both sides of the old mirror, and write each copy to the corresponding side of the new mirror. (If the old and new vdevs have a different number of children, we will do this as best as possible.) Even though we aren't verifying checksums, this ensures that as long as there's a good copy of the data, we'll have a good copy after the removal, even if there's silent damage to one side of the mirror. If we're removing a mirror that has some silent damage, we'll have exactly the same damage in the new location (assuming that the new location is also a mirror). 2. When we read from an indirect vdev that points to a mirror vdev, we only consider one copy of the data. This can lead to reduced effective redundancy, because we might read a bad copy of the data from one side of the mirror, and not retry the other, good side of the mirror. Note that the problem is not with the removal process, but rather after the removal has completed (having copied correct data to both sides of the mirror), if one side of the new mirror is silently damaged, we encounter the problem when reading the relocated data via the indirect vdev. Also note that the problem doesn't occur when ZFS knows that one side of the mirror is bad, e.g. when a disk entirely fails or is offlined. The impact is that reads (from indirect vdevs that point to mirrors) may return a checksum error even though the good data exists on one side of the mirror, and scrub doesn't repair all data on the mirror (if some of it is pointed to via an indirect vdev). The fix for this is complicated by "split blocks" - one logical block may be split into two (or more) pieces with each piece moved to a different new location. In this case we need to read all versions of each split (one from each side of the mirror), and figure out which combination of versions results in the correct checksum, and then repair the incorrect versions. This ensures that we supply the same redundancy whether you use device removal or not. For example, if a mirror has small silent errors on all of its children, we can still reconstruct the correct data, as long as those errors are at sufficiently-separated offsets (specifically, separated by the largest block size - default of 128KB, but up to 16MB). Porting notes: * A new indirect vdev check was moved from dsl_scan_needs_resilver_cb() to dsl_scan_needs_resilver(), which was added to ZoL as part of the sequential scrub work. * Passed NULL for zfs_ereport_post_checksum()'s zbookmark_phys_t parameter. The extra parameter is unique to ZoL. * When posting indirect checksum errors the ABD can be passed directly, zfs_ereport_post_checksum() is not yet ABD-aware in OpenZFS. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Ported-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9290 OpenZFS-commit: https://github.com/openzfs/openzfs/pull/591 Closes #6900
2018-02-13 19:37:56 +00:00
spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER);
vdev_t *vd =
vdev_lookup_top(spa, spa->spa_vdev_removal->svr_vdev_id);
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
spa->spa_dspace -= spa_deflate(spa) ?
vd->vdev_stat.vs_dspace : vd->vdev_stat.vs_space;
OpenZFS 9290 - device removal reduces redundancy of mirrors Mirrors are supposed to provide redundancy in the face of whole-disk failure and silent damage (e.g. some data on disk is not right, but ZFS hasn't detected the whole device as being broken). However, the current device removal implementation bypasses some of the mirror's redundancy. Note that in no case is incorrect data returned, but we might get a checksum error when we should have been able to find the right data. There are two underlying problems: 1. When we remove a mirror device, we only read one side of the mirror. Since we can't verify the checksum, this side may be silently bad, but the good data is on the other side of the mirror (which we didn't read). This can cause the removal to "bake in" the busted data – all copies of the data in the new location are the same, busted version, while we left the good version behind. The fix for this is to read and copy both sides of the mirror. If the old and new vdevs are mirrors, we will read both sides of the old mirror, and write each copy to the corresponding side of the new mirror. (If the old and new vdevs have a different number of children, we will do this as best as possible.) Even though we aren't verifying checksums, this ensures that as long as there's a good copy of the data, we'll have a good copy after the removal, even if there's silent damage to one side of the mirror. If we're removing a mirror that has some silent damage, we'll have exactly the same damage in the new location (assuming that the new location is also a mirror). 2. When we read from an indirect vdev that points to a mirror vdev, we only consider one copy of the data. This can lead to reduced effective redundancy, because we might read a bad copy of the data from one side of the mirror, and not retry the other, good side of the mirror. Note that the problem is not with the removal process, but rather after the removal has completed (having copied correct data to both sides of the mirror), if one side of the new mirror is silently damaged, we encounter the problem when reading the relocated data via the indirect vdev. Also note that the problem doesn't occur when ZFS knows that one side of the mirror is bad, e.g. when a disk entirely fails or is offlined. The impact is that reads (from indirect vdevs that point to mirrors) may return a checksum error even though the good data exists on one side of the mirror, and scrub doesn't repair all data on the mirror (if some of it is pointed to via an indirect vdev). The fix for this is complicated by "split blocks" - one logical block may be split into two (or more) pieces with each piece moved to a different new location. In this case we need to read all versions of each split (one from each side of the mirror), and figure out which combination of versions results in the correct checksum, and then repair the incorrect versions. This ensures that we supply the same redundancy whether you use device removal or not. For example, if a mirror has small silent errors on all of its children, we can still reconstruct the correct data, as long as those errors are at sufficiently-separated offsets (specifically, separated by the largest block size - default of 128KB, but up to 16MB). Porting notes: * A new indirect vdev check was moved from dsl_scan_needs_resilver_cb() to dsl_scan_needs_resilver(), which was added to ZoL as part of the sequential scrub work. * Passed NULL for zfs_ereport_post_checksum()'s zbookmark_phys_t parameter. The extra parameter is unique to ZoL. * When posting indirect checksum errors the ABD can be passed directly, zfs_ereport_post_checksum() is not yet ABD-aware in OpenZFS. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Ported-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9290 OpenZFS-commit: https://github.com/openzfs/openzfs/pull/591 Closes #6900
2018-02-13 19:37:56 +00:00
spa_config_exit(spa, SCL_VDEV, FTAG);
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
}
2008-11-20 20:01:55 +00:00
}
/*
* Return the failure mode that has been set to this pool. The default
* behavior will be to block all I/Os when a complete failure occurs.
*/
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
uint64_t
2008-11-20 20:01:55 +00:00
spa_get_failmode(spa_t *spa)
{
return (spa->spa_failmode);
}
boolean_t
spa_suspended(spa_t *spa)
{
return (spa->spa_suspended != ZIO_SUSPEND_NONE);
}
2008-11-20 20:01:55 +00:00
uint64_t
spa_version(spa_t *spa)
{
return (spa->spa_ubsync.ub_version);
}
boolean_t
spa_deflate(spa_t *spa)
{
return (spa->spa_deflate);
}
metaslab_class_t *
spa_normal_class(spa_t *spa)
{
return (spa->spa_normal_class);
}
metaslab_class_t *
spa_log_class(spa_t *spa)
{
return (spa->spa_log_class);
}
metaslab_class_t *
spa_special_class(spa_t *spa)
{
return (spa->spa_special_class);
}
metaslab_class_t *
spa_dedup_class(spa_t *spa)
{
return (spa->spa_dedup_class);
}
/*
* Locate an appropriate allocation class
*/
metaslab_class_t *
spa_preferred_class(spa_t *spa, uint64_t size, dmu_object_type_t objtype,
uint_t level, uint_t special_smallblk)
{
if (DMU_OT_IS_ZIL(objtype)) {
if (spa->spa_log_class->mc_groups != 0)
return (spa_log_class(spa));
else
return (spa_normal_class(spa));
}
boolean_t has_special_class = spa->spa_special_class->mc_groups != 0;
if (DMU_OT_IS_DDT(objtype)) {
if (spa->spa_dedup_class->mc_groups != 0)
return (spa_dedup_class(spa));
else if (has_special_class && zfs_ddt_data_is_special)
return (spa_special_class(spa));
else
return (spa_normal_class(spa));
}
/* Indirect blocks for user data can land in special if allowed */
if (level > 0 && (DMU_OT_IS_FILE(objtype) || objtype == DMU_OT_ZVOL)) {
if (has_special_class && zfs_user_indirect_is_special)
return (spa_special_class(spa));
else
return (spa_normal_class(spa));
}
if (DMU_OT_IS_METADATA(objtype) || level > 0) {
if (has_special_class)
return (spa_special_class(spa));
else
return (spa_normal_class(spa));
}
/*
* Allow small file blocks in special class in some cases (like
* for the dRAID vdev feature). But always leave a reserve of
* zfs_special_class_metadata_reserve_pct exclusively for metadata.
*/
if (DMU_OT_IS_FILE(objtype) &&
has_special_class && size <= special_smallblk) {
metaslab_class_t *special = spa_special_class(spa);
uint64_t alloc = metaslab_class_get_alloc(special);
uint64_t space = metaslab_class_get_space(special);
uint64_t limit =
(space * (100 - zfs_special_class_metadata_reserve_pct))
/ 100;
if (alloc < limit)
return (special);
}
return (spa_normal_class(spa));
}
void
spa_evicting_os_register(spa_t *spa, objset_t *os)
{
mutex_enter(&spa->spa_evicting_os_lock);
list_insert_head(&spa->spa_evicting_os_list, os);
mutex_exit(&spa->spa_evicting_os_lock);
}
void
spa_evicting_os_deregister(spa_t *spa, objset_t *os)
{
mutex_enter(&spa->spa_evicting_os_lock);
list_remove(&spa->spa_evicting_os_list, os);
cv_broadcast(&spa->spa_evicting_os_cv);
mutex_exit(&spa->spa_evicting_os_lock);
}
void
spa_evicting_os_wait(spa_t *spa)
{
mutex_enter(&spa->spa_evicting_os_lock);
while (!list_is_empty(&spa->spa_evicting_os_list))
cv_wait(&spa->spa_evicting_os_cv, &spa->spa_evicting_os_lock);
mutex_exit(&spa->spa_evicting_os_lock);
dmu_buf_user_evict_wait();
}
2008-11-20 20:01:55 +00:00
int
spa_max_replication(spa_t *spa)
{
/*
* As of SPA_VERSION == SPA_VERSION_DITTO_BLOCKS, we are able to
* handle BPs with more than one DVA allocated. Set our max
* replication level accordingly.
