summaryrefslogtreecommitdiffstats
path: root/lib
Commit message (Collapse)AuthorAgeFilesLines
* ICP: Improve AES-GCM performanceAttila Fülöp2020-02-102-1/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently SIMD accelerated AES-GCM performance is limited by two factors: a. The need to disable preemption and interrupts and save the FPU state before using it and to do the reverse when done. Due to the way the code is organized (see (b) below) we have to pay this price twice for each 16 byte GCM block processed. b. Most processing is done in C, operating on single GCM blocks. The use of SIMD instructions is limited to the AES encryption of the counter block (AES-NI) and the Galois multiplication (PCLMULQDQ). This leads to the FPU not being fully utilized for crypto operations. To solve (a) we do crypto processing in larger chunks while owning the FPU. An `icp_gcm_avx_chunk_size` module parameter was introduced to make this chunk size tweakable. It defaults to 32 KiB. This step alone roughly doubles performance. (b) is tackled by porting and using the highly optimized openssl AES-GCM assembler routines, which do all the processing (CTR, AES, GMULT) in a single routine. Both steps together result in up to 32x reduction of the time spend in the en/decryption routines, leading up to approximately 12x throughput increase for large (128 KiB) blocks. Lastly, this commit changes the default encryption algorithm from AES-CCM to AES-GCM when setting the `encryption=on` property. Reviewed-By: Brian Behlendorf <[email protected]> Reviewed-By: Jason King <[email protected]> Reviewed-By: Tom Caputi <[email protected]> Reviewed-By: Richard Laager <[email protected]> Signed-off-by: Attila Fülöp <[email protected]> Closes #9749
* Fix static data to link with -fno-commonRomain Dolbeau2020-02-062-1/+3
| | | | | | | | -fno-common is the new default in GCC 10, replacing -fcommon in GCC <= 9, so static data must only be allocated once. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Romain Dolbeau <[email protected]> Closes #9943
* async zvol minor node creation interferes with receiveMatthew Ahrens2020-02-031-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we finish a zfs receive, dmu_recv_end_sync() calls zvol_create_minors(async=TRUE). This kicks off some other threads that create the minor device nodes (in /dev/zvol/poolname/...). These async threads call zvol_prefetch_minors_impl() and zvol_create_minor(), which both call dmu_objset_own(), which puts a "long hold" on the dataset. Since the zvol minor node creation is asynchronous, this can happen after the `ZFS_IOC_RECV[_NEW]` ioctl and `zfs receive` process have completed. After the first receive ioctl has completed, userland may attempt to do another receive into the same dataset (e.g. the next incremental stream). This second receive and the asynchronous minor node creation can interfere with one another in several different ways, because they both require exclusive access to the dataset: 1. When the second receive is finishing up, dmu_recv_end_check() does dsl_dataset_handoff_check(), which can fail with EBUSY if the async minor node creation already has a "long hold" on this dataset. This causes the 2nd receive to fail. 2. The async udev rule can fail if zvol_id and/or systemd-udevd try to open the device while the the second receive's async attempt at minor node creation owns the dataset (via zvol_prefetch_minors_impl). This causes the minor node (/dev/zd*) to exist, but the udev-generated /dev/zvol/... to not exist. 3. The async minor node creation can silently fail with EBUSY if the first receive's zvol_create_minor() trys to own the dataset while the second receive's zvol_prefetch_minors_impl already owns the dataset. To address these problems, this change synchronously creates the minor node. To avoid the lock ordering problems that the asynchrony was introduced to fix (see #3681), we create the minor nodes from open context, with no locks held, rather than from syncing contex as was originally done. Implementation notes: We generally do not need to traverse children or prefetch anything (e.g. when running the recv, snapshot, create, or clone subcommands of zfs). We only need recursion when importing/opening a pool and when loading encryption keys. The existing recursive, asynchronous, prefetching code is preserved for use in these cases. Channel programs may need to create zvol minor nodes, when creating a snapshot of a zvol with the snapdev property set. We figure out what snapshots are created when running