- 16 May, 2022 40 commits
-
-
Filipe Manana authored
When doing a NOWAIT direct IO write, if we can NOCOW then it means we can proceed with the non-blocking, NOWAIT path. However reserving the metadata space and qgroup meta space can often result in blocking - flushing delalloc, wait for ordered extents to complete, trigger transaction commits, etc, going against the semantics of a NOWAIT write. So make the NOWAIT write path to try to reserve all the metadata it needs without resulting in a blocking behaviour - if we get -ENOSPC or -EDQUOT then return -EAGAIN to make the caller fallback to a blocking direct IO write. This is part of a patchset comprised of the following patches: btrfs: avoid blocking on page locks with nowait dio on compressed range btrfs: avoid blocking nowait dio when locking file range btrfs: avoid double nocow check when doing nowait dio writes btrfs: stop allocating a path when checking if cross reference exists btrfs: free path at can_nocow_extent() before checking for checksum items btrfs: release path earlier at can_nocow_extent() btrfs: avoid blocking when allocating context for nowait dio read/write btrfs: avoid blocking on space revervation when doing nowait dio writes The following test was run before and after applying this patchset: $ cat io-uring-nodatacow-test.sh #!/bin/bash DEV=/dev/sdc MNT=/mnt/sdc MOUNT_OPTIONS="-o ssd -o nodatacow" MKFS_OPTIONS="-R free-space-tree -O no-holes" NUM_JOBS=4 FILE_SIZE=8G RUN_TIME=300 cat <<EOF > /tmp/fio-job.ini [io_uring_rw] rw=randrw fsync=0 fallocate=posix group_reporting=1 direct=1 ioengine=io_uring iodepth=64 bssplit=4k/20:8k/20:16k/20:32k/10:64k/10:128k/5:256k/5:512k/5:1m/5 filesize=$FILE_SIZE runtime=$RUN_TIME time_based filename=foobar directory=$MNT numjobs=$NUM_JOBS thread EOF echo performance | \ tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor umount $MNT &> /dev/null mkfs.btrfs -f $MKFS_OPTIONS $DEV &> /dev/null mount $MOUNT_OPTIONS $DEV $MNT fio /tmp/fio-job.ini umount $MNT The test was run a 12 cores box with 64G of ram, using a non-debug kernel config (Debian's default config) and a spinning disk. Result before the patchset: READ: bw=407MiB/s (427MB/s), 407MiB/s-407MiB/s (427MB/s-427MB/s), io=119GiB (128GB), run=300175-300175msec WRITE: bw=407MiB/s (427MB/s), 407MiB/s-407MiB/s (427MB/s-427MB/s), io=119GiB (128GB), run=300175-300175msec Result after the patchset: READ: bw=436MiB/s (457MB/s), 436MiB/s-436MiB/s (457MB/s-457MB/s), io=128GiB (137GB), run=300044-300044msec WRITE: bw=435MiB/s (456MB/s), 435MiB/s-435MiB/s (456MB/s-456MB/s), io=128GiB (137GB), run=300044-300044msec That's about +7.2% throughput for reads and +6.9% for writes. Signed-off-by:Filipe Manana <fdmanana@suse.com> Signed-off-by:
David Sterba <dsterba@suse.com>
-
Filipe Manana authored
When doing a NOWAIT direct IO read/write, we allocate a context object (struct btrfs_dio_data) with GFP_NOFS, which can result in blocking waiting for memory allocation (GFP_NOFS is __GFP_RECLAIM | __GFP_IO). This is undesirable for the NOWAIT semantics, so do the allocation with GFP_NOWAIT if we are serving a NOWAIT request and if the allocation fails return -EAGAIN, so that the caller can fallback to a blocking context and retry with a non-blocking write. Signed-off-by:
-
Filipe Manana authored
At can_nocow_extent(), we are releasing the path only after checking if the block group that has the target extent is read only, and after checking if there's delalloc in the range in case our extent is a preallocated extent. The read only extent check can be expensive if we have a very large filesystem with many block groups, as well as the check for delalloc in the inode's io_tree in case the io_tree is big due to IO on other file ranges. Our path is holding a read lock on a leaf and there's no need to keep the lock while doing those two checks, so release the path before doing them, immediately after the last use of the leaf. Signed-off-by:
-
Filipe Manana authored
When we look for checksum items, through csum_exist_in_range(), at can_nocow_extent(), we no longer need the path that we have previously allocated. Through csum_exist_in_range() -> btrfs_lookup_csums_range(), we also end up allocating a path, so we are adding unnecessary extra memory usage. So free the path before calling csum_exist_in_range(). Signed-off-by:
-
Filipe Manana authored
At btrfs_cross_ref_exist() we always allocate a path, but we really don't need to because all its callers (only 2) already have an allocated path that is not being used when they call btrfs_cross_ref_exist(). So change btrfs_cross_ref_exist() to take a path as an argument and update both its callers to pass in the unused path they have when they call btrfs_cross_ref_exist(). Signed-off-by:
-
Filipe Manana authored
When doing a NOWAIT direct IO write we are checking twice if we can COW into the target file range using can_nocow_extent() - once at the very beginning of the write path, at btrfs_write_check() via check_nocow_nolock(), and later again at btrfs_get_blocks_direct_write(). The can_nocow_extent() function does a lot of expensive things - searching for the file extent item in the inode's subvolume tree, searching for the extent item in the extent tree, checking delayed references, etc, so it isn't a very cheap call. We can remove the first check at btrfs_write_check(), and add there a quick check to verify if the inode has the NODATACOW or PREALLOC flags, and quickly bail out if it doesn't have neither of those flags, as that means we have to COW and therefore can't comply with the NOWAIT semantics. After this we do only one call to can_nocow_extent(), while we are at btrfs_get_blocks_direct_write(), where we have already locked the file range and we did a try lock on the range before, at btrfs_dio_iomap_begin() (since the previous patch in the series). Signed-off-by:
-
Filipe Manana authored
If we are doing a NOWAIT direct IO read/write, we can block when locking the file range at btrfs_dio_iomap_begin(), as it's possible the range (or a part of it) is already locked by another task (mmap writes, another direct IO read/write racing with us, fiemap, etc). We are also waiting for completion of any ordered extent we find in the range, which also can block us for a significant amount of time. There's also the incorrect fallback to buffered IO (returning -ENOTBLK) when we are dealing with a NOWAIT request and we can't proceed. In this case we should be returning -EAGAIN, as falling back to buffered IO can result in blocking for many different reasons, so that the caller can delegate a retry to a context where blocking is more acceptable. Fix these cases by: 1) Doing a try lock on the file range and failing with -EAGAIN if we can not lock right away; 2) Fail with -EAGAIN if we find an ordered extent; 3) Return -EAGAIN instead of -ENOTBLK when we need to fallback to buffered IO and we have a NOWAIT request. This will also allow us to avoid a duplicated check that verifies if we are able to do a NOCOW write for NOWAIT direct IO writes, done in the next patch. Signed-off-by:
-
Filipe Manana authored
If we are doing NOWAIT direct IO read/write and our inode has compressed extents, we call filemap_fdatawrite_range() against the range in order to wait for compressed writeback to complete, since the generic code at iomap_dio_rw() calls filemap_write_and_wait_range() once, which is not enough to wait for compressed writeback to complete. This call to filemap_fdatawrite_range() can block on page locks, since the first writepages() on a range that we will try to compress results only in queuing a work to compress the data while holding the pages locked. Even though the generic code at iomap_dio_rw() will do the right thing and return -EAGAIN for NOWAIT requests in case there are pages in the range, we can still end up at btrfs_dio_iomap_begin() with pages in the range because either of the following can happen: 1) Memory mapped writes, as we haven't locked the range yet; 2) Buffered reads might have started, which lock the pages, and we do the filemap_fdatawrite_range() call before locking the file range. So don't call filemap_fdatawrite_range() at btrfs_dio_iomap_begin() if we are doing a NOWAIT read/write. Instead call filemap_range_needs_writeback() to check if there are any locked, dirty, or under writeback pages, and return -EAGAIN if that's the case. Signed-off-by:
-
Jonathan Lassoff authored