*/
if (spa_version(spa) < SPA_VERSION_DITTO_BLOCKS)
return (1);
return (MIN(SPA_DVAS_PER_BP, spa_max_replication_override));
}
int
spa_prev_software_version(spa_t *spa)
{
return (spa->spa_prev_software_version);
}
uint64_t
spa_deadman_synctime(spa_t *spa)
{
return (spa->spa_deadman_synctime);
}
Add TRIM support UNMAP/TRIM support is a frequently-requested feature to help prevent performance from degrading on SSDs and on various other SAN-like storage back-ends. By issuing UNMAP/TRIM commands for sectors which are no longer allocated the underlying device can often more efficiently manage itself. This TRIM implementation is modeled on the `zpool initialize` feature which writes a pattern to all unallocated space in the pool. The new `zpool trim` command uses the same vdev_xlate() code to calculate what sectors are unallocated, the same per- vdev TRIM thread model and locking, and the same basic CLI for a consistent user experience. The core difference is that instead of writing a pattern it will issue UNMAP/TRIM commands for those extents. The zio pipeline was updated to accommodate this by adding a new ZIO_TYPE_TRIM type and associated spa taskq. This new type makes is straight forward to add the platform specific TRIM/UNMAP calls to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs. This makes it possible to largely avoid changing the pipieline, one exception is that TRIM zio's may exceed the 16M block size limit since they contain no data. In addition to the manual `zpool trim` command, a background automatic TRIM was added and is controlled by the 'autotrim' property. It relies on the exact same infrastructure as the manual TRIM. However, instead of relying on the extents in a metaslab's ms_allocatable range tree, a ms_trim tree is kept per metaslab. When 'autotrim=on', ranges added back to the ms_allocatable tree are also added to the ms_free tree. The ms_free tree is then periodically consumed by an autotrim thread which systematically walks a top level vdev's metaslabs. Since the automatic TRIM will skip ranges it considers too small there is value in occasionally running a full `zpool trim`. This may occur when the freed blocks are small and not enough time was allowed to aggregate them. An automatic TRIM and a manual `zpool trim` may be run concurrently, in which case the automatic TRIM will yield to the manual TRIM. Reviewed-by: Jorgen Lundman <lundman@lundman.net> Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed-by: Matt Ahrens <mahrens@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com> Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com> Contributions-by: Tim Chase <tim@chase2k.com> Contributions-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #8419 Closes #598
2019-03-29 16:13:20 +00:00
spa_autotrim_t
spa_get_autotrim(spa_t *spa)
{
return (spa->spa_autotrim);
}
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
uint64_t
spa_deadman_ziotime(spa_t *spa)
{
return (spa->spa_deadman_ziotime);
}
uint64_t
spa_get_deadman_failmode(spa_t *spa)
{
return (spa->spa_deadman_failmode);
}
void
spa_set_deadman_failmode(spa_t *spa, const char *failmode)
{
if (strcmp(failmode, "wait") == 0)
spa->spa_deadman_failmode = ZIO_FAILURE_MODE_WAIT;
else if (strcmp(failmode, "continue") == 0)
spa->spa_deadman_failmode = ZIO_FAILURE_MODE_CONTINUE;
else if (strcmp(failmode, "panic") == 0)
spa->spa_deadman_failmode = ZIO_FAILURE_MODE_PANIC;
else
spa->spa_deadman_failmode = ZIO_FAILURE_MODE_WAIT;
}
void
spa_set_deadman_ziotime(hrtime_t ns)
{
spa_t *spa = NULL;
if (spa_mode_global != SPA_MODE_UNINIT) {
mutex_enter(&spa_namespace_lock);
while ((spa = spa_next(spa)) != NULL)
spa->spa_deadman_ziotime = ns;
mutex_exit(&spa_namespace_lock);
}
}
void
spa_set_deadman_synctime(hrtime_t ns)
{
spa_t *spa = NULL;
if (spa_mode_global != SPA_MODE_UNINIT) {
mutex_enter(&spa_namespace_lock);
while ((spa = spa_next(spa)) != NULL)
spa->spa_deadman_synctime = ns;
mutex_exit(&spa_namespace_lock);
}
}
2008-11-20 20:01:55 +00:00
uint64_t
dva_get_dsize_sync(spa_t *spa, const dva_t *dva)
2008-11-20 20:01:55 +00:00
{
uint64_t asize = DVA_GET_ASIZE(dva);
uint64_t dsize = asize;
2008-11-20 20:01:55 +00:00
ASSERT(spa_config_held(spa, SCL_ALL, RW_READER) != 0);
2008-11-20 20:01:55 +00:00
if (asize != 0 && spa->spa_deflate) {
vdev_t *vd = vdev_lookup_top(spa, DVA_GET_VDEV(dva));
if (vd != NULL)
dsize = (asize >> SPA_MINBLOCKSHIFT) *
vd->vdev_deflate_ratio;
2008-11-20 20:01:55 +00:00
}
return (dsize);
}
uint64_t
bp_get_dsize_sync(spa_t *spa, const blkptr_t *bp)
{
uint64_t dsize = 0;
for (int d = 0; d < BP_GET_NDVAS(bp); d++)
dsize += dva_get_dsize_sync(spa, &bp->blk_dva[d]);
return (dsize);
}
uint64_t
bp_get_dsize(spa_t *spa, const blkptr_t *bp)
{
uint64_t dsize = 0;
spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER);
for (int d = 0; d < BP_GET_NDVAS(bp); d++)
dsize += dva_get_dsize_sync(spa, &bp->blk_dva[d]);
spa_config_exit(spa, SCL_VDEV, FTAG);
return (dsize);
2008-11-20 20:01:55 +00:00
}
uint64_t
spa_dirty_data(spa_t *spa)
{
return (spa->spa_dsl_pool->dp_dirty_total);
}
/*
* ==========================================================================
* SPA Import Progress Routines
* ==========================================================================
*/
typedef struct spa_import_progress {
uint64_t pool_guid; /* unique id for updates */
char *pool_name;
spa_load_state_t spa_load_state;
uint64_t mmp_sec_remaining; /* MMP activity check */
uint64_t spa_load_max_txg; /* rewind txg */
procfs_list_node_t smh_node;
} spa_import_progress_t;
spa_history_list_t *spa_import_progress_list = NULL;
static int
spa_import_progress_show_header(struct seq_file *f)
{
seq_printf(f, "%-20s %-14s %-14s %-12s %s\n", "pool_guid",
"load_state", "multihost_secs", "max_txg",
"pool_name");
return (0);
}
static int
spa_import_progress_show(struct seq_file *f, void *data)
{
spa_import_progress_t *sip = (spa_import_progress_t *)data;
seq_printf(f, "%-20llu %-14llu %-14llu %-12llu %s\n",
(u_longlong_t)sip->pool_guid, (u_longlong_t)sip->spa_load_state,
(u_longlong_t)sip->mmp_sec_remaining,
(u_longlong_t)sip->spa_load_max_txg,
(sip->pool_name ? sip->pool_name : "-"));
return (0);
}
/* Remove oldest elements from list until there are no more than 'size' left */
static void
spa_import_progress_truncate(spa_history_list_t *shl, unsigned int size)
{
spa_import_progress_t *sip;
while (shl->size > size) {
sip = list_remove_head(&shl->procfs_list.pl_list);
if (sip->pool_name)
spa_strfree(sip->pool_name);
kmem_free(sip, sizeof (spa_import_progress_t));
shl->size--;
}
IMPLY(size == 0, list_is_empty(&shl->procfs_list.pl_list));
}
static void
spa_import_progress_init(void)
{
spa_import_progress_list = kmem_zalloc(sizeof (spa_history_list_t),
KM_SLEEP);
spa_import_progress_list->size = 0;
spa_import_progress_list->procfs_list.pl_private =
spa_import_progress_list;
procfs_list_install("zfs",
"import_progress",
0644,
&spa_import_progress_list->procfs_list,
spa_import_progress_show,
spa_import_progress_show_header,
NULL,
offsetof(spa_import_progress_t, smh_node));
}
static void
spa_import_progress_destroy(void)
{
spa_history_list_t *shl = spa_import_progress_list;
procfs_list_uninstall(&shl->procfs_list);
spa_import_progress_truncate(shl, 0);
procfs_list_destroy(&shl->procfs_list);
kmem_free(shl, sizeof (spa_history_list_t));
}
int
spa_import_progress_set_state(uint64_t pool_guid,
spa_load_state_t load_state)
{
spa_history_list_t *shl = spa_import_progress_list;
spa_import_progress_t *sip;
int error = ENOENT;
if (shl->size == 0)
return (0);
mutex_enter(&shl->procfs_list.pl_lock);
for (sip = list_tail(&shl->procfs_list.pl_list); sip != NULL;
sip = list_prev(&shl->procfs_list.pl_list, sip)) {
if (sip->pool_guid == pool_guid) {
sip->spa_load_state = load_state;
error = 0;
break;
}
}
mutex_exit(&shl->procfs_list.pl_lock);
return (error);
}
int
spa_import_progress_set_max_txg(uint64_t pool_guid, uint64_t load_max_txg)
{
spa_history_list_t *shl = spa_import_progress_list;
spa_import_progress_t *sip;
int error = ENOENT;
if (shl->size == 0)
return (0);
mutex_enter(&shl->procfs_list.pl_lock);
for (sip = list_tail(&shl->procfs_list.pl_list); sip != NULL;
sip = list_prev(&shl->procfs_list.pl_list, sip)) {
if (sip->pool_guid == pool_guid) {
sip->spa_load_max_txg = load_max_txg;
error = 0;
break;
}
}
mutex_exit(&shl->procfs_list.pl_lock);
return (error);
}
int
spa_import_progress_set_mmp_check(uint64_t pool_guid,
uint64_t mmp_sec_remaining)
{
spa_history_list_t *shl = spa_import_progress_list;
spa_import_progress_t *sip;
int error = ENOENT;
if (shl->size == 0)
return (0);
mutex_enter(&shl->procfs_list.pl_lock);
for (sip = list_tail(&shl->procfs_list.pl_list); sip != NULL;
sip = list_prev(&shl->procfs_list.pl_list, sip)) {
if (sip->pool_guid == pool_guid) {
sip->mmp_sec_remaining = mmp_sec_remaining;
error = 0;
break;
}
}
mutex_exit(&shl->procfs_list.pl_lock);
return (error);
}
/*
* A new import is in progress, add an entry.
*/
void
spa_import_progress_add(spa_t *spa)
{
spa_history_list_t *shl = spa_import_progress_list;
spa_import_progress_t *sip;
char *poolname = NULL;
sip = kmem_zalloc(sizeof (spa_import_progress_t), KM_SLEEP);
sip->pool_guid = spa_guid(spa);
(void) nvlist_lookup_string(spa->spa_config, ZPOOL_CONFIG_POOL_NAME,
&poolname);
if (poolname == NULL)
poolname = spa_name(spa);
sip->pool_name = spa_strdup(poolname);
sip->spa_load_state = spa_load_state(spa);
mutex_enter(&shl->procfs_list.pl_lock);
procfs_list_add(&shl->procfs_list, sip);
shl->size++;
mutex_exit(&shl->procfs_list.pl_lock);
}
void
spa_import_progress_remove(uint64_t pool_guid)
{
spa_history_list_t *shl = spa_import_progress_list;
spa_import_progress_t *sip;
mutex_enter(&shl->procfs_list.pl_lock);
for (sip = list_tail(&shl->procfs_list.pl_list); sip != NULL;
sip = list_prev(&shl->procfs_list.pl_list, sip)) {
if (sip->pool_guid == pool_guid) {
if (sip->pool_name)
spa_strfree(sip->pool_name);
list_remove(&shl->procfs_list.pl_list, sip);
shl->size--;
kmem_free(sip, sizeof (spa_import_progress_t));
break;
}
}
mutex_exit(&shl->procfs_list.pl_lock);
}
2008-11-20 20:01:55 +00:00
/*
* ==========================================================================
* Initialization and Termination
* ==========================================================================
*/
static int
spa_name_compare(const void *a1, const void *a2)
{
const spa_t *s1 = a1;
const spa_t *s2 = a2;
int s;
s = strcmp(s1->spa_name, s2->spa_name);
Performance optimization of AVL tree comparator functions perf: 2.75x faster ddt_entry_compare() First 256bits of ddt_key_t is a block checksum, which are expected to be close to random data. Hence, on average, comparison only needs to look at first few bytes of the keys. To reduce number of conditional jump instructions, the result is computed as: sign(memcmp(k1, k2)). Sign of an integer 'a' can be obtained as: `(0 < a) - (a < 0)` := {-1, 0, 1} , which is computed efficiently. Synthetic performance evaluation of original and new algorithm over 1G random keys on 2.6GHz Intel(R) Xeon(R) CPU E5-2660 v3: old 6.85789 s new 2.49089 s perf: 2.8x faster vdev_queue_offset_compare() and vdev_queue_timestamp_compare() Compute the result directly instead of using conditionals perf: zfs_range_compare() Speedup between 1.1x - 2.5x, depending on compiler version and optimization level. perf: spa_error_entry_compare() `bcmp()` is not suitable for comparator use. Use `memcmp()` instead. perf: 2.8x faster metaslab_compare() and metaslab_rangesize_compare() perf: 2.8x faster zil_bp_compare() perf: 2.8x faster mze_compare() perf: faster dbuf_compare() perf: faster compares in spa_misc perf: 2.8x faster layout_hash_compare() perf: 2.8x faster space_reftree_compare() perf: libzfs: faster avl tree comparators perf: guid_compare() perf: dsl_deadlist_compare() perf: perm_set_compare() perf: 2x faster range_tree_seg_compare() perf: faster unique_compare() perf: faster vdev_cache _compare() perf: faster vdev_uberblock_compare() perf: faster fuid _compare() perf: faster zfs_znode_hold_compare() Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Richard Elling <richard.elling@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5033
2016-08-27 18:12:53 +00:00
Reduce loaded range tree memory usage This patch implements a new tree structure for ZFS, and uses it to store range trees more efficiently. The new structure is approximately a B-tree, though there are some small differences from the usual characterizations. The tree has core nodes and leaf nodes; each contain data elements, which the elements in the core nodes acting as separators between its children. The difference between core and leaf nodes is that the core nodes have an array of children, while leaf nodes don't. Every node in the tree may be only partially full; in most cases, they are all at least 50% full (in terms of element count) except for the root node, which can be less full. Underfull nodes will steal from their neighbors or merge to remain full enough, while overfull nodes will split in two. The data elements are contained in tree-controlled buffers; they are copied into these on insertion, and overwritten on deletion. This means that the elements are not independently allocated, which reduces overhead, but also means they can't be shared between trees (and also that pointers to them are only valid until a side-effectful tree operation occurs). The overhead varies based on how dense the tree is, but is usually on the order of about 50% of the element size; the per-node overheads are very small, and so don't make a significant difference. The trees can accept arbitrary records; they accept a size and a comparator to allow them to be used for a variety of purposes. The new trees replace the AVL trees used in the range trees today. Currently, the range_seg_t structure contains three 8 byte integers of payload and two 24 byte avl_tree_node_ts to handle its storage in both an offset-sorted tree and a size-sorted tree (total size: 64 bytes). In the new model, the range seg structures are usually two 4 byte integers, but a separate one needs to exist for the size-sorted and offset-sorted tree. Between the raw size, the 50% overhead, and the double storage, the new btrees are expected to use 8*1.5*2 = 24 bytes per record, or 33.3% as much memory as the AVL trees (this is for the purposes of storing metaslab range trees; for other purposes, like scrubs, they use ~50% as much memory). We reduced the size of the payload in the range segments by teaching range trees about starting offsets and shifts; since metaslabs have a fixed starting offset, and they all operate in terms of disk sectors, we can store the ranges using 4-byte integers as long as the size of the metaslab divided by the sector size is less than 2^32. For 512-byte sectors, this is a 2^41 (or 2TB) metaslab, which with the default settings corresponds to a 256PB disk. 4k sector disks can handle metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not anticipate disks of this size in the near future, there should be almost no cases where metaslabs need 64-byte integers to store their ranges. We do still have the capability to store 64-byte integer ranges to account for cases where we are storing per-vdev (or per-dnode) trees, which could reasonably go above the limits discussed. We also do not store fill information in the compact version of the node, since it is only used for sorted scrub. We also optimized the metaslab loading process in various other ways to offset some inefficiencies in the btree model. While individual operations (find, insert, remove_from) are faster for the btree than they are for the avl tree, remove usually requires a find operation, while in the AVL tree model the element itself suffices. Some clever changes actually caused an overall speedup in metaslab loading; we use approximately 40% less cpu to load metaslabs in our tests on Illumos. Another memory and performance optimization was achieved by changing what is stored in the size-sorted trees. When a disk is heavily fragmented, the df algorithm used by default in ZFS will almost always find a number of small regions in its initial cursor-based search; it will usually only fall back to the size-sorted tree to find larger regions. If we increase the size of the cursor-based search slightly, and don't store segments that are smaller than a tunable size floor in the size-sorted tree, we can further cut memory usage down to below 20% of what the AVL trees store. This also results in further reductions in CPU time spent loading metaslabs. The 16KiB size floor was chosen because it results in substantial memory usage reduction while not usually resulting in situations where we can't find an appropriate chunk with the cursor and are forced to use an oversized chunk from the size-sorted tree. In addition, even if we do have to use an oversized chunk from the size-sorted tree, the chunk would be too small to use for ZIL allocations, so it isn't as big of a loss as it might otherwise be. And often, more small allocations will follow the initial one, and the cursor search will now find the remainder of the chunk we didn't use all of and use it for subsequent allocations. Practical testing has shown little or no change in fragmentation as a result of this change. If the size-sorted tree becomes empty while the offset sorted one still has entries, it will load all the entries from the offset sorted tree and disregard the size floor until it is unloaded again. This operation occurs rarely with the default setting, only on incredibly thoroughly fragmented pools. There are some other small changes to zdb to teach it to handle btrees, but nothing major. Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed by: Sebastien Roy seb@delphix.com Reviewed-by: Igor Kozhukhov <igor@dilos.org> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Closes #9181