the LUA program in syncing context. In this case we need to remember what snapshots were created, and then try to create their minor nodes from open context, after the LUA code has completed. There are additional zvol use cases that asynchronously own the dataset, which can cause similar problems. E.g. changing the volmode or snapdev properties. These are less problematic because they are not recursive and don't touch datasets that are not involved in the operation, there is still potential for interference with subsequent operations. In the future, these cases should be similarly converted to create the zvol minor node synchronously from open context. The async tasks of removing and renaming minors do not own the objset, so they do not have this problem. However, it may make sense to also convert these operations to happen synchronously from open context, in the future. Reviewed-by: Paul Dagnelie <[email protected]> Reviewed-by: Prakash Surya <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matthew Ahrens <[email protected]> External-issue: DLPX-65948 Closes #7863 Closes #9885
* Add AltiVec RAID-ZRomain Dolbeau2020-01-232-0/+46
| | | | | | | | | | | | | Implements the RAID-Z function using AltiVec SIMD. This is basically the NEON code translated to AltiVec. Note that the 'fletcher' algorithm requires 64-bits operations, and the initial implementations of AltiVec (PPC74xx a.k.a. G4, PPC970 a.k.a. G5) only has up to 32-bits operations, so no 'fletcher'. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Romain Dolbeau <[email protected]> Closes #9539
* Disable get_numeric_property for xattr on FreeBSDMatthew Macy2020-01-211-0/+2
| | | | | | | | | | FreeBSD doesn't have a mount flag for determining the disposition of xattr. Disable so that it is fetched by the default route so that 'zfs get xattr' returns the correct value. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9862
* Re-consolidate zio_delay_interruptMatthew Macy2020-01-211-1/+0
| | | | | | | | With recent SPL changes there is no longer any need for a per platform version. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9860
* Unify target_cpu handlingBrian Behlendorf2020-01-173-17/+15
| | | | | | | | | | | | | Over the years several slightly different approaches were used in the Makefiles to determine the target architecture. This change updates both the build system and Makefile to handle this in a consistent fashion. TARGET_CPU is set to i386, x86_64, powerpc, aarch6 or sparc64 and made available in the Makefiles to be used as appropriate. Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #9848
* libzfs: add zfs_mount_at() functionKyle Evans2020-01-141-8/+37
| | | | | | | | | | | | | | zfs_mount_at() mounts a dataset at an arbitrary mountpoint rather than at the configured mountpoint. This may be used by consumers that wish to temporarily expose a dataset at another mountpoint without altering dataset/pool properties. This will be used by FreeBSD's libbe be_mount(), which mounts a boot environment at an arbitrary mountpoint. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Kyle Evans <[email protected]> Closes #9833
* Change http://zfsonlinux.org links to https://zfsonlinux.orgBrian Behlendorf2020-01-133-3/+3
| | | | | | | | | | | Update the project website links contained in to repository to reference the secure https://zfsonlinux.org address. Reviewed-By: Richard Laager <[email protected]> Reviewed-by: George Melikov <[email protected]> Reviewed-by: Garrett Fields <[email protected]> Reviewed-by: Kjeld Schouten <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #9837
* Add 'zfs send --saved' flagTom Caputi2020-01-102-22/+141
| | | | | | | | | | | | | | | | | | This commit adds the --saved (-S) to the 'zfs send' command. This flag allows a user to send a partially received dataset, which can be useful when migrating a backup server to new hardware. This flag is compatible with resumable receives, so even if the saved send is interrupted, it can be resumed. The flag does not require any user / kernel ABI changes or any new feature flags in the send stream format. Reviewed-by: Paul Dagnelie <[email protected]> Reviewed-by: Alek Pinchuk <[email protected]> Reviewed-by: Paul Zuchowski <[email protected]> Reviewed-by: Christian Schwarz <[email protected]> Reviewed-by: Matt Ahrens <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Tom Caputi <[email protected]> Closes #9007
* libspl: declare aok extern in headerNick Black2019-12-262-1/+3