In order for end users to quickly react to new issues that come up in production, it is proving useful to leverage this printk indexing system. This printk index enables kernel developers to use calls to printk() with changeable ad-hoc format strings, while still enabling end users to detect changes and develop a semi-stable interface for detecting and parsing these messages. So that detailed Btrfs messages are captured by this printk index, this patch wraps btrfs_printk and btrfs_handle_fs_error with macros. Example of the generated list: https://lore.kernel.org/lkml/12588e13d51a9c3bf59467d3fc1ac2162f1275c1.1647539056.git.jof@thejof.comSigned-off-by:
Jonathan Lassoff <jof@thejof.com> Reviewed-by:
-
Qu Wenruo authored
Btrfs doesn't check whether the tree block respects the root owner. This means, if a tree block referred by a parent in extent tree, but has owner of 5, btrfs can still continue reading the tree block, as long as it doesn't trigger other sanity checks. Normally this is fine, but combined with the empty tree check in check_leaf(), if we hit an empty extent tree, but the root node has csum tree owner, we can let such extent buffer to sneak in. Shrink the hole by: - Do extra eb owner check at tree read time - Make sure the root owner extent buffer exactly matches the root id. Unfortunately we can't yet completely patch the hole, there are several call sites can't pass all info we need: - For reloc/log trees Their owner is key::offset, not key::objectid. We need the full root key to do that accurate check. For now, we just skip the ownership check for those trees. - For add_data_references() of relocation That call site doesn't have any parent/ownership info, as all the bytenrs are all from btrfs_find_all_leafs(). - For direct backref items walk Direct backref items records the parent bytenr directly, thus unlike indirect backref item, we don't do a full tree search. Thus in that case, we don't have full parent owner to check. For the later two cases, they all pass 0 as @owner_root, thus we can skip those cases if @owner_root is 0. Signed-off-by:
Qu Wenruo <wqu@suse.com> Reviewed-by:
-
Filipe Manana authored
We have four different scenarios where we don't expect to find ordered extents after locking a file range: 1) During plain fallocate; 2) During hole punching; 3) During zero range; 4) During reflinks (both cloning and deduplication). This is because in all these cases we follow the pattern: 1) Lock the inode's VFS lock in exclusive mode; 2) Lock the inode's i_mmap_lock in exclusive node, to serialize with mmap writes; 3) Flush delalloc in a file range and wait for all ordered extents to complete - both done through btrfs_wait_ordered_range(); 4) Lock the file range in the inode's io_tree. So add a helper that asserts that we don't have ordered extents for a given range. Make the four scenarios listed above use this helper after locking the respective file range. Signed-off-by:
-
Filipe Manana authored
For hole punching and zero range we have this loop that checks if we have ordered extents after locking the file range, and if so unlock the range, wait for ordered extents, and retry until we don't find more ordered extents. This logic was needed in the past because: 1) Direct IO writes within the i_size boundary did not take the inode's VFS lock. This was because that lock used to be a mutex, then some years ago it was switched to a rw semaphore (commit 9902af79 ("parallel lookups: actual switch to rwsem")), and then btrfs was changed to take the VFS inode's lock in shared mode for writes that don't cross the i_size boundary (commit e9adabb9 ("btrfs: use shared lock for direct writes within EOF")); 2) We could race with memory mapped writes, because memory mapped writes don't acquire the inode's VFS lock. We don't have that race anymore, as we have a rw semaphore to synchronize memory mapped writes with fallocate (and reflinking too). That change happened with commit 8d9b4a16 ("btrfs: exclude mmap from happening during all fallocate operations"). So stop looking for ordered extents after locking the file range when doing hole punching and zero range operations. Signed-off-by:
-
Filipe Manana authored