2019-10-09 17:36:03 +00:00
return (TREE_ISIGN(s));
2008-11-20 20:01:55 +00:00
}
void
spa_boot_init(void)
2008-11-20 20:01:55 +00:00
{
spa_config_load();
}
void
spa_init(spa_mode_t mode)
2008-11-20 20:01:55 +00:00
{
mutex_init(&spa_namespace_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa_spare_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa_l2cache_lock, NULL, MUTEX_DEFAULT, NULL);
cv_init(&spa_namespace_cv, NULL, CV_DEFAULT, NULL);
avl_create(&spa_namespace_avl, spa_name_compare, sizeof (spa_t),
offsetof(spa_t, spa_avl));
avl_create(&spa_spare_avl, spa_spare_compare, sizeof (spa_aux_t),
offsetof(spa_aux_t, aux_avl));
avl_create(&spa_l2cache_avl, spa_l2cache_compare, sizeof (spa_aux_t),
offsetof(spa_aux_t, aux_avl));
2009-01-15 21:59:39 +00:00
spa_mode_global = mode;
2008-11-20 20:01:55 +00:00
#ifndef _KERNEL
if (spa_mode_global != SPA_MODE_READ && dprintf_find_string("watch")) {
struct sigaction sa;
sa.sa_flags = SA_SIGINFO;
sigemptyset(&sa.sa_mask);
sa.sa_sigaction = arc_buf_sigsegv;
if (sigaction(SIGSEGV, &sa, NULL) == -1) {
perror("could not enable watchpoints: "
"sigaction(SIGSEGV, ...) = ");
} else {
arc_watch = B_TRUE;
}
}
#endif
Add linux events This topic branch leverages the Solaris style FMA call points in ZFS to create a user space visible event notification system under Linux. This new system is called zevent and it unifies all previous Solaris style ereports and sysevent notifications. Under this Linux specific scheme when a sysevent or ereport event occurs an nvlist describing the event is created which looks almost exactly like a Solaris ereport. These events are queued up in the kernel when they occur and conditionally logged to the console. It is then up to a user space application to consume the events and do whatever it likes with them. To make this possible the existing /dev/zfs ABI has been extended with two new ioctls which behave as follows. * ZFS_IOC_EVENTS_NEXT Get the next pending event. The kernel will keep track of the last event consumed by the file descriptor and provide the next one if available. If no new events are available the ioctl() will block waiting for the next event. This ioctl may also be called in a non-blocking mode by setting zc.zc_guid = ZEVENT_NONBLOCK. In the non-blocking case if no events are available ENOENT will be returned. It is possible that ESHUTDOWN will be returned if the ioctl() is called while module unloading is in progress. And finally ENOMEM may occur if the provided nvlist buffer is not large enough to contain the entire event. * ZFS_IOC_EVENTS_CLEAR Clear are events queued by the kernel. The kernel will keep a fairly large number of recent events queued, use this ioctl to clear the in kernel list. This will effect all user space processes consuming events. The zpool command has been extended to use this events ABI with the 'events' subcommand. You may run 'zpool events -v' to output a verbose log of all recent events. This is very similar to the Solaris 'fmdump -ev' command with the key difference being it also includes what would be considered sysevents under Solaris. You may also run in follow mode with the '-f' option. To clear the in kernel event queue use the '-c' option. $ sudo cmd/zpool/zpool events -fv TIME CLASS May 13 2010 16:31:15.777711000 ereport.fs.zfs.config.sync class = "ereport.fs.zfs.config.sync" ena = 0x40982b7897700001 detector = (embedded nvlist) version = 0x0 scheme = "zfs" pool = 0xed976600de75dfa6 (end detector) time = 0x4bec8bc3 0x2e5aed98 pool = "zpios" pool_guid = 0xed976600de75dfa6 pool_context = 0x0 While the 'zpool events' command is handy for interactive debugging it is not expected to be the primary consumer of zevents. This ABI was primarily added to facilitate the addition of a user space monitoring daemon. This daemon would consume all events posted by the kernel and based on the type of event perform an action. For most events simply forwarding them on to syslog is likely enough. But this interface also cleanly allows for more sophisticated actions to be taken such as generating an email for a failed drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:42:43 +00:00
fm_init();
zfs_refcount_init();
2008-11-20 20:01:55 +00:00
unique_init();
Reduce loaded range tree memory usage This patch implements a new tree structure for ZFS, and uses it to store range trees more efficiently. The new structure is approximately a B-tree, though there are some small differences from the usual characterizations. The tree has core nodes and leaf nodes; each contain data elements, which the elements in the core nodes acting as separators between its children. The difference between core and leaf nodes is that the core nodes have an array of children, while leaf nodes don't. Every node in the tree may be only partially full; in most cases, they are all at least 50% full (in terms of element count) except for the root node, which can be less full. Underfull nodes will steal from their neighbors or merge to remain full enough, while overfull nodes will split in two. The data elements are contained in tree-controlled buffers; they are copied into these on insertion, and overwritten on deletion. This means that the elements are not independently allocated, which reduces overhead, but also means they can't be shared between trees (and also that pointers to them are only valid until a side-effectful tree operation occurs). The overhead varies based on how dense the tree is, but is usually on the order of about 50% of the element size; the per-node overheads are very small, and so don't make a significant difference. The trees can accept arbitrary records; they accept a size and a comparator to allow them to be used for a variety of purposes. The new trees replace the AVL trees used in the range trees today. Currently, the range_seg_t structure contains three 8 byte integers of payload and two 24 byte avl_tree_node_ts to handle its storage in both an offset-sorted tree and a size-sorted tree (total size: 64 bytes). In the new model, the range seg structures are usually two 4 byte integers, but a separate one needs to exist for the size-sorted and offset-sorted tree. Between the raw size, the 50% overhead, and the double storage, the new btrees are expected to use 8*1.5*2 = 24 bytes per record, or 33.3% as much memory as the AVL trees (this is for the purposes of storing metaslab range trees; for other purposes, like scrubs, they use ~50% as much memory). We reduced the size of the payload in the range segments by teaching range trees about starting offsets and shifts; since metaslabs have a fixed starting offset, and they all operate in terms of disk sectors, we can store the ranges using 4-byte integers as long as the size of the metaslab divided by the sector size is less than 2^32. For 512-byte sectors, this is a 2^41 (or 2TB) metaslab, which with the default settings corresponds to a 256PB disk. 4k sector disks can handle metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not anticipate disks of this size in the near future, there should be almost no cases where metaslabs need 64-byte integers to store their ranges. We do still have the capability to store 64-byte integer ranges to account for cases where we are storing per-vdev (or per-dnode) trees, which could reasonably go above the limits discussed. We also do not store fill information in the compact version of the node, since it is only used for sorted scrub. We also optimized the metaslab loading process in various other ways to offset some inefficiencies in the btree model. While individual operations (find, insert, remove_from) are faster for the btree than they are for the avl tree, remove usually requires a find operation, while in the AVL tree model the element itself suffices. Some clever changes actually caused an overall speedup in metaslab loading; we use approximately 40% less cpu to load metaslabs in our tests on Illumos. Another memory and performance optimization was achieved by changing what is stored in the size-sorted trees. When a disk is heavily fragmented, the df algorithm used by default in ZFS will almost always find a number of small regions in its initial cursor-based search; it will usually only fall back to the size-sorted tree to find larger regions. If we increase the size of the cursor-based search slightly, and don't store segments that are smaller than a tunable size floor in the size-sorted tree, we can further cut memory usage down to below 20% of what the AVL trees store. This also results in further reductions in CPU time spent loading metaslabs. The 16KiB size floor was chosen because it results in substantial memory usage reduction while not usually resulting in situations where we can't find an appropriate chunk with the cursor and are forced to use an oversized chunk from the size-sorted tree. In addition, even if we do have to use an oversized chunk from the size-sorted tree, the chunk would be too small to use for ZIL allocations, so it isn't as big of a loss as it might otherwise be. And often, more small allocations will follow the initial one, and the cursor search will now find the remainder of the chunk we didn't use all of and use it for subsequent allocations. Practical testing has shown little or no change in fragmentation as a result of this change. If the size-sorted tree becomes empty while the offset sorted one still has entries, it will load all the entries from the offset sorted tree and disregard the size floor until it is unloaded again. This operation occurs rarely with the default setting, only on incredibly thoroughly fragmented pools. There are some other small changes to zdb to teach it to handle btrees, but nothing major. Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed by: Sebastien Roy seb@delphix.com Reviewed-by: Igor Kozhukhov <igor@dilos.org> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Closes #9181
2019-10-09 17:36:03 +00:00
zfs_btree_init();
metaslab_stat_init();
ddt_init();
2008-11-20 20:01:55 +00:00
zio_init();
dmu_init();
zil_init();
vdev_cache_stat_init();
vdev_mirror_stat_init();
SIMD implementation of vdev_raidz generate and reconstruct routines This is a new implementation of RAIDZ1/2/3 routines using x86_64 scalar, SSE, and AVX2 instruction sets. Included are 3 parity generation routines (P, PQ, and PQR) and 7 reconstruction routines, for all RAIDZ level. On module load, a quick benchmark of supported routines will select the fastest for each operation and they will be used at runtime. Original implementation is still present and can be selected via module parameter. Patch contains: - specialized gen/rec routines for all RAIDZ levels, - new scalar raidz implementation (unrolled), - two x86_64 SIMD implementations (SSE and AVX2 instructions sets), - fastest routines selected on module load (benchmark). - cmd/raidz_test - verify and benchmark all implementations - added raidz_test to the ZFS Test Suite New zfs module parameters: - zfs_vdev_raidz_impl (str): selects the implementation to use. On module load, the parameter will only accept first 3 options, and the other implementations can be set once module is finished loading. Possible values for this option are: "fastest" - use the fastest math available "original" - use the original raidz code "scalar" - new scalar impl "sse" - new SSE impl if available "avx2" - new AVX2 impl if available See contents of `/sys/module/zfs/parameters/zfs_vdev_raidz_impl` to get the list of supported values. If an implementation is not supported on the system, it will not be shown. Currently selected option is enclosed in `[]`. Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4328
2016-04-25 08:04:31 +00:00
vdev_raidz_math_init();
vdev_file_init();
2008-11-20 20:01:55 +00:00
zfs_prop_init();
zpool_prop_init();
zpool_feature_init();
2008-11-20 20:01:55 +00:00
spa_config_load();
l2arc_start();
scan_init();
qat_init();
spa_import_progress_init();
2008-11-20 20:01:55 +00:00
}
void
spa_fini(void)
{
l2arc_stop();
2008-11-20 20:01:55 +00:00
spa_evict_all();
vdev_file_fini();
2008-11-20 20:01:55 +00:00
vdev_cache_stat_fini();
vdev_mirror_stat_fini();
SIMD implementation of vdev_raidz generate and reconstruct routines This is a new implementation of RAIDZ1/2/3 routines using x86_64 scalar, SSE, and AVX2 instruction sets. Included are 3 parity generation routines (P, PQ, and PQR) and 7 reconstruction routines, for all RAIDZ level. On module load, a quick benchmark of supported routines will select the fastest for each operation and they will be used at runtime. Original implementation is still present and can be selected via module parameter. Patch contains: - specialized gen/rec routines for all RAIDZ levels, - new scalar raidz implementation (unrolled), - two x86_64 SIMD implementations (SSE and AVX2 instructions sets), - fastest routines selected on module load (benchmark). - cmd/raidz_test - verify and benchmark all implementations - added raidz_test to the ZFS Test Suite New zfs module parameters: - zfs_vdev_raidz_impl (str): selects the implementation to use. On module load, the parameter will only accept first 3 options, and the other implementations can be set once module is finished loading. Possible values for this option are: "fastest" - use the fastest math available "original" - use the original raidz code "scalar" - new scalar impl "sse" - new SSE impl if available "avx2" - new AVX2 impl if available See contents of `/sys/module/zfs/parameters/zfs_vdev_raidz_impl` to get the list of supported values. If an implementation is not supported on the system, it will not be shown. Currently selected option is enclosed in `[]`. Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4328
2016-04-25 08:04:31 +00:00
vdev_raidz_math_fini();
2008-11-20 20:01:55 +00:00
zil_fini();
dmu_fini();
zio_fini();
ddt_fini();