| | | | | | | | | | Rather than defining a new instance of 'aok' in every compilation unit which includes this header, there is a single instance defined in zone.c, and the header now only declares an extern. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Paul Zuchowski <[email protected]> Signed-off-by: Nick Black <[email protected]> Closes #9752
* Colorize zpool status outputTony Hutter2019-12-191-0/+92
| | | | | | | | | | | | | | | | | | | If the ZFS_COLOR env variable is set, then use ANSI color output in zpool status: - Column headers are bold - Degraded or offline pools/vdevs are yellow - Non-zero error counters and faulted vdevs/pools are red - The 'status:' and 'action:' sections are yellow if they're displaying a warning. This also includes a new 'faketty' function in libtest.shlib that is compatible with FreeBSD (code provided by @freqlabs). Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Tony Hutter <[email protected]> Closes #9340
* cppcheck: (error) Memory leak: vtocBrian Behlendorf2019-12-181-14/+19
| | | | | | | | | | | | | | | | | Resolve the reported memory leak by using a dedicated local vptr variable to store the pointer reported by calloc(). Only assign the passed **vtoc function argument on success, in all other cases vptr is freed. [lib/libefi/rdwr_efi.c:403]: (error) Memory leak: vtoc [lib/libefi/rdwr_efi.c:422]: (error) Memory leak: vtoc [lib/libefi/rdwr_efi.c:440]: (error) Memory leak: vtoc [lib/libefi/rdwr_efi.c:454]: (error) Memory leak: vtoc [lib/libefi/rdwr_efi.c:470]: (error) Memory leak: vtoc Reviewed-by: Tony Hutter <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #9732
* Add FreeBSD jail support hooksMatthew Macy2019-12-111-2/+2
| | | | | | | | | | | | Add the 'zfs jail/unjail' subcommands along with the relevant documentation from FreeBSD. This feature is not supported on Linux and still requires the match kernel ioctls which will be included when the FreeBSD platform code is integrated. Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Signed-off-by: Ryan Moeller <[email protected]> Closes #9686
* Replace ASSERTV macro with compiler annotationMatthew Macy2019-12-052-2/+4
| | | | | | | | | | | Remove the ASSERTV macro and handle suppressing unused compiler warnings for variables only in ASSERTs using the __attribute__((unused)) compiler annotation. The annotation is understood by both gcc and clang. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9671
* Refactor zfs_context.h to build on FreeBSDMatthew Macy2019-12-043-9/+4
| | | | | | | | | | | | - on Linux move Linux specific headers to zfs_context_os.h - on FreeBSD move FreeBSD specific definitions to zfs_context_os.h - remove duplicate tsd_ definitions - remove unused AT_TYPE Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Don Brady <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9668
* Increase allowed 'special_small_blocks' maximum valueBrian Behlendorf2019-12-031-4/+13
| | | | | | | | | | | | | | There may be circumstances where it's desirable that all blocks in a specified dataset be stored on the special device. Relax the artificial 128K limit and allow the special_small_blocks property to be set up to 1M. When blocks >1MB have been enabled via the zfs_max_recordsize module option, this limit is increased accordingly. Reviewed-by: Don Brady <[email protected]> Reviewed-by: Kjeld Schouten <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #9131 Closes #9355
* Make asm-x86_64/atomic.S build on FreeBSDMatthew Macy2019-11-301-1/+5
| | | | | | | | Include the required headers for FreeBSD. Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9634
* Mark Linux fallocate extensions as specific to LinuxMatthew Macy2019-11-301-0/+4
| | | | | | | | | fallocate(2) is a Linux-specific system call which in unavailable on other platforms. Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9633
* Implement -A (ignore ASSERTs) for zdbPaul Zuchowski2019-11-272-1/+10
| | | | | | | | | | | | The command line switch -A (ignore ASSERTs) has always been available in zdb but was never connected up to the correct global variable. There are times when you need zdb to ignore asserts and keep dumping out whatever information it can get despite the ASSERT(s) failing. It was always intended to be part of zdb but was incomplete. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Paul Zuchowski <[email protected]> Closes #9610
* Add zfs_file_* interface, remove vnodesMatthew Macy2019-11-213-284/+376