When doing hole punching we are flushing delalloc and waiting for ordered extents to complete before locking the inode (VFS lock and the btrfs specific i_mmap_lock). This is fine because even if a write happens after we call btrfs_wait_ordered_range() and before we lock the inode (call btrfs_inode_lock()), we will notice the write at btrfs_punch_hole_lock_range() and flush delalloc and wait for its ordered extent. We can however make this simpler by locking first the inode an then call btrfs_wait_ordered_range(), which will allow us to remove the ordered extent lookup logic from btrfs_punch_hole_lock_range() in the next patch. It also makes the behaviour the same as plain fallocate, hole punching and reflinks. Signed-off-by:
-
Filipe Manana authored
For fallocate() we have this loop that checks if we have ordered extents after locking the file range, and if so unlock the range, wait for ordered extents, and retry until we don't find more ordered extents. This logic was needed in the past because: 1) Direct IO writes within the i_size boundary did not take the inode's VFS lock. This was because that lock used to be a mutex, then some years ago it was switched to a rw semaphore (commit 9902af79 ("parallel lookups: actual switch to rwsem")), and then btrfs was changed to take the VFS inode's lock in shared mode for writes that don't cross the i_size boundary (commit e9adabb9 ("btrfs: use shared lock for direct writes within EOF")); 2) We could race with memory mapped writes, because memory mapped writes don't acquire the inode's VFS lock. We don't have that race anymore, as we have a rw semaphore to synchronize memory mapped writes with fallocate (and reflinking too). That change happened with commit 8d9b4a16 ("btrfs: exclude mmap from happening during all fallocate operations"). So stop looking for ordered extents after locking the file range when doing a plain fallocate. Signed-off-by:
-
Filipe Manana authored
When starting a reflink operation we have these calls to inode_dio_wait() which used to be needed because direct IO writes that don't cross the i_size boundary did not take the inode's VFS lock, so we could race with them and end up with ordered extents in target range after calling btrfs_wait_ordered_range(). However that is not the case anymore, because the inode's VFS lock was changed from a mutex to a rw semaphore, by commit 9902af79 ("parallel lookups: actual switch to rwsem"), and several years later we started to lock the inode's VFS lock in shared mode for direct IO writes that don't cross the i_size boundary (commit e9adabb9 ("btrfs: use shared lock for direct writes within EOF")). So remove those inode_dio_wait() calls. Signed-off-by:
-
Filipe Manana authored
When starting a fallocate zero range operation, before getting the first extent map for the range, we make a call to inode_dio_wait(). This logic was needed in the past because direct IO writes within the i_size boundary did not take the inode's VFS lock. This was because that lock used to be a mutex, then some years ago it was switched to a rw semaphore (by commit 9902af79 ("parallel lookups: actual switch to rwsem")), and then btrfs was changed to take the VFS inode's lock in shared mode for writes that don't cross the i_size boundary (done in commit e9adabb9 ("btrfs: use shared lock for direct writes within EOF")). The lockless direct IO writes could result in a race with the zero range operation, resulting in the later getting a stale extent map for the range. So remove this no longer needed call to inode_dio_wait(), as fallocate takes the inode's VFS lock in exclusive mode and direct IO writes within i_size take that same lock in shared mode. Signed-off-by:
-
Filipe Manana authored
During a plain fallocate, we always start by reserving an amount of data space that matches the length of the range passed to fallocate. When we already have extents allocated in that range, we may end up trying to reserve a lot more data space then we need, which can result in several undesired behaviours: 1) We fail with -ENOSPC. For example the passed range has a length of 1G, but there's only one hole with a size of 1M in that range; 2) We temporarily reserve excessive data space that could be used by other operations happening concurrently; 3) By reserving much more data space then we need, we can end up doing expensive things like triggering dellaloc for other inodes, waiting for the ordered extents to complete, trigger transaction commits, allocate new block groups, etc. Example: $ cat test.sh #!