Reduce loaded range tree memory usage This patch implements a new tree structure for ZFS, and uses it to store range trees more efficiently. The new structure is approximately a B-tree, though there are some small differences from the usual characterizations. The tree has core nodes and leaf nodes; each contain data elements, which the elements in the core nodes acting as separators between its children. The difference between core and leaf nodes is that the core nodes have an array of children, while leaf nodes don't. Every node in the tree may be only partially full; in most cases, they are all at least 50% full (in terms of element count) except for the root node, which can be less full. Underfull nodes will steal from their neighbors or merge to remain full enough, while overfull nodes will split in two. The data elements are contained in tree-controlled buffers; they are copied into these on insertion, and overwritten on deletion. This means that the elements are not independently allocated, which reduces overhead, but also means they can't be shared between trees (and also that pointers to them are only valid until a side-effectful tree operation occurs). The overhead varies based on how dense the tree is, but is usually on the order of about 50% of the element size; the per-node overheads are very small, and so don't make a significant difference. The trees can accept arbitrary records; they accept a size and a comparator to allow them to be used for a variety of purposes. The new trees replace the AVL trees used in the range trees today. Currently, the range_seg_t structure contains three 8 byte integers of payload and two 24 byte avl_tree_node_ts to handle its storage in both an offset-sorted tree and a size-sorted tree (total size: 64 bytes). In the new model, the range seg structures are usually two 4 byte integers, but a separate one needs to exist for the size-sorted and offset-sorted tree. Between the raw size, the 50% overhead, and the double storage, the new btrees are expected to use 8*1.5*2 = 24 bytes per record, or 33.3% as much memory as the AVL trees (this is for the purposes of storing metaslab range trees; for other purposes, like scrubs, they use ~50% as much memory). We reduced the size of the payload in the range segments by teaching range trees about starting offsets and shifts; since metaslabs have a fixed starting offset, and they all operate in terms of disk sectors, we can store the ranges using 4-byte integers as long as the size of the metaslab divided by the sector size is less than 2^32. For 512-byte sectors, this is a 2^41 (or 2TB) metaslab, which with the default settings corresponds to a 256PB disk. 4k sector disks can handle metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not anticipate disks of this size in the near future, there should be almost no cases where metaslabs need 64-byte integers to store their ranges. We do still have the capability to store 64-byte integer ranges to account for cases where we are storing per-vdev (or per-dnode) trees, which could reasonably go above the limits discussed. We also do not store fill information in the compact version of the node, since it is only used for sorted scrub. We also optimized the metaslab loading process in various other ways to offset some inefficiencies in the btree model. While individual operations (find, insert, remove_from) are faster for the btree than they are for the avl tree, remove usually requires a find operation, while in the AVL tree model the element itself suffices. Some clever changes actually caused an overall speedup in metaslab loading; we use approximately 40% less cpu to load metaslabs in our tests on Illumos. Another memory and performance optimization was achieved by changing what is stored in the size-sorted trees. When a disk is heavily fragmented, the df algorithm used by default in ZFS will almost always find a number of small regions in its initial cursor-based search; it will usually only fall back to the size-sorted tree to find larger regions. If we increase the size of the cursor-based search slightly, and don't store segments that are smaller than a tunable size floor in the size-sorted tree, we can further cut memory usage down to below 20% of what the AVL trees store. This also results in further reductions in CPU time spent loading metaslabs. The 16KiB size floor was chosen because it results in substantial memory usage reduction while not usually resulting in situations where we can't find an appropriate chunk with the cursor and are forced to use an oversized chunk from the size-sorted tree. In addition, even if we do have to use an oversized chunk from the size-sorted tree, the chunk would be too small to use for ZIL allocations, so it isn't as big of a loss as it might otherwise be. And often, more small allocations will follow the initial one, and the cursor search will now find the remainder of the chunk we didn't use all of and use it for subsequent allocations. Practical testing has shown little or no change in fragmentation as a result of this change. If the size-sorted tree becomes empty while the offset sorted one still has entries, it will load all the entries from the offset sorted tree and disregard the size floor until it is unloaded again. This operation occurs rarely with the default setting, only on incredibly thoroughly fragmented pools. There are some other small changes to zdb to teach it to handle btrees, but nothing major. Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed by: Sebastien Roy seb@delphix.com Reviewed-by: Igor Kozhukhov <igor@dilos.org> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Paul Dagnelie <pcd@delphix.com> Closes #9181
2019-10-09 17:36:03 +00:00
metaslab_stat_fini();
zfs_btree_fini();
2008-11-20 20:01:55 +00:00
unique_fini();
zfs_refcount_fini();
Add linux events This topic branch leverages the Solaris style FMA call points in ZFS to create a user space visible event notification system under Linux. This new system is called zevent and it unifies all previous Solaris style ereports and sysevent notifications. Under this Linux specific scheme when a sysevent or ereport event occurs an nvlist describing the event is created which looks almost exactly like a Solaris ereport. These events are queued up in the kernel when they occur and conditionally logged to the console. It is then up to a user space application to consume the events and do whatever it likes with them. To make this possible the existing /dev/zfs ABI has been extended with two new ioctls which behave as follows. * ZFS_IOC_EVENTS_NEXT Get the next pending event. The kernel will keep track of the last event consumed by the file descriptor and provide the next one if available. If no new events are available the ioctl() will block waiting for the next event. This ioctl may also be called in a non-blocking mode by setting zc.zc_guid = ZEVENT_NONBLOCK. In the non-blocking case if no events are available ENOENT will be returned. It is possible that ESHUTDOWN will be returned if the ioctl() is called while module unloading is in progress. And finally ENOMEM may occur if the provided nvlist buffer is not large enough to contain the entire event. * ZFS_IOC_EVENTS_CLEAR Clear are events queued by the kernel. The kernel will keep a fairly large number of recent events queued, use this ioctl to clear the in kernel list. This will effect all user space processes consuming events. The zpool command has been extended to use this events ABI with the 'events' subcommand. You may run 'zpool events -v' to output a verbose log of all recent events. This is very similar to the Solaris 'fmdump -ev' command with the key difference being it also includes what would be considered sysevents under Solaris. You may also run in follow mode with the '-f' option. To clear the in kernel event queue use the '-c' option. $ sudo cmd/zpool/zpool events -fv TIME CLASS May 13 2010 16:31:15.777711000 ereport.fs.zfs.config.sync class = "ereport.fs.zfs.config.sync" ena = 0x40982b7897700001 detector = (embedded nvlist) version = 0x0 scheme = "zfs" pool = 0xed976600de75dfa6 (end detector) time = 0x4bec8bc3 0x2e5aed98 pool = "zpios" pool_guid = 0xed976600de75dfa6 pool_context = 0x0 While the 'zpool events' command is handy for interactive debugging it is not expected to be the primary consumer of zevents. This ABI was primarily added to facilitate the addition of a user space monitoring daemon. This daemon would consume all events posted by the kernel and based on the type of event perform an action. For most events simply forwarding them on to syslog is likely enough. But this interface also cleanly allows for more sophisticated actions to be taken such as generating an email for a failed drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:42:43 +00:00
fm_fini();
scan_fini();
qat_fini();
spa_import_progress_destroy();
2008-11-20 20:01:55 +00:00
avl_destroy(&spa_namespace_avl);
avl_destroy(&spa_spare_avl);
avl_destroy(&spa_l2cache_avl);
cv_destroy(&spa_namespace_cv);
mutex_destroy(&spa_namespace_lock);
mutex_destroy(&spa_spare_lock);
mutex_destroy(&spa_l2cache_lock);
}
/*
* Return whether this pool has slogs. No locking needed.
* It's not a problem if the wrong answer is returned as it's only for
* performance and not correctness
*/
boolean_t
spa_has_slogs(spa_t *spa)
{
return (spa->spa_log_class->mc_rotor != NULL);
}
spa_log_state_t
spa_get_log_state(spa_t *spa)
{
return (spa->spa_log_state);
}
void
spa_set_log_state(spa_t *spa, spa_log_state_t state)
{
spa->spa_log_state = state;
}
boolean_t
spa_is_root(spa_t *spa)
{
return (spa->spa_is_root);
}
2009-01-15 21:59:39 +00:00
boolean_t
spa_writeable(spa_t *spa)
{
return (!!(spa->spa_mode & SPA_MODE_WRITE) && spa->spa_trust_config);
2009-01-15 21:59:39 +00:00
}
/*
* Returns true if there is a pending sync task in any of the current
* syncing txg, the current quiescing txg, or the current open txg.
*/
boolean_t
spa_has_pending_synctask(spa_t *spa)
{
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
return (!txg_all_lists_empty(&spa->spa_dsl_pool->dp_sync_tasks) ||
!txg_all_lists_empty(&spa->spa_dsl_pool->dp_early_sync_tasks));
}
spa_mode_t
2009-01-15 21:59:39 +00:00
spa_mode(spa_t *spa)
{
return (spa->spa_mode);
}
uint64_t
spa_bootfs(spa_t *spa)
{
return (spa->spa_bootfs);
}
uint64_t
spa_delegation(spa_t *spa)
{
return (spa->spa_delegation);
}
objset_t *
spa_meta_objset(spa_t *spa)
{
return (spa->spa_meta_objset);
}
enum zio_checksum
spa_dedup_checksum(spa_t *spa)
{
return (spa->spa_dedup_checksum);
}
/*
* Reset pool scan stat per scan pass (or reboot).
*/
void
spa_scan_stat_init(spa_t *spa)
{
/* data not stored on disk */
spa->spa_scan_pass_start = gethrestime_sec();
if (dsl_scan_is_paused_scrub(spa->spa_dsl_pool->dp_scan))
spa->spa_scan_pass_scrub_pause = spa->spa_scan_pass_start;
else
spa->spa_scan_pass_scrub_pause = 0;
spa->spa_scan_pass_scrub_spent_paused = 0;
spa->spa_scan_pass_exam = 0;
spa->spa_scan_pass_issued = 0;
vdev_scan_stat_init(spa->spa_root_vdev);
}
/*
* Get scan stats for zpool status reports
*/
int
spa_scan_get_stats(spa_t *spa, pool_scan_stat_t *ps)
{
dsl_scan_t *scn = spa->spa_dsl_pool ? spa->spa_dsl_pool->dp_scan : NULL;
if (scn == NULL || scn->scn_phys.scn_func == POOL_SCAN_NONE)
return (SET_ERROR(ENOENT));
bzero(ps, sizeof (pool_scan_stat_t));
/* data stored on disk */
ps->pss_func = scn->scn_phys.scn_func;
ps->pss_state = scn->scn_phys.scn_state;
ps->pss_start_time = scn->scn_phys.scn_start_time;
ps->pss_end_time = scn->scn_phys.scn_end_time;
ps->pss_to_examine = scn->scn_phys.scn_to_examine;
ps->pss_examined = scn->scn_phys.scn_examined;
ps->pss_to_process = scn->scn_phys.scn_to_process;
ps->pss_processed = scn->scn_phys.scn_processed;
ps->pss_errors = scn->scn_phys.scn_errors;
/* data not stored on disk */
ps->pss_pass_exam = spa->spa_scan_pass_exam;
ps->pss_pass_start = spa->spa_scan_pass_start;
ps->pss_pass_scrub_pause = spa->spa_scan_pass_scrub_pause;
ps->pss_pass_scrub_spent_paused = spa->spa_scan_pass_scrub_spent_paused;
ps->pss_pass_issued = spa->spa_scan_pass_issued;
ps->pss_issued =
scn->scn_issued_before_pass + spa->spa_scan_pass_issued;
return (0);
}
Illumos 5027 - zfs large block support 5027 zfs large block support Reviewed by: Alek Pinchuk <pinchuk.alek@gmail.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com> Reviewed by: Richard Elling <richard.elling@richardelling.com> Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/5027 https://github.com/illumos/illumos-gate/commit/b515258 Porting Notes: * Included in this patch is a tiny ISP2() cleanup in zio_init() from Illumos 5255. * Unlike the upstream Illumos commit this patch does not impose an arbitrary 128K block size limit on volumes. Volumes, like filesystems, are limited by the zfs_max_recordsize=1M module option. * By default the maximum record size is limited to 1M by the module option zfs_max_recordsize. This value may be safely increased up to 16M which is the largest block size supported by the on-disk format. At the moment, 1M blocks clearly offer a significant performance improvement but the benefits of going beyond this for the majority of workloads are less clear. * The illumos version of this patch increased DMU_MAX_ACCESS to 32M. This was determined not to be large enough when using 16M blocks because the zfs_make_xattrdir() function will fail (EFBIG) when assigning a TX. This was immediately observed under Linux because all newly created files must have a security xattr created and that was failing. Therefore, we've set DMU_MAX_ACCESS to 64M. * On 32-bit platforms a hard limit of 1M is set for blocks due to the limited virtual address space. We should be able to relax this one the ABD patches are merged. Ported-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #354
2014-11-03 20:15:08 +00:00
int
spa_maxblocksize(spa_t *spa)
{
if (spa_feature_is_enabled(spa, SPA_FEATURE_LARGE_BLOCKS))
return (SPA_MAXBLOCKSIZE);
else
return (SPA_OLD_MAXBLOCKSIZE);
}
OpenZFS 7614, 9064 - zfs device evacuation/removal OpenZFS 7614 - zfs device evacuation/removal OpenZFS 9064 - remove_mirror should wait for device removal to complete This project allows top-level vdevs to be removed from the storage pool with "zpool remove", reducing the total amount of storage in the pool. This operation copies all allocated regions of the device to be removed onto other devices, recording the mapping from old to new location. After the removal is complete, read and free operations to the removed (now "indirect") vdev must be remapped and performed at the new location on disk. The indirect mapping table is kept in memory whenever the pool is loaded, so there is minimal performance overhead when doing operations on the indirect vdev. The size of the in-memory mapping table will be reduced when its entries become "obsolete" because they are no longer used by any block pointers in the pool. An entry becomes obsolete when all the blocks that use it are freed. An entry can also become obsolete when all the snapshots that reference it are deleted, and the block pointers that reference it have been "remapped" in all filesystems/zvols (and clones). Whenever an indirect block is written, all the block pointers in it will be "remapped" to their new (concrete) locations if possible. This process can be accelerated by using the "zfs remap" command to proactively rewrite all indirect blocks that reference indirect (removed) vdevs. Note that when a device is removed, we do not verify the checksum of the data that is copied. This makes the process much faster, but if it were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be possible to copy the wrong data, when we have the correct data on e.g. the other side of the mirror. At the moment, only mirrors and simple top-level vdevs can be removed and no removal is allowed if any of the top-level vdevs are raidz. Porting Notes: * Avoid zero-sized kmem_alloc() in vdev_compact_children(). The device evacuation code adds a dependency that vdev_compact_children() be able to properly empty the vdev_child array by setting it to NULL and zeroing vdev_children. Under Linux, kmem_alloc() and related functions return a sentinel pointer rather than NULL for zero-sized allocations. * Remove comment regarding "mpt" driver where zfs_remove_max_segment is initialized to SPA_MAXBLOCKSIZE. Change zfs_condense_indirect_commit_entry_delay_ticks to zfs_condense_indirect_commit_entry_delay_ms for consistency with most other tunables in which delays are specified in ms. * ZTS changes: Use set_tunable rather than mdb Use zpool sync as appropriate Use sync_pool instead of sync Kill jobs during test_removal_with_operation to allow unmount/export Don't add non-disk names such as "mirror" or "raidz" to $DISKS Use $TEST_BASE_DIR instead of /tmp Increase HZ from 100 to 1000 which is more common on Linux removal_multiple_indirection.ksh Reduce iterations in order to not time out on the code coverage builders. removal_resume_export: Functionally, the test case is correct but there exists a race where the kernel thread hasn't been fully started yet and is not visible. Wait for up to 1 second for the removal thread to be started before giving up on it. Also, increase the amount of data copied in order that the removal not finish before the export has a chance to fail. * MMP compatibility, the concept of concrete versus non-concrete devices has slightly changed the semantics of vdev_writeable(). Update mmp_random_leaf_impl() accordingly. * Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool feature which is not supported by OpenZFS. * Added support for new vdev removal tracepoints. * Test cases removal_with_zdb and removal_condense_export have been intentionally disabled. When run manually they pass as intended, but when running in the automated test environment they produce unreliable results on the latest Fedora release. They may work better once the upstream pool import refectoring is merged into ZoL at which point they will be re-enabled. Authored by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Alex Reece <alex@delphix.com> Reviewed-by: George Wilson <george.wilson@delphix.com> Reviewed-by: John Kennedy <john.kennedy@delphix.com> Reviewed-by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Richard Laager <rlaager@wiktel.com> Reviewed by: Tim Chase <tim@chase2k.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Garrett D'Amore <garrett@damore.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://www.illumos.org/issues/7614 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1eb Closes #6900
2016-09-22 16:30:13 +00:00
/*
* Returns the txg that the last device removal completed. No indirect mappings
* have been added since this txg.