| | | | | | | | | | | | | | | | | | Provide a common zfs_file_* interface which can be implemented on all platforms to perform normal file access from either the kernel module or the libzpool library. This allows all non-portable vnode_t usage in the common code to be replaced by the new portable zfs_file_t. The associated vnode and kobj compatibility functions, types, and macros have been removed from the SPL. Moving forward, vnodes should only be used in platform specific code when provided by the native operating system. Reviewed-by: Sean Eric Fagan <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9556
* Remove inappropiate error message suggesting to use '-r'InsanePrawn2019-11-151-4/+2
| | | | | | | | | | Removes an incorrect error message from libzfs that suggests applying '-r' when a zfs subcommand is called with a filesystem path while expecting either a snapshot or bookmark path. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Kjeld Schouten <[email protected]> Signed-off-by: InsanePrawn <[email protected]> Closes #9574
* Fix `zpool create -o <property>` error messageBrian Behlendorf2019-11-131-2/+0
| | | | | | | | | | | | | | | | | | When `zpool create -o <property>` is run without root permissions and the pool property requested is not specifically enumerated in zpool_valid_proplist(). Then an incorrect error message referring to an invalid property is printed rather than the expected permission denied error. Specifying a pool property at create time should be handled the same way as filesystem properties in zfs_valid_proplist(). There should not be default zfs_error_aux() set for properties which are not listed. Reviewed-by: loli10K <[email protected]> Reviewed-by: Kjeld Schouten <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #9550 Closes #9568
* Add wrapper stub for zfs_cmd ioctl to libzpoolMatthew Macy2019-11-125-8/+41
| | | | | | | | | | FreeBSD needs a wrapper for handling zfs_cmd ioctls. In libzfs this is handled by zfs_ioctl. However, here we need to wrap the call directly. Reviewed-by: Allan Jude <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9511
* Allow platform dependent path stripping for vdevsRyan Moeller2019-11-112-76/+84
| | | | | | | | | | | | On Linux the full path preceding devices is stripped when formatting vdev names. On FreeBSD we only want to strip "/dev/". Hide the implementation details of path stripping behind zfs_strip_path(). Make zfs_strip_partition_path() static in Linux implementation while here, since it is never used outside of the file it is defined in. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ryan Moeller <[email protected]> Closes #9565
* Preliminary support for RV64GRomain Dolbeau2019-11-062-2/+21
| | | | | | | This adds basic support for RISC-V, specifically RV64G. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Romain Dolbeau <[email protected]> Closes #9540
* Enable use of DTRACE_PROBE* macros in "spl" modulePrakash Surya2019-11-013-1/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | This change modifies some of the infrastructure for enabling the use of the DTRACE_PROBE* macros, such that we can use tehm in the "spl" module. Currently, when the DTRACE_PROBE* macros are used, they get expanded to create new functions, and these dynamically generated functions become part of the "zfs" module. Since the "spl" module does not depend on the "zfs" module, the use of DTRACE_PROBE* in the "spl" module would result in undefined symbols being used in the "spl" module. Specifically, DTRACE_PROBE* would turn into a function call, and the function being called would be a symbol only contained in the "zfs" module; which results in a linker and/or runtime error. Thus, this change adds the necessary logic to the "spl" module, to mirror the tracing functionality available to the "zfs" module. After this change, we'll have a "trace_zfs.h" header file which defines the probes available only to the "zfs" module, and a "trace_spl.h" header file which defines the probes available only to the "spl" module. Reviewed by: Brad Lewis <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Prakash Surya <[email protected]> Closes #9525
* Move sha2.h to platform codeMatthew Macy2019-10-312-0/+152
| | | | | | | | | FreeBSD has its own sha routines that the port uses. Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9530
* Add wrapper for Linux BLKFLSBUF ioctlMatthew Macy2019-10-281-0/+6
| | | | | | | | | | FreeBSD has no analog. Buffered block devices were removed a decade plus ago. Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9508
* Fix zpool history unbounded memory usageChunwei Chen2019-10-281-9/+11