/bin/bash DEV=/dev/sdj MNT=/mnt/sdj mkfs.btrfs -f -b 1g $DEV mount $DEV $MNT # Create a file with a size of 600M and two holes, one at [200M, 201M[ # and another at [401M, 402M[ xfs_io -f -c "pwrite -S 0xab 0 200M" \ -c "pwrite -S 0xcd 201M 200M" \ -c "pwrite -S 0xef 402M 198M" \ $MNT/foobar # Now call fallocate against the whole file range, see if it fails # with -ENOSPC or not - it shouldn't since we only need to allocate # 2M of data space. xfs_io -c "falloc 0 600M" $MNT/foobar umount $MNT $ ./test.sh (...) wrote 209715200/209715200 bytes at offset 0 200 MiB, 51200 ops; 0.8063 sec (248.026 MiB/sec and 63494.5831 ops/sec) wrote 209715200/209715200 bytes at offset 210763776 200 MiB, 51200 ops; 0.8053 sec (248.329 MiB/sec and 63572.3172 ops/sec) wrote 207618048/207618048 bytes at offset 421527552 198 MiB, 50688 ops; 0.7925 sec (249.830 MiB/sec and 63956.5548 ops/sec) fallocate: No space left on device $ So fix this by not allocating an amount of data space that matches the length of the range passed to fallocate. Instead allocate an amount of data space that corresponds to the sum of the sizes of each hole found in the range. This reservation now happens after we have locked the file range, which is safe since we know at this point there's no delalloc in the range because we've taken the inode's VFS lock in exclusive mode, we have taken the inode's i_mmap_lock in exclusive mode, we have flushed delalloc and waited for all ordered extents in the range to complete. This type of failure actually seems to happen in practice with systemd, and we had at least one report about this in a very long thread which is referenced by the Link tag below. Link: https://lore.kernel.org/linux-btrfs/bdJVxLiFr_PyQSXRUbZJfFW_jAjsGgoMetqPHJMbg-hdy54Xt_ZHhRetmnJ6cJ99eBlcX76wy-AvWwV715c3YndkxneSlod11P1hlaADx0s=@protonmail.com/Signed-off-by: -
Sweet Tea Dorminy authored
According to the tree checker, "all xattrs with a given objectid follow the inode with that objectid in the tree" is an invariant. This was broken by the recent change "btrfs: move common inode creation code into btrfs_create_new_inode()", which moved acl creation and property inheritance (stored in xattrs) to before inode insertion into the tree. As a result, under certain timings, the xattrs could be written to the tree before the inode, causing the tree checker to report violation of the invariant. Move property inheritance and acl creation back to their old ordering after the inode insertion. Suggested-by:
Omar Sandoval <osandov@osandov.com> Reported-by:
Naohiro Aota <naohiro.aota@wdc.com> Signed-off-by:
Sweet Tea Dorminy <sweettea-kernel@dorminy.me> Signed-off-by:
-
Omar Sandoval authored
All of our inode creation code paths duplicate the calls to btrfs_init_inode_security() and btrfs_add_link(). Subvolume creation additionally duplicates property inheritance and the call to btrfs_set_inode_index(). Fix this by moving the common code into btrfs_create_new_inode(). This accomplishes a few things at once: 1. It reduces code duplication. 2. It allows us to set up the inode completely before inserting the inode item, removing calls to btrfs_update_inode(). 3. It fixes a leak of an inode on disk in some error cases. For example, in btrfs_create(), if btrfs_new_inode() succeeds, then we have inserted an inode item and its inode ref. However, if something after that fails (e.g., btrfs_init_inode_security()), then we end the transaction and then decrement the link count on the inode. If the transaction is committed and the system crashes before the failed inode is deleted, then we leak that inode on disk. Instead, this refactoring aborts the transaction when we can't recover more gracefully. 4. It exposes various ways that subvolume creation diverges from mkdir in terms of inheriting flags, properties, permissions, and POSIX ACLs, a lot of which appears to be accidental. This patch explicitly does _not_ change the existing non-standard behavior, but it makes those differences more clear in the code and documents them so that we can discuss whether they should be changed. Reviewed-by:
Omar Sandoval <osandov@fb.com> Signed-off-by:
-
Omar Sandoval authored