*/
uint64_t
spa_get_last_removal_txg(spa_t *spa)
{
uint64_t vdevid;
uint64_t ret = -1ULL;
spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER);
/*
* sr_prev_indirect_vdev is only modified while holding all the
* config locks, so it is sufficient to hold SCL_VDEV as reader when
* examining it.
*/
vdevid = spa->spa_removing_phys.sr_prev_indirect_vdev;
while (vdevid != -1ULL) {
vdev_t *vd = vdev_lookup_top(spa, vdevid);
vdev_indirect_births_t *vib = vd->vdev_indirect_births;
ASSERT3P(vd->vdev_ops, ==, &vdev_indirect_ops);
/*
* If the removal did not remap any data, we don't care.
*/
if (vdev_indirect_births_count(vib) != 0) {
ret = vdev_indirect_births_last_entry_txg(vib);
break;
}
vdevid = vd->vdev_indirect_config.vic_prev_indirect_vdev;
}
spa_config_exit(spa, SCL_VDEV, FTAG);
IMPLY(ret != -1ULL,
spa_feature_is_active(spa, SPA_FEATURE_DEVICE_REMOVAL));
return (ret);
}
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
int
spa_maxdnodesize(spa_t *spa)
{
if (spa_feature_is_enabled(spa, SPA_FEATURE_LARGE_DNODE))
return (DNODE_MAX_SIZE);
else
return (DNODE_MIN_SIZE);
}
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 03:20:35 +00:00
boolean_t
spa_multihost(spa_t *spa)
{
return (spa->spa_multihost ? B_TRUE : B_FALSE);
}
uint32_t
spa_get_hostid(spa_t *spa)
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 03:20:35 +00:00
{
return (spa->spa_hostid);
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 03:20:35 +00:00
}
OpenZFS 9075 - Improve ZFS pool import/load process and corrupted pool recovery Some work has been done lately to improve the debugability of the ZFS pool load (and import) process. This includes: 7638 Refactor spa_load_impl into several functions 8961 SPA load/import should tell us why it failed 7277 zdb should be able to print zfs_dbgmsg's To iterate on top of that, there's a few changes that were made to make the import process more resilient and crash free. One of the first tasks during the pool load process is to parse a config provided from userland that describes what devices the pool is composed of. A vdev tree is generated from that config, and then all the vdevs are opened. The Meta Object Set (MOS) of the pool is accessed, and several metadata objects that are necessary to load the pool are read. The exact configuration of the pool is also stored inside the MOS. Since the configuration provided from userland is external and might not accurately describe the vdev tree of the pool at the txg that is being loaded, it cannot be relied upon to safely operate the pool. For that reason, the configuration in the MOS is read early on. In the past, the two configurations were compared together and if there was a mismatch then the load process was aborted and an error was returned. The latter was a good way to ensure a pool does not get corrupted, however it made the pool load process needlessly fragile in cases where the vdev configuration changed or the userland configuration was outdated. Since the MOS is stored in 3 copies, the configuration provided by userland doesn't have to be perfect in order to read its contents. Hence, a new approach has been adopted: The pool is first opened with the untrusted userland configuration just so that the real configuration can be read from the MOS. The trusted MOS configuration is then used to generate a new vdev tree and the pool is re-opened. When the pool is opened with an untrusted configuration, writes are disabled to avoid accidentally damaging it. During reads, some sanity checks are performed on block pointers to see if each DVA points to a known vdev; when the configuration is untrusted, instead of panicking the system if those checks fail we simply avoid issuing reads to the invalid DVAs. This new two-step pool load process now allows rewinding pools accross vdev tree changes such as device replacement, addition, etc. Loading a pool from an external config file in a clustering environment also becomes much safer now since the pool will import even if the config is outdated and didn't, for instance, register a recent device addition. With this code in place, it became relatively easy to implement a long-sought-after feature: the ability to import a pool with missing top level (i.e. non-redundant) devices. Note that since this almost guarantees some loss of data, this feature is for now restricted to a read-only import. Porting notes (ZTS): * Fix 'make dist' target in zpool_import * The maximum path length allowed by tar is 99 characters. Several of the new test cases exceeded this limit resulting in them not being included in the tarball. Shorten the names slightly. * Set/get tunables using accessor functions. * Get last synced txg via the "zfs_txg_history" mechanism. * Clear zinject handlers in cleanup for import_cache_device_replaced and import_rewind_device_replaced in order that the zpool can be exported if there is an error. * Increase FILESIZE to 8G in zfs-test.sh to allow for a larger ext4 file system to be created on ZFS_DISK2. Also, there's no need to partition ZFS_DISK2 at all. The partitioning had already been disabled for multipath devices. Among other things, the partitioning steals some space from the ext4 file system, makes it difficult to accurately calculate the paramters to parted and can make some of the tests fail. * Increase FS_SIZE and FILE_SIZE in the zpool_import test configuration now that FILESIZE is larger. * Write more data in order that device evacuation take lonnger in a couple tests. * Use mkdir -p to avoid errors when the directory already exists. * Remove use of sudo in import_rewind_config_changed. Authored by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Andrew Stormont <andyjstormont@gmail.com> Approved by: Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9075 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/619c0123 Closes #7459
2016-07-22 14:39:36 +00:00
boolean_t
spa_trust_config(spa_t *spa)
{
return (spa->spa_trust_config);
}
uint64_t
spa_missing_tvds_allowed(spa_t *spa)
{
return (spa->spa_missing_tvds_allowed);
}
Log Spacemap Project = Motivation At Delphix we've seen a lot of customer systems where fragmentation is over 75% and random writes take a performance hit because a lot of time is spend on I/Os that update on-disk space accounting metadata. Specifically, we seen cases where 20% to 40% of sync time is spend after sync pass 1 and ~30% of the I/Os on the system is spent updating spacemaps. The problem is that these pools have existed long enough that we've touched almost every metaslab at least once, and random writes scatter frees across all metaslabs every TXG, thus appending to their spacemaps and resulting in many I/Os. To give an example, assuming that every VDEV has 200 metaslabs and our writes fit within a single spacemap block (generally 4K) we have 200 I/Os. Then if we assume 2 levels of indirection, we need 400 additional I/Os and since we are talking about metadata for which we keep 2 extra copies for redundancy we need to triple that number, leading to a total of 1800 I/Os per VDEV every TXG. We could try and decrease the number of metaslabs so we have less I/Os per TXG but then each metaslab would cover a wider range on disk and thus would take more time to be loaded in memory from disk. In addition, after it's loaded, it's range tree would consume more memory. Another idea would be to just increase the spacemap block size which would allow us to fit more entries within an I/O block resulting in fewer I/Os per metaslab and a speedup in loading time. The problem is still that we don't deal with the number of I/Os going up as the number of metaslabs is increasing and the fact is that we generally write a lot to a few metaslabs and a little to the rest of them. Thus, just increasing the block size would actually waste bandwidth because we won't be utilizing our bigger block size. = About this patch This patch introduces the Log Spacemap project which provides the solution to the above problem while taking into account all the aforementioned tradeoffs. The details on how it achieves that can be found in the references sections below and in the code (see Big Theory Statement in spa_log_spacemap.c). Even though the change is fairly constraint within the metaslab and lower-level SPA codepaths, there is a side-change that is user-facing. The change is that VDEV IDs from VDEV holes will no longer be reused. To give some background and reasoning for this, when a log device is removed and its VDEV structure was replaced with a hole (or was compacted; if at the end of the vdev array), its vdev_id could be reused by devices added after that. Now with the pool-wide space maps recording the vdev ID, this behavior can cause problems (e.g. is this entry referring to a segment in the new vdev or the removed log?). Thus, to simplify things the ID reuse behavior is gone and now vdev IDs for top-level vdevs are truly unique within a pool. = Testing The illumos implementation of this feature has been used internally for a year and has been in production for ~6 months. For this patch specifically there don't seem to be any regressions introduced to ZTS and I have been running zloop for a week without any related problems. = Performance Analysis (Linux Specific) All performance results and analysis for illumos can be found in the links of the references. Redoing the same experiments in Linux gave similar results. Below are the specifics of the Linux run. After the pool reached stable state the percentage of the time spent in pass 1 per TXG was 64% on average for the stock bits while the log spacemap bits stayed at 95% during the experiment (graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png). Sync times per TXG were 37.6 seconds on average for the stock bits and 22.7 seconds for the log spacemap bits (related graph: sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result the log spacemap bits were able to push more TXGs, which is also the reason why all graphs quantified per TXG have more entries for the log spacemap bits. Another interesting aspect in terms of txg syncs is that the stock bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8, and 20% reach 9. The log space map bits reached sync pass 4 in 79% of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This emphasizes the fact that not only we spend less time on metadata but we also iterate less times to convergence in spa_sync() dirtying objects. [related graphs: stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png] Finally, the improvement in IOPs that the userland gains from the change is approximately 40%. There is a consistent win in IOPS as you can see from the graphs below but the absolute amount of improvement that the log spacemap gives varies within each minute interval. sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png = Porting to Other Platforms For people that want to port this commit to other platforms below is a list of ZoL commits that this patch depends on: Make zdb results for checkpoint tests consistent db587941c5ff6dea01932bb78f70db63cf7f38ba Update vdev_is_spacemap_addressable() for new spacemap encoding 419ba5914552c6185afbe1dd17b3ed4b0d526547 Simplify spa_sync by breaking it up to smaller functions 8dc2197b7b1e4d7ebc1420ea30e51c6541f1d834 Factor metaslab_load_wait() in metaslab_load() b194fab0fb6caad18711abccaff3c69ad8b3f6d3 Rename range_tree_verify to range_tree_verify_not_present df72b8bebe0ebac0b20e0750984bad182cb6564a Change target size of metaslabs from 256GB to 16GB c853f382db731e15a87512f4ef1101d14d778a55 zdb -L should skip leak detection altogether 21e7cf5da89f55ce98ec1115726b150e19eefe89 vs_alloc can underflow in L2ARC vdevs 7558997d2f808368867ca7e5234e5793446e8f3f Simplify log vdev removal code 6c926f426a26ffb6d7d8e563e33fc176164175cb Get rid of space_map_update() for ms_synced_length 425d3237ee88abc53d8522a7139c926d278b4b7f Introduce auxiliary metaslab histograms 928e8ad47d3478a3d5d01f0dd6ae74a9371af65e Error path in metaslab_load_impl() forgets to drop ms_sync_lock 8eef997679ba54547f7d361553d21b3291f41ae7 = References Background, Motivation, and Internals of the Feature - OpenZFS 2017 Presentation: youtu.be/jj2IxRkl5bQ - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project Flushing Algorithm Internals & Performance Results (Illumos Specific) - Blogpost: sdimitro.github.io/post/zfs-lsm-flushing/ - OpenZFS 2018 Presentation: youtu.be/x6D2dHRjkxw - Slides: slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm Upstream Delphix Issues: DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320 DLPX-63385 Reviewed-by: Sean Eric Fagan <sef@ixsystems.com> Reviewed-by: Matt Ahrens <matt@delphix.com> Reviewed-by: George Wilson <gwilson@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com> Closes #8442
2019-07-16 17:11:49 +00:00
space_map_t *
spa_syncing_log_sm(spa_t *spa)
{
return (spa->spa_syncing_log_sm);
}
OpenZFS 9075 - Improve ZFS pool import/load process and corrupted pool recovery Some work has been done lately to improve the debugability of the ZFS pool load (and import) process. This includes: 7638 Refactor spa_load_impl into several functions 8961 SPA load/import should tell us why it failed 7277 zdb should be able to print zfs_dbgmsg's To iterate on top of that, there's a few changes that were made to make the import process more resilient and crash free. One of the first tasks during the pool load process is to parse a config provided from userland that describes what devices the pool is composed of. A vdev tree is generated from that config, and then all the vdevs are opened. The Meta Object Set (MOS) of the pool is accessed, and several metadata objects that are necessary to load the pool are read. The exact configuration of the pool is also stored inside the MOS. Since the configuration provided from userland is external and might not accurately describe the vdev tree of the pool at the txg that is being loaded, it cannot be relied upon to safely operate the pool. For that reason, the configuration in the MOS is read early on. In the past, the two configurations were compared together and if there was a mismatch then the load process was aborted and an error was returned. The latter was a good way to ensure a pool does not get corrupted, however it made the pool load process needlessly fragile in cases where the vdev configuration changed or the userland configuration was outdated. Since the MOS is stored in 3 copies, the configuration provided by userland doesn't have to be perfect in order to read its contents. Hence, a new approach has been adopted: The pool is first opened with the untrusted userland configuration just so that the real configuration can be read from the MOS. The trusted MOS configuration is then used to generate a new vdev tree and the pool is re-opened. When the pool is opened with an untrusted configuration, writes are disabled to avoid accidentally damaging it. During reads, some sanity checks are performed on block pointers to see if each DVA points to a known vdev; when the configuration is untrusted, instead of panicking the system if those checks fail we simply avoid issuing reads to the invalid DVAs. This new two-step pool load process now allows rewinding pools accross vdev tree changes such as device replacement, addition, etc. Loading a pool from an external config file in a clustering environment also becomes much safer now since the pool will import even if the config is outdated and didn't, for instance, register a recent device addition. With this code in place, it became relatively easy to implement a long-sought-after feature: the ability to import a pool with missing top level (i.e. non-redundant) devices. Note that since this almost guarantees some loss of data, this feature is for now restricted to a read-only import. Porting notes (ZTS): * Fix 'make dist' target in zpool_import * The maximum path length allowed by tar is 99 characters. Several of the new test cases exceeded this limit resulting in them not being included in the tarball. Shorten the names slightly. * Set/get tunables using accessor functions. * Get last synced txg via the "zfs_txg_history" mechanism. * Clear zinject handlers in cleanup for import_cache_device_replaced and import_rewind_device_replaced in order that the zpool can be exported if there is an error. * Increase FILESIZE to 8G in zfs-test.sh to allow for a larger ext4 file system to be created on ZFS_DISK2. Also, there's no need to partition ZFS_DISK2 at all. The partitioning had already been disabled for multipath devices. Among other things, the partitioning steals some space from the ext4 file system, makes it difficult to accurately calculate the paramters to parted and can make some of the tests fail. * Increase FS_SIZE and FILE_SIZE in the zpool_import test configuration now that FILESIZE is larger. * Write more data in order that device evacuation take lonnger in a couple tests. * Use mkdir -p to avoid errors when the directory already exists. * Remove use of sudo in import_rewind_config_changed. Authored by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Andrew Stormont <andyjstormont@gmail.com> Approved by: Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9075 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/619c0123 Closes #7459
2016-07-22 14:39:36 +00:00
void
spa_set_missing_tvds(spa_t *spa, uint64_t missing)
{
spa->spa_missing_tvds = missing;
}
/*
* Return the pool state string ("ONLINE", "DEGRADED", "SUSPENDED", etc).