| | | | | | | | | | | | In original implementation, zpool history will read the whole history before printing anything, causing memory usage goes unbounded. We fix this by breaking it into read-print iterations. Reviewed-by: Tom Caputi <[email protected]> Reviewed-by: Matt Ahrens <[email protected]> Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Chunwei Chen <[email protected]> Closes #9516
* Remove unneeded header from libzpool/kernel.cMatthew Macy2019-10-262-36/+0
| | | | | | | | | The sys/signal.h header doesn't exist on FreeBSD, nor is it needed on Linux. Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9510
* Fix header guard typoMatthew Macy2019-10-261-1/+1
| | | | | | | Fix header guard typo accidentally introduced by #9497. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9514
* Remove unused headers from uu_string.cMatthew Macy2019-10-251-2/+0
| | | | | | | | The malloc.h include is gratuitous and runs in to the following error on FreeBSD: Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9509
* Move platform dependent errno aliasesMatthew Macy2019-10-253-1/+12
| | | | | | | | | EBADE, EBADR, and ENOANO do not exist on FreeBSD The libspl errno.h is similarly platform dependent. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9498
* Remove sdt.hMatthew Macy2019-10-251-0/+23
| | | | | | | | | It's mostly a noop on ZoL and it conflicts with platforms that support dtrace. Remove this header to resolve the conflict. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9497
* Fix incremental recursive encrypted receiveTom Caputi2019-10-241-5/+17
| | | | | | | | | | | | | | | | | | | | Currently, incremental recursive encrypted receives fail to work for any snapshot after the first. The reason for this is because the check in zfs_setup_cmdline_props() did not properly realize that when the user attempts to use '-x encryption' in this situation, they are not really overriding the existing encryption property and instead are attempting to prevent it from changing. This resulted in an error message stating: "encryption property 'encryption' cannot be set or excluded for raw or incremental streams". This problem is fixed by updating the logic to expect this use case. Reviewed-by: loli10K <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Igor Kozhukhov <[email protected]> Signed-off-by: Tom Caputi <[email protected]> Closes #9494
* Linux 4.14, 4.19, 5.0+ compat: SIMD save/restoreBrian Behlendorf2019-10-241-1/+2
| | | | | | | | | | | | | | | | | | | | Contrary to initial testing we cannot rely on these kernels to invalidate the per-cpu FPU state and restore the FPU registers. Nor can we guarantee that the kernel won't modify the FPU state which we saved in the task struck. Therefore, the kfpu_begin() and kfpu_end() functions have been updated to save and restore the FPU state using our own dedicated per-cpu FPU state variables. This has the additional advantage of allowing us to use the FPU again in user threads. So we remove the code which was added to use task queues to ensure some functions ran in kernel threads. Reviewed-by: Fabian Grünbichler <[email protected]> Reviewed-by: Tony Hutter <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Issue #9346 Closes #9403
* Use zfs_ioctl with zfs_cmd_t in libzfsMatthew Macy2019-10-237-27/+27
| | | | | | | | | Consistently use the `zfs_ioctl()` wrapper since `ioctl()` cannot be called directly due to differing semantics between platforms. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9492
* Use platform independent error code for libzfs_run_process_implMatthew Macy2019-10-231-1/+1
| | | | | | Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9493
* Use correct format string when printing int8Matthew Macy2019-10-201-1/+1
| | | | | | | | Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Reviewed-by: George Melikov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9486
* OpenZFS restructuring - ARC memory pressureMatthew Macy2019-10-181-0/+1
| | | | | | | | | | | Factor Linux specific memory pressure handling out of ARC. Each platform will have different available interfaces for managing memory pressure. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9472
* Move zio_delay_interrupt to platform codeMatthew Macy2019-10-131-0/+1
| | | | | | | | FreeBSD has its own implementation as do other platforms. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9439
* Modify sharenfs=on default behaviorBrian Behlendorf2019-10-131-2/+3