The various inode creation code paths do not account for the compression property, POSIX ACLs, or the parent inode item when starting a transaction. Fix it by refactoring all of these code paths to use a new function, btrfs_new_inode_prepare(), which computes the correct number of items. To do so, it needs to know whether POSIX ACLs will be created, so move the ACL creation into that function. To reduce the number of arguments that need to be passed around for inode creation, define struct btrfs_new_inode_args containing all of the relevant information. btrfs_new_inode_prepare() will also be a good place to set up the fscrypt context and encrypted filename in the future. Reviewed-by:
-
Omar Sandoval authored
btrfs_{mknod,create,mkdir}() are now identical other than the inode initialization and some inconsequential function call order differences. Factor out the common code to reduce code duplication. Reviewed-by: -
Omar Sandoval authored
Instead of calling new_inode() and inode_init_owner() inside of btrfs_new_inode(), do it in the callers. This allows us to pass in just the inode instead of the mnt_userns and mode and removes the need for memalloc_nofs_{save,restores}() since we do it before starting a transaction. In create_subvol(), it also means we no longer have to look up the inode again to instantiate it. This also paves the way for some more cleanups in later patches. This also removes the comments about Smack checking i_op, which are no longer true since commit 5d6c3191 ("xattr: Add __vfs_{get,set,remove}xattr helpers"). Now it checks inode->i_opflags & IOP_XATTR, which is set based on sb->s_xattr. Signed-off-by: -
Qu Wenruo authored
Although we have btrfs_extent_buffer_leak_debug_check() (enabled by CONFIG_BTRFS_DEBUG option) to detect and warn QA testers that we have some extent buffer leakage, it's just pr_err(), not noisy enough for fstests to cache. So here we trigger a WARN_ON() if the allocated_ebs list is not empty. Reviewed-by:
-
Anand Jain authored
In the function btrfs_dev_replace_finishing, we dereferenced fs_info->fs_devices 6 times. Use keep local variable for that. Reviewed-by:
Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by:
Anand Jain <anand.jain@oracle.com> Reviewed-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
Marcos Paulo de Souza <mpdesouza@suse.com> Signed-off-by:
Gabriel Niebler <gniebler@suse.com> Reviewed-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
This function can be simplified by refactoring to use the new iterator macro. No functional changes. Signed-off-by:
-
Gabriel Niebler authored
There is a common pattern when searching for a key in btrfs: * Call btrfs_search_slot to find the slot for the key * Enter an endless loop: * If the found slot is larger than the no. of items in the current leaf, check the next leaf * If it's still not found in the next leaf, terminate the loop * Otherwise do something with the found key * Increment the current slot and continue To reduce code duplication, we can replace this code pattern with an iterator macro, similar to the existing for_each_X macros found elsewhere in the kernel. This also makes the code easier to understand for newcomers by putting a name to the encapsulated functionality. Signed-off-by: -
Qu Wenruo authored
Since the subpage support for scrub, one page no longer always represents one sector, thus scrub_bio::pagev and scrub_bio::sector_count are no longer accurate. Rename them to scrub_bio::sectors and scrub_bio::sector_count respectively. This also involves scrub_ctx::pages_per_bio and other macros involved. Now the renaming of pages involved in scrub is be finished. Signed-off-by:
-
Qu Wenruo authored
Since the subpage support of scrub, scrub_sector is in fact just representing one sector. Thus the name scrub_page is no longer correct, rename it to scrub_sector. This also involves the following renames: - spage -> sector Normally we would just replace "page" with "sector" and result something like "ssector". But the repeating 's' is not really eye friendly. So here we just simple use "sector", as there is nothing from MM layer called "sector" to cause any confusion. - scrub_parity::spages -> sectors_list Normally we use plural to indicate an array, not a list. Rename it to @sectors_list to be more explicit on the list part. - Also reformat and update comments that get changed Signed-off-by:
-