*/
const char *
spa_state_to_name(spa_t *spa)
{
2019-05-23 21:28:53 +00:00
ASSERT3P(spa, !=, NULL);
/*
* it is possible for the spa to exist, without root vdev
* as the spa transitions during import/export
*/
vdev_t *rvd = spa->spa_root_vdev;
if (rvd == NULL) {
return ("TRANSITIONING");
}
vdev_state_t state = rvd->vdev_state;
vdev_aux_t aux = rvd->vdev_stat.vs_aux;
if (spa_suspended(spa) &&
(spa_get_failmode(spa) != ZIO_FAILURE_MODE_CONTINUE))
return ("SUSPENDED");
switch (state) {
case VDEV_STATE_CLOSED:
case VDEV_STATE_OFFLINE:
return ("OFFLINE");
case VDEV_STATE_REMOVED:
return ("REMOVED");
case VDEV_STATE_CANT_OPEN:
if (aux == VDEV_AUX_CORRUPT_DATA || aux == VDEV_AUX_BAD_LOG)
return ("FAULTED");
else if (aux == VDEV_AUX_SPLIT_POOL)
return ("SPLIT");
else
return ("UNAVAIL");
case VDEV_STATE_FAULTED:
return ("FAULTED");
case VDEV_STATE_DEGRADED:
return ("DEGRADED");
case VDEV_STATE_HEALTHY:
return ("ONLINE");
default:
break;
}
return ("UNKNOWN");
}
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
boolean_t
spa_top_vdevs_spacemap_addressable(spa_t *spa)
{
vdev_t *rvd = spa->spa_root_vdev;
for (uint64_t c = 0; c < rvd->vdev_children; c++) {
if (!vdev_is_spacemap_addressable(rvd->vdev_child[c]))
return (B_FALSE);
}
return (B_TRUE);
}
boolean_t
spa_has_checkpoint(spa_t *spa)
{
return (spa->spa_checkpoint_txg != 0);
}
boolean_t
spa_importing_readonly_checkpoint(spa_t *spa)
{
return ((spa->spa_import_flags & ZFS_IMPORT_CHECKPOINT) &&
spa->spa_mode == SPA_MODE_READ);
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
}
uint64_t
spa_min_claim_txg(spa_t *spa)
{
uint64_t checkpoint_txg = spa->spa_uberblock.ub_checkpoint_txg;
if (checkpoint_txg != 0)
return (checkpoint_txg + 1);
return (spa->spa_first_txg);
}
/*
* If there is a checkpoint, async destroys may consume more space from
* the pool instead of freeing it. In an attempt to save the pool from
* getting suspended when it is about to run out of space, we stop
* processing async destroys.
*/
boolean_t
spa_suspend_async_destroy(spa_t *spa)
{
dsl_pool_t *dp = spa_get_dsl(spa);
uint64_t unreserved = dsl_pool_unreserved_space(dp,
ZFS_SPACE_CHECK_EXTRA_RESERVED);
uint64_t used = dsl_dir_phys(dp->dp_root_dir)->dd_used_bytes;
uint64_t avail = (unreserved > used) ? (unreserved - used) : 0;
if (spa_has_checkpoint(spa) && avail == 0)
return (B_TRUE);
return (B_FALSE);
}
Update build system and packaging Minimal changes required to integrate the SPL sources in to the ZFS repository build infrastructure and packaging. Build system and packaging: * Renamed SPL_* autoconf m4 macros to ZFS_*. * Removed redundant SPL_* autoconf m4 macros. * Updated the RPM spec files to remove SPL package dependency. * The zfs package obsoletes the spl package, and the zfs-kmod package obsoletes the spl-kmod package. * The zfs-kmod-devel* packages were updated to add compatibility symlinks under /usr/src/spl-x.y.z until all dependent packages can be updated. They will be removed in a future release. * Updated copy-builtin script for in-kernel builds. * Updated DKMS package to include the spl.ko. * Updated stale AUTHORS file to include all contributors. * Updated stale COPYRIGHT and included the SPL as an exception. * Renamed README.markdown to README.md * Renamed OPENSOLARIS.LICENSE to LICENSE. * Renamed DISCLAIMER to NOTICE. Required code changes: * Removed redundant HAVE_SPL macro. * Removed _BOOT from nvpairs since it doesn't apply for Linux. * Initial header cleanup (removal of empty headers, refactoring). * Remove SPL repository clone/build from zimport.sh. * Use of DEFINE_RATELIMIT_STATE and DEFINE_SPINLOCK removed due to build issues when forcing C99 compilation. * Replaced legacy ACCESS_ONCE with READ_ONCE. * Include needed headers for `current` and `EXPORT_SYMBOL`. Reviewed-by: Tony Hutter <hutter2@llnl.gov> Reviewed-by: Olaf Faaland <faaland1@llnl.gov> Reviewed-by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Pavel Zakharov <pavel.zakharov@delphix.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> TEST_ZIMPORT_SKIP="yes" Closes #7556
2018-02-16 01:53:18 +00:00
#if defined(_KERNEL)
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
int
param_set_deadman_failmode_common(const char *val)
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
{
spa_t *spa = NULL;
char *p;
if (val == NULL)
return (SET_ERROR(EINVAL));
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
if ((p = strchr(val, '\n')) != NULL)
*p = '\0';
if (strcmp(val, "wait") != 0 && strcmp(val, "continue") != 0 &&
strcmp(val, "panic"))
return (SET_ERROR(EINVAL));
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
if (spa_mode_global != SPA_MODE_UNINIT) {
mutex_enter(&spa_namespace_lock);
while ((spa = spa_next(spa)) != NULL)
spa_set_deadman_failmode(spa, val);
mutex_exit(&spa_namespace_lock);
}
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
return (0);
Extend deadman logic The intent of this patch is extend the existing deadman code such that it's flexible enough to be used by both ztest and on production systems. The proposed changes include: * Added a new `zfs_deadman_failmode` module option which is used to dynamically control the behavior of the deadman. It's loosely modeled after, but independant from, the pool failmode property. It can be set to wait, continue, or panic. * wait - Wait for the "hung" I/O (default) * continue - Attempt to recover from a "hung" I/O * panic - Panic the system * Added a new `zfs_deadman_ziotime_ms` module option which is analogous to `zfs_deadman_synctime_ms` except instead of applying to a pool TXG sync it applies to zio_wait(). A default value of 300s is used to define a "hung" zio. * The ztest deadman thread has been re-enabled by default, aligned with the upstream OpenZFS code, and then extended to terminate the process when it takes significantly longer to complete than expected. * The -G option was added to ztest to print the internal debug log when a fatal error is encountered. This same option was previously added to zdb in commit fa603f82. Update zloop.sh to unconditionally pass -G to obtain additional debugging. * The FM_EREPORT_ZFS_DELAY event which was previously posted when the deadman detect a "hung" pool has been replaced by a new dedicated FM_EREPORT_ZFS_DEADMAN event. * The proposed recovery logic attempts to restart a "hung" zio by calling zio_interrupt() on any outstanding leaf zios. We may want to further restrict this to zios in either the ZIO_STAGE_VDEV_IO_START or ZIO_STAGE_VDEV_IO_DONE stages. Calling zio_interrupt() is expected to only be useful for cases when an IO has been submitted to the physical device but for some reasonable the completion callback hasn't been called by the lower layers. This shouldn't be possible but has been observed and may be caused by kernel/driver bugs. * The 'zfs_deadman_synctime_ms' default value was reduced from 1000s to 600s. * Depending on how ztest fails there may be no cache file to move. This should not be considered fatal, collect the logs which are available and carry on. * Add deadman test cases for spa_deadman() and zio_wait(). * Increase default zfs_deadman_checktime_ms to 60s. Reviewed-by: Tim Chase <tim@chase2k.com> Reviewed by: Thomas Caputi <tcaputi@datto.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6999
2017-12-18 22:06:07 +00:00
}
#endif
/* Namespace manipulation */
EXPORT_SYMBOL(spa_lookup);
EXPORT_SYMBOL(spa_add);
EXPORT_SYMBOL(spa_remove);
EXPORT_SYMBOL(spa_next);
/* Refcount functions */
EXPORT_SYMBOL(spa_open_ref);
EXPORT_SYMBOL(spa_close);
EXPORT_SYMBOL(spa_refcount_zero);
/* Pool configuration lock */
EXPORT_SYMBOL(spa_config_tryenter);
EXPORT_SYMBOL(spa_config_enter);
EXPORT_SYMBOL(spa_config_exit);
EXPORT_SYMBOL(spa_config_held);
/* Pool vdev add/remove lock */
EXPORT_SYMBOL(spa_vdev_enter);
EXPORT_SYMBOL(spa_vdev_exit);
/* Pool vdev state change lock */
EXPORT_SYMBOL(spa_vdev_state_enter);
EXPORT_SYMBOL(spa_vdev_state_exit);
/* Accessor functions */
EXPORT_SYMBOL(spa_shutting_down);
EXPORT_SYMBOL(spa_get_dsl);
EXPORT_SYMBOL(spa_get_rootblkptr);
EXPORT_SYMBOL(spa_set_rootblkptr);
EXPORT_SYMBOL(spa_altroot);
EXPORT_SYMBOL(spa_sync_pass);
EXPORT_SYMBOL(spa_name);
EXPORT_SYMBOL(spa_guid);
EXPORT_SYMBOL(spa_last_synced_txg);
EXPORT_SYMBOL(spa_first_txg);
EXPORT_SYMBOL(spa_syncing_txg);
EXPORT_SYMBOL(spa_version);
EXPORT_SYMBOL(spa_state);
EXPORT_SYMBOL(spa_load_state);
EXPORT_SYMBOL(spa_freeze_txg);
EXPORT_SYMBOL(spa_get_dspace);
EXPORT_SYMBOL(spa_update_dspace);
EXPORT_SYMBOL(spa_deflate);
EXPORT_SYMBOL(spa_normal_class);
EXPORT_SYMBOL(spa_log_class);
EXPORT_SYMBOL(spa_special_class);
EXPORT_SYMBOL(spa_preferred_class);
EXPORT_SYMBOL(spa_max_replication);
EXPORT_SYMBOL(spa_prev_software_version);
EXPORT_SYMBOL(spa_get_failmode);
EXPORT_SYMBOL(spa_suspended);
EXPORT_SYMBOL(spa_bootfs);
EXPORT_SYMBOL(spa_delegation);
EXPORT_SYMBOL(spa_meta_objset);
Illumos 5027 - zfs large block support 5027 zfs large block support Reviewed by: Alek Pinchuk <pinchuk.alek@gmail.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com> Reviewed by: Richard Elling <richard.elling@richardelling.com> Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/5027 https://github.com/illumos/illumos-gate/commit/b515258 Porting Notes: * Included in this patch is a tiny ISP2() cleanup in zio_init() from Illumos 5255. * Unlike the upstream Illumos commit this patch does not impose an arbitrary 128K block size limit on volumes. Volumes, like filesystems, are limited by the zfs_max_recordsize=1M module option. * By default the maximum record size is limited to 1M by the module option zfs_max_recordsize. This value may be safely increased up to 16M which is the largest block size supported by the on-disk format. At the moment, 1M blocks clearly offer a significant performance improvement but the benefits of going beyond this for the majority of workloads are less clear. * The illumos version of this patch increased DMU_MAX_ACCESS to 32M. This was determined not to be large enough when using 16M blocks because the zfs_make_xattrdir() function will fail (EFBIG) when assigning a TX. This was immediately observed under Linux because all newly created files must have a security xattr created and that was failing. Therefore, we've set DMU_MAX_ACCESS to 64M. * On 32-bit platforms a hard limit of 1M is set for blocks due to the limited virtual address space. We should be able to relax this one the ABD patches are merged. Ported-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #354
2014-11-03 20:15:08 +00:00
EXPORT_SYMBOL(spa_maxblocksize);
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
EXPORT_SYMBOL(spa_maxdnodesize);
/* Miscellaneous support routines */
EXPORT_SYMBOL(spa_guid_exists);
EXPORT_SYMBOL(spa_strdup);
EXPORT_SYMBOL(spa_strfree);
EXPORT_SYMBOL(spa_get_random);
EXPORT_SYMBOL(spa_generate_guid);
EXPORT_SYMBOL(snprintf_blkptr);
EXPORT_SYMBOL(spa_freeze);
EXPORT_SYMBOL(spa_upgrade);
EXPORT_SYMBOL(spa_evict_all);
EXPORT_SYMBOL(spa_lookup_by_guid);
EXPORT_SYMBOL(spa_has_spare);
EXPORT_SYMBOL(dva_get_dsize_sync);
EXPORT_SYMBOL(bp_get_dsize_sync);
EXPORT_SYMBOL(bp_get_dsize);
EXPORT_SYMBOL(spa_has_slogs);
EXPORT_SYMBOL(spa_is_root);
EXPORT_SYMBOL(spa_writeable);
EXPORT_SYMBOL(spa_mode);
EXPORT_SYMBOL(spa_namespace_lock);
OpenZFS 9075 - Improve ZFS pool import/load process and corrupted pool recovery Some work has been done lately to improve the debugability of the ZFS pool load (and import) process. This includes: 7638 Refactor spa_load_impl into several functions 8961 SPA load/import should tell us why it failed 7277 zdb should be able to print zfs_dbgmsg's To iterate on top of that, there's a few changes that were made to make the import process more resilient and crash free. One of the first tasks during the pool load process is to parse a config provided from userland that describes what devices the pool is composed of. A vdev tree is generated from that config, and then all the vdevs are opened. The Meta Object Set (MOS) of the pool is accessed, and several metadata objects that are necessary to load the pool are read. The exact configuration of the pool is also stored inside the MOS. Since the configuration provided from userland is external and might not accurately describe the vdev tree of the pool at the txg that is being loaded, it cannot be relied upon to safely operate the pool. For that reason, the configuration in the MOS is read early on. In the past, the two configurations were compared together and if there was a mismatch then the load process was aborted and an error was returned. The latter was a good way to ensure a pool does not get corrupted, however it made the pool load process needlessly fragile in cases where the vdev configuration changed or the userland configuration was outdated. Since the MOS is stored in 3 copies, the configuration provided by userland doesn't have to be perfect in order to read its contents. Hence, a new approach has been adopted: The pool is first opened with the untrusted userland configuration just so that the real configuration can be read from the MOS. The trusted MOS configuration is then used to generate a new vdev tree and the pool is re-opened. When the pool is opened with an untrusted configuration, writes are disabled to avoid accidentally