| | | | | | | | | | | | While it may sometimes be convenient to export an NFS filesystem with no_root_squash it should not be the default behavior. Align the default behavior with the Linux NFS server defaults. To restore the previous behavior use 'zfs set sharenfs="no_root_squash,..."'. Reviewed-by: loli10K <[email protected]> Reviewed-by: Richard Laager <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #9397 Closes #9425
* Implement ZPOOL_IMPORT_UDEV_TIMEOUT_MSRichard Yao2019-10-111-1/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since 0.7.0, zpool import would unconditionally block on udev for 30 seconds. This introduced a regression in initramfs environments that lack udev (particularly mdev based environments), yet use a zfs userland tools intended for the system that had been built against udev. Gentoo's genkernel is the main example, although custom user initramfs environments would be similarly impacted unless special builds of the ZFS userland utilities were done for them. Such environments already have their own mechanisms for blocking until device nodes are ready (such as genkernel's scandelay parameter), so it is unnecessary for zpool import to block on a non-existent udev until a timeout is reached inside of them. Rather than trying to intelligently determine whether udev is available on the system to avoid unnecessarily blocking in such environments, it seems best to just allow the environment to override the timeout. I propose that we add an environment variable called ZPOOL_IMPORT_UDEV_TIMEOUT_MS. Setting it to 0 would restore the 0.6.x behavior that was more desirable in mdev based initramfs environments. This allows the system user land utilities to be reused when building mdev-based initramfs archives. Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Georgy Yakovlev <[email protected]> Signed-off-by: Richard Yao <[email protected]> Closes #9436
* Reduce loaded range tree memory usagePaul Dagnelie2019-10-095-5/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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 <[email protected]> Reviewed-by: Matt Ahrens <[email protected]> Reviewed by: Sebastien Roy [email protected] Reviewed-by: Igor Kozhukhov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Paul Dagnelie <[email protected]> Closes #9181
* OpenZFS restructuring - libzfsMatthew Macy2019-10-0310-797/+980
| | | | | | | | | Factor Linux specific functionality out of libzfs. Reviewed-by: Allan Jude <[email protected]> Reviewed-by: Jorgen Lundman <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matthew Macy <[email protected]> Closes #9377
* OpenZFS restructuring - libzutilMatthew Macy2019-10-036-1321/+1501
| | | | | | | | Factor Linux specific functionality out of libzutil. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Signed-off-by: Ryan Moeller <[email protected]> Closes #9356
* OpenZFS restructuring - libsplMatthew Macy2019-10-0236-307/+103
| | | | | | | | | Factor Linux specific pieces out of libspl. Reviewed-by: Ryan Moeller <[email protected]> Reviewed-by: Sean Eric Fagan <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9336
* OpenZFS restructuring - zpoolMatthew Macy2019-09-301-0/+15
| | | | | | | | | | | Factor Linux specific functions out of the zpool command. Reviewed-by: Allan Jude <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Reviewed-by: Sean Eric Fagan <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: loli10K <[email protected]> Signed-off-by: Matt Macy <[email protected]> Closes #9333
* Fix encryption hierarchy issues with zfs recv -dTom Caputi2019-09-251-44/+40
| | | | | | | | | | | | | | | | | | | | | Currently, the recv_fix_encryption_hierarchy() function accepts 'destsnap' as one of its parameters. Originally, this was intended to be the top-level dataset of a receive (whether or not the receive was recursive). Unfortunately, this parameter actually is simply the input that is passed in from the command line. When the user specifies 'zfs recv -d', this string is actually only the name of the receiving pool since the rest of the name is derived from the send stream. This causes the function to fail, leaving some datasets with an invalid encryption hierarchy. This patch resolves this problem by passing in the top_zfs variable instead. In order to make this work, this patch also includes some changes that ensure the value is always present when we need it. Reviewed-by: loli10K <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Tom Caputi <[email protected]> Closes #9273 Closes #9309
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 15:29:08 +0000