damaging it. During reads, some sanity checks are performed on block pointers to see if each DVA points to a known vdev; when the configuration is untrusted, instead of panicking the system if those checks fail we simply avoid issuing reads to the invalid DVAs. This new two-step pool load process now allows rewinding pools accross vdev tree changes such as device replacement, addition, etc. Loading a pool from an external config file in a clustering environment also becomes much safer now since the pool will import even if the config is outdated and didn't, for instance, register a recent device addition. With this code in place, it became relatively easy to implement a long-sought-after feature: the ability to import a pool with missing top level (i.e. non-redundant) devices. Note that since this almost guarantees some loss of data, this feature is for now restricted to a read-only import. Porting notes (ZTS): * Fix 'make dist' target in zpool_import * The maximum path length allowed by tar is 99 characters. Several of the new test cases exceeded this limit resulting in them not being included in the tarball. Shorten the names slightly. * Set/get tunables using accessor functions. * Get last synced txg via the "zfs_txg_history" mechanism. * Clear zinject handlers in cleanup for import_cache_device_replaced and import_rewind_device_replaced in order that the zpool can be exported if there is an error. * Increase FILESIZE to 8G in zfs-test.sh to allow for a larger ext4 file system to be created on ZFS_DISK2. Also, there's no need to partition ZFS_DISK2 at all. The partitioning had already been disabled for multipath devices. Among other things, the partitioning steals some space from the ext4 file system, makes it difficult to accurately calculate the paramters to parted and can make some of the tests fail. * Increase FS_SIZE and FILE_SIZE in the zpool_import test configuration now that FILESIZE is larger. * Write more data in order that device evacuation take lonnger in a couple tests. * Use mkdir -p to avoid errors when the directory already exists. * Remove use of sudo in import_rewind_config_changed. Authored by: Pavel Zakharov <pavel.zakharov@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: Andrew Stormont <andyjstormont@gmail.com> Approved by: Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9075 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/619c0123 Closes #7459
2016-07-22 14:39:36 +00:00
EXPORT_SYMBOL(spa_trust_config);
EXPORT_SYMBOL(spa_missing_tvds_allowed);
EXPORT_SYMBOL(spa_set_missing_tvds);
EXPORT_SYMBOL(spa_state_to_name);
OpenZFS 9166 - zfs storage pool checkpoint Details about the motivation of this feature and its usage can be found in this blogpost: https://sdimitro.github.io/post/zpool-checkpoint/ A lightning talk of this feature can be found here: https://www.youtube.com/watch?v=fPQA8K40jAM Implementation details can be found in big block comment of spa_checkpoint.c Side-changes that are relevant to this commit but not explained elsewhere: * renames members of "struct metaslab trees to be shorter without losing meaning * space_map_{alloc,truncate}() accept a block size as a parameter. The reason is that in the current state all space maps that we allocate through the DMU use a global tunable (space_map_blksz) which defauls to 4KB. This is ok for metaslab space maps in terms of bandwirdth since they are scattered all over the disk. But for other space maps this default is probably not what we want. Examples are device removal's vdev_obsolete_sm or vdev_chedkpoint_sm from this review. Both of these have a 1:1 relationship with each vdev and could benefit from a bigger block size. Porting notes: * The part of dsl_scan_sync() which handles async destroys has been moved into the new dsl_process_async_destroys() function. * Remove "VERIFY(!(flags & FWRITE))" in "kernel.c" so zhack can write to block device backed pools. * ZTS: * Fix get_txg() in zpool_sync_001_pos due to "checkpoint_txg". * Don't use large dd block sizes on /dev/urandom under Linux in checkpoint_capacity. * Adopt Delphix-OS's setting of 4 (spa_asize_inflation = SPA_DVAS_PER_BP + 1) for the checkpoint_capacity test to speed its attempts to fill the pool * Create the base and nested pools with sync=disabled to speed up the "setup" phase. * Clear labels in test pool between checkpoint tests to avoid duplicate pool issues. * The import_rewind_device_replaced test has been marked as "known to fail" for the reasons listed in its DISCLAIMER. * New module parameters: zfs_spa_discard_memory_limit, zfs_remove_max_bytes_pause (not documented - debugging only) vdev_max_ms_count (formerly metaslabs_per_vdev) vdev_min_ms_count Authored by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com> Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: John Kennedy <john.kennedy@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Richard Lowe <richlowe@richlowe.net> Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Tim Chase <tim@chase2k.com> OpenZFS-issue: https://illumos.org/issues/9166 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7159fdb8 Closes #7570
2016-12-16 22:11:29 +00:00
EXPORT_SYMBOL(spa_importing_readonly_checkpoint);
EXPORT_SYMBOL(spa_min_claim_txg);
EXPORT_SYMBOL(spa_suspend_async_destroy);
EXPORT_SYMBOL(spa_has_checkpoint);
EXPORT_SYMBOL(spa_top_vdevs_spacemap_addressable);
ZFS_MODULE_PARAM(zfs, zfs_, flags, UINT, ZMOD_RW,
"Set additional debugging flags");
Swap DTRACE_PROBE* with Linux tracepoints This patch leverages Linux tracepoints from within the ZFS on Linux code base. It also refactors the debug code to bring it back in sync with Illumos. The information exported via tracepoints can be used for a variety of reasons (e.g. debugging, tuning, general exploration/understanding, etc). It is advantageous to use Linux tracepoints as the mechanism to export this kind of information (as opposed to something else) for a number of reasons: * A number of external tools can make use of our tracepoints "automatically" (e.g. perf, systemtap) * Tracepoints are designed to be extremely cheap when disabled * It's one of the "accepted" ways to export this kind of information; many other kernel subsystems use tracepoints too. Unfortunately, though, there are a few caveats as well: * Linux tracepoints appear to only be available to GPL licensed modules due to the way certain kernel functions are exported. Thus, to actually make use of the tracepoints introduced by this patch, one might have to patch and re-compile the kernel; exporting the necessary functions to non-GPL modules. * Prior to upstream kernel version v3.14-rc6-30-g66cc69e, Linux tracepoints are not available for unsigned kernel modules (tracepoints will get disabled due to the module's 'F' taint). Thus, one either has to sign the zfs kernel module prior to loading it, or use a kernel versioned v3.14-rc6-30-g66cc69e or newer. Assuming the above two requirements are satisfied, lets look at an example of how this patch can be used and what information it exposes (all commands run as 'root'): # list all zfs tracepoints available $ ls /sys/kernel/debug/tracing/events/zfs enable filter zfs_arc__delete zfs_arc__evict zfs_arc__hit zfs_arc__miss zfs_l2arc__evict zfs_l2arc__hit zfs_l2arc__iodone zfs_l2arc__miss zfs_l2arc__read zfs_l2arc__write zfs_new_state__mfu zfs_new_state__mru # enable all zfs tracepoints, clear the tracepoint ring buffer $ echo 1 > /sys/kernel/debug/tracing/events/zfs/enable $ echo 0 > /sys/kernel/debug/tracing/trace # import zpool called 'tank', inspect tracepoint data (each line was # truncated, they're too long for a commit message otherwise) $ zpool import tank $ cat /sys/kernel/debug/tracing/trace | head -n35 # tracer: nop # # entries-in-buffer/entries-written: 1219/1219 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr... z_rd_int/0-30156 [003] .... 91344.200611: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201173: zfs_arc__miss: hdr... z_rd_int/1-30157 [003] .... 91344.201756: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201795: zfs_arc__miss: hdr... z_rd_int/2-30158 [003] .... 91344.202099: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202126: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202130: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202134: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202146: zfs_arc__miss: hdr... z_rd_int/3-30159 [003] .... 91344.202457: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202484: zfs_arc__miss: hdr... z_rd_int/4-30160 [003] .... 91344.202866: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202891: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203034: zfs_arc__miss: hdr... z_rd_iss/1-30149 [001] .... 91344.203749: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.203789: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203878: zfs_arc__miss: hdr... z_rd_iss/3-30151 [001] .... 91344.204315: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.204332: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204337: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204352: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204356: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204360: zfs_arc__hit: hdr ... To highlight the kind of detailed information that is being exported using this infrastructure, I've taken the first tracepoint line from the output above and reformatted it such that it fits in 80 columns: lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr { dva 0x1:0x40082 birth 15491 cksum0 0x163edbff3a flags 0x640 datacnt 1 type 1 size 2048 spa 3133524293419867460 state_type 0 access 0 mru_hits 0 mru_ghost_hits 0 mfu_hits 0 mfu_ghost_hits 0 l2_hits 0 refcount 1 } bp { dva0 0x1:0x40082 dva1 0x1:0x3000e5 dva2 0x1:0x5a006e cksum 0x163edbff3a:0x75af30b3dd6:0x1499263ff5f2b:0x288bd118815e00 lsize 2048 } zb { objset 0 object 0 level -1 blkid 0 } For the specific tracepoint shown here, 'zfs_arc__miss', data is exported detailing the arc_buf_hdr_t (hdr), blkptr_t (bp), and zbookmark_t (zb) that caused the ARC miss (down to the exact DVA!). This kind of precise and detailed information can be extremely valuable when trying to answer certain kinds of questions. For anybody unfamiliar but looking to build on this, I found the XFS source code along with the following three web links to be extremely helpful: * http://lwn.net/Articles/379903/ * http://lwn.net/Articles/381064/ * http://lwn.net/Articles/383362/ I should also node the more "boring" aspects of this patch: * The ZFS_LINUX_COMPILE_IFELSE autoconf macro was modified to support a sixth paramter. This parameter is used to populate the contents of the new conftest.h file. If no sixth parameter is provided, conftest.h will be empty. * The ZFS_LINUX_TRY_COMPILE_HEADER autoconf macro was introduced. This macro is nearly identical to the ZFS_LINUX_TRY_COMPILE macro, except it has support for a fifth option that is then passed as the sixth parameter to ZFS_LINUX_COMPILE_IFELSE. These autoconf changes were needed to test the availability of the Linux tracepoint macros. Due to the odd nature of the Linux tracepoint macro API, a separate ".h" must be created (the path and filename is used internally by the kernel's define_trace.h file). * The HAVE_DECLARE_EVENT_CLASS autoconf macro was introduced. This is to determine if we can safely enable the Linux tracepoint functionality. We need to selectively disable the tracepoint code due to the kernel exporting certain functions as GPL only. Without this check, the build process will fail at link time. In addition, the SET_ERROR macro was modified into a tracepoint as well. To do this, the 'sdt.h' file was moved into the 'include/sys' directory and now contains a userspace portion and a kernel space portion. The dprintf and zfs_dbgmsg* interfaces are now implemented as tracepoint as well. Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-06-13 17:54:48 +00:00
ZFS_MODULE_PARAM(zfs, zfs_, recover, INT, ZMOD_RW,
"Set to attempt to recover from fatal errors");
Swap DTRACE_PROBE* with Linux tracepoints This patch leverages Linux tracepoints from within the ZFS on Linux code base. It also refactors the debug code to bring it back in sync with Illumos. The information exported via tracepoints can be used for a variety of reasons (e.g. debugging, tuning, general exploration/understanding, etc). It is advantageous to use Linux tracepoints as the mechanism to export this kind of information (as opposed to something else) for a number of reasons: * A number of external tools can make use of our tracepoints "automatically" (e.g. perf, systemtap) * Tracepoints are designed to be extremely cheap when disabled * It's one of the "accepted" ways to export this kind of information; many other kernel subsystems use tracepoints too. Unfortunately, though, there are a few caveats as well: * Linux tracepoints appear to only be available to GPL licensed modules due to the way certain kernel functions are exported. Thus, to actually make use of the tracepoints introduced by this patch, one might have to patch and re-compile the kernel; exporting the necessary functions to non-GPL modules. * Prior to upstream kernel version v3.14-rc6-30-g66cc69e, Linux tracepoints are not available for unsigned kernel modules (tracepoints will get disabled due to the module's 'F' taint). Thus, one either has to sign the zfs kernel module prior to loading it, or use a kernel versioned v3.14-rc6-30-g66cc69e or newer. Assuming the above two requirements are satisfied, lets look at an example of how this patch can be used and what information it exposes (all commands run as 'root'): # list all zfs tracepoints available $ ls /sys/kernel/debug/tracing/events/zfs enable filter zfs_arc__delete zfs_arc__evict zfs_arc__hit zfs_arc__miss zfs_l2arc__evict zfs_l2arc__hit zfs_l2arc__iodone zfs_l2arc__miss zfs_l2arc__read zfs_l2arc__write zfs_new_state__mfu zfs_new_state__mru # enable all zfs tracepoints, clear the tracepoint ring buffer $ echo 1 > /sys/kernel/debug/tracing/events/zfs/enable $ echo 0 > /sys/kernel/debug/tracing/trace # import zpool called 'tank', inspect tracepoint data (each line was # truncated, they're too long for a commit message otherwise) $ zpool import tank $ cat /sys/kernel/debug/tracing/trace | head -n35 # tracer: nop # # entries-in-buffer/entries-written: 1219/1219 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr... z_rd_int/0-30156 [003] .... 91344.200611: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201173: zfs_arc__miss: hdr... z_rd_int/1-30157 [003] .... 91344.201756: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201795: zfs_arc__miss: hdr... z_rd_int/2-30158 [003] .... 91344.202099: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202126: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202130: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202134: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202146: zfs_arc__miss: hdr... z_rd_int/3-30159 [003] .... 91344.202457: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202484: zfs_arc__miss: hdr... z_rd_int/4-30160 [003] .... 91344.202866: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202891: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203034: zfs_arc__miss: hdr... z_rd_iss/1-30149 [001] .... 91344.203749: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.203789: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203878: zfs_arc__miss: hdr... z_rd_iss/3-30151 [001] .... 91344.204315: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.204332: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204337: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204352: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204356: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204360: zfs_arc__hit: hdr ... To highlight the kind of detailed information that is being exported using this infrastructure, I've taken the first tracepoint line from the output above and reformatted it such that it fits in 80 columns: lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr { dva 0x1:0x40082 birth 15491 cksum0 0x163edbff3a flags 0x640 datacnt 1 type 1 size 2048 spa 3133524293419867460 state_type 0 access 0 mru_hits 0 mru_ghost_hits 0 mfu_hits 0 mfu_ghost_hits 0 l2_hits 0 refcount 1 } bp { dva0 0x1:0x40082 dva1 0x1:0x3000e5 dva2 0x1:0x5a006e cksum 0x163edbff3a:0x75af30b3dd6:0x1499263ff5f2b:0x288bd118815e00 lsize 2048 } zb { objset 0 object 0 level -1 blkid 0 } For the specific tracepoint shown here, 'zfs_arc__miss', data is exported detailing the arc_buf_hdr_t (hdr), blkptr_t (bp), and zbookmark_t (zb) that caused the ARC miss (down to the exact DVA!). This kind of precise and detailed information can be extremely valuable when trying to answer certain kinds of questions. For anybody unfamiliar but looking to build on this, I found the XFS source code along with the following three web links to be extremely helpful: * http://lwn.net/Articles/379903/ * http://lwn.net/Articles/381064/ * http://lwn.net/Articles/383362/ I should also node the more "boring" aspects of this patch: * The ZFS_LINUX_COMPILE_IFELSE autoconf macro was modified to support a sixth paramter. This parameter is used to populate the contents of the new conftest.h file. If no sixth parameter is provided, conftest.h will be empty. * The ZFS_LINUX_TRY_COMPILE_HEADER autoconf macro was introduced. This macro is nearly identical to the ZFS_LINUX_TRY_COMPILE macro, except it has support for a fifth option that is then passed as the sixth parameter to ZFS_LINUX_COMPILE_IFELSE. These autoconf changes were needed to test the availability of the Linux tracepoint macros. Due to the odd nature of the Linux tracepoint macro API, a separate ".h" must be created (the path and filename is used internally by the kernel's define_trace.h file). * The HAVE_DECLARE_EVENT_CLASS autoconf macro was introduced. This is to determine if we can safely enable the Linux tracepoint functionality. We need to selectively disable the tracepoint code due to the kernel exporting certain functions as GPL only. Without this check, the build process will fail at link time. In addition, the SET_ERROR macro was modified into a tracepoint as well. To do this, the 'sdt.h' file was moved into the 'include/sys' directory and now contains a userspace portion and a kernel space portion. The dprintf and zfs_dbgmsg* interfaces are now implemented as tracepoint as well. Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-06-13 17:54:48 +00:00
ZFS_MODULE_PARAM(zfs, zfs_, free_leak_on_eio, INT, ZMOD_RW,
Swap DTRACE_PROBE* with Linux tracepoints This patch leverages Linux tracepoints from within the ZFS on Linux code base. It also refactors the debug code to bring it back in sync with Illumos. The information exported via tracepoints can be used for a variety of reasons (e.g. debugging, tuning, general exploration/understanding, etc). It is advantageous to use Linux tracepoints as the mechanism to export this kind of information (as opposed to something else) for a number of reasons: * A number of external tools can make use of our tracepoints "automatically" (e.g. perf, systemtap) * Tracepoints are designed to be extremely cheap when disabled * It's one of the "accepted" ways to export this kind of information; many other kernel subsystems use tracepoints too. Unfortunately, though, there are a few caveats as well: * Linux tracepoints appear to only be available to GPL licensed modules due to the way certain kernel functions are exported. Thus, to actually make use of the tracepoints introduced by this patch, one might have to patch and re-compile the kernel; exporting the necessary functions to non-GPL modules. * Prior to upstream kernel version v3.14-rc6-30-g66cc69e, Linux tracepoints are not available for unsigned kernel modules (tracepoints will get disabled due to the module's 'F' taint). Thus, one either has to sign the zfs kernel module prior to loading it, or use a kernel versioned v3.14-rc6-30-g66cc69e or newer. Assuming the above two requirements are satisfied, lets look at an example of how this patch can be used and what information it exposes (all commands run as 'root'): # list all zfs tracepoints available $ ls /sys/kernel/debug/tracing/events/zfs enable filter zfs_arc__delete zfs_arc__evict zfs_arc__hit zfs_arc__miss zfs_l2arc__evict zfs_l2arc__hit zfs_l2arc__iodone zfs_l2arc__miss zfs_l2arc__read zfs_l2arc__write zfs_new_state__mfu zfs_new_state__mru # enable all zfs tracepoints, clear the tracepoint ring buffer $ echo 1 > /sys/kernel/debug/tracing/events/zfs/enable $ echo 0 > /sys/kernel/debug/tracing/trace # import zpool called 'tank', inspect tracepoint data (each line was # truncated, they're too long for a commit message otherwise) $ zpool import tank $ cat /sys/kernel/debug/tracing/trace | head -n35 # tracer: nop # # entries-in-buffer/entries-written: 1219/1219 #P:8 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr... z_rd_int/0-30156 [003] .... 91344.200611: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201173: zfs_arc__miss: hdr... z_rd_int/1-30157 [003] .... 91344.201756: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.201795: zfs_arc__miss: hdr... z_rd_int/2-30158 [003] .... 91344.202099: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202126: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202130: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202134: zfs_arc__hit: hdr ... lt-zpool-30132 [003] .... 91344.202146: zfs_arc__miss: hdr... z_rd_int/3-30159 [003] .... 91344.202457: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202484: zfs_arc__miss: hdr... z_rd_int/4-30160 [003] .... 91344.202866: zfs_new_state__mru... lt-zpool-30132 [003] .... 91344.202891: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203034: zfs_arc__miss: hdr... z_rd_iss/1-30149 [001] .... 91344.203749: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.203789: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.203878: zfs_arc__miss: hdr... z_rd_iss/3-30151 [001] .... 91344.204315: zfs_new_state__mru... lt-zpool-30132 [001] .... 91344.204332: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204337: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204352: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204356: zfs_arc__hit: hdr ... lt-zpool-30132 [001] .... 91344.204360: zfs_arc__hit: hdr ... To highlight the kind of detailed information that is being exported using this infrastructure, I've taken the first tracepoint line from the output above and reformatted it such that it fits in 80 columns: lt-zpool-30132 [003] .... 91344.200050: zfs_arc__miss: hdr { dva 0x1:0x40082 birth 15491 cksum0 0x163edbff3a flags 0x640 datacnt 1 type 1 size 2048 spa 3133524293419867460 state_type 0 access 0 mru_hits 0 mru_ghost_hits 0 mfu_hits 0 mfu_ghost_hits 0 l2_hits 0 refcount 1 } bp { dva0 0x1:0x40082 dva1 0x1:0x3000e5 dva2 0x1:0x5a006e cksum 0x163edbff3a:0x75af30b3dd6:0x1499263ff5f2b:0x288bd118815e00 lsize 2048 } zb { objset 0 object 0 level -1 blkid 0 } For the specific tracepoint shown here, 'zfs_arc__miss', data is exported detailing the arc_buf_hdr_t (hdr), blkptr_t (bp), and zbookmark_t (zb) that caused the ARC miss (down to the exact DVA!). This kind of precise and detailed information can be extremely valuable when trying to answer certain kinds of questions. For anybody unfamiliar but looking to build on this, I found the XFS source code along with the following three web links to be extremely helpful: * http://lwn.net/Articles/379903/ * http://lwn.net/Articles/381064/ * http://lwn.net/Articles/383362/ I should also node the more "boring" aspects of this patch: * The ZFS_LINUX_COMPILE_IFELSE autoconf macro was modified to support a sixth paramter. This parameter is used to populate the contents of the new conftest.h file. If no sixth parameter is provided, conftest.h will be empty. * The ZFS_LINUX_TRY_COMPILE_HEADER autoconf macro was introduced. This macro is nearly identical to the ZFS_LINUX_TRY_COMPILE macro, except it has support for a fifth option that is then passed as the sixth parameter to ZFS_LINUX_COMPILE_IFELSE. These autoconf changes were needed to test the availability of the Linux tracepoint macros. Due to the odd nature of the Linux tracepoint macro API, a separate ".h" must be created (the path and filename is used internally by the kernel's define_trace.h file). * The HAVE_DECLARE_EVENT_CLASS autoconf macro was introduced. This is to determine if we can safely enable the Linux tracepoint functionality. We need to selectively disable the tracepoint code due to the kernel exporting certain functions as GPL only. Without this check, the build process will fail at link time. In addition, the SET_ERROR macro was modified into a tracepoint as well. To do this, the 'sdt.h' file was moved into the 'include/sys' directory and now contains a userspace portion and a kernel space portion. The dprintf and zfs_dbgmsg* interfaces are now implemented as tracepoint as well. Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-06-13 17:54:48 +00:00
"Set to ignore IO errors during free and permanently leak the space");
ZFS_MODULE_PARAM(zfs, zfs_, deadman_checktime_ms, ULONG, ZMOD_RW,
"Dead I/O check interval in milliseconds");
ZFS_MODULE_PARAM(zfs, zfs_, deadman_enabled, INT, ZMOD_RW,
"Enable deadman timer");
ZFS_MODULE_PARAM(zfs_spa, spa_, asize_inflation, INT, ZMOD_RW,
"SPA size estimate multiplication factor");
ZFS_MODULE_PARAM(zfs, zfs_, ddt_data_is_special, INT, ZMOD_RW,
"Place DDT data into the special class");
ZFS_MODULE_PARAM(zfs, zfs_, user_indirect_is_special, INT, ZMOD_RW,
"Place user data indirect blocks into the special class");
/* BEGIN CSTYLED */
ZFS_MODULE_PARAM_CALL(zfs_deadman, zfs_deadman_, failmode,
param_set_deadman_failmode, param_get_charp, ZMOD_RW,
"Failmode for deadman timer");
ZFS_MODULE_PARAM_CALL(zfs_deadman, zfs_deadman_, synctime_ms,
param_set_deadman_synctime, param_get_ulong, ZMOD_RW,
"Pool sync expiration time in milliseconds");
ZFS_MODULE_PARAM_CALL(zfs_deadman, zfs_deadman_, ziotime_ms,
param_set_deadman_ziotime, param_get_ulong, ZMOD_RW,
"IO expiration time in milliseconds");
ZFS_MODULE_PARAM(zfs, zfs_, special_class_metadata_reserve_pct, INT, ZMOD_RW,
"Small file blocks in special vdevs depends on this much "
"free space available");
/* END CSTYLED */
ZFS_MODULE_PARAM_CALL(zfs_spa, spa_, slop_shift, param_set_slop_shift,
param_get_int, ZMOD_RW, "Reserved free space in pool");