- 18 Nov, 2019 40 commits
-
-
Nikolay Borisov authored
The old name name was an awful misnomer because it didn't really find the oldest super backup per-se but rather its slot. For example if we have: slot0: gen - 2 slot1: gen - 1 slot2: gen slot3: empty init_backup_root_slot will return slot3 and not slot0. The new name is more appropriate since the function doesn't care whether there is a valid backup in the returned slot or not. Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
This function has been superseded by previous commits and is no longer used so just remove it. Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
Since the filesystem is not well formed and no trees are loaded it's pointless holding the objectid_mutex. Just remove its usage. Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
The code responsible for reading and initializing tree roots is scattered in open_ctree among 2 labels, emulating a loop. This is rather confusing to reason about. Instead, factor the code to a new function, init_tree_roots which implements the same logical flow. There are a couple of notable differences, namely: * Instead of using next_backup_root it's using the newly introduced read_backup_root. * If read_backup_root returns an error init_tree_roots propagates the error and there is no special handling of that case e.g. the code jumps straight to 'fail_tree_roots' label. The old code, however, was (erroneously) jumping to 'fail_block_groups' label if next_backup_root did fail, this was unnecessary since the tree roots init logic doesn't modify the state of block groups. Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
This function will replace next_root_backup with a much saner/cleaner interface. Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
It's no longer needed following cleanups around find_newest_backup_root Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
Backup roots are always written in a circular manner. By definition we can only ever have 1 backup root whose generation equals to that of the superblock. Hence, the 'if' in the for loop will trigger at most once. This is sufficient to return the newest backup root. Furthermore the newest_gen parameter is always set to the generation of the superblock. This value can be obtained from the fs_info. This patch removes the unnecessary code dealing with the wraparound case and makes 'newest_gen' a local variable. Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Filipe Manana authored
The inode delalloc mutex was added a long time ago by commit f248679e ("Btrfs: add a delalloc mutex to inodes for delalloc reservations"), and the reason for its introduction is not very clear from the change log. It claims it solves bogus warnings from lockdep, however it lacks an example report/warning from lockdep, or any explanation. Since we have enough concurrentcy protection from the locks of the space info and block reserve objects, and such lockdep warnings don't seem to exist anymore (at least on a 5.3 kernel I couldn't get them with fstests, ltp, fs_mark, etc), remove it, simplifying things a bit and decreasing the size of the btrfs_inode structure. With some quick fio tests doing direct IO and mmap writes I couldn't observe any significant performance increase either (direct IO writes that don't increase the file's size don't hold the inode's lock for their entire duration and mmap writes don't hold the inode's lock at all), which are the only type of writes that could see any performance gain due to less serialization. Review feedback from Josef: The problem was taking the i_mutex in mmap, which is how I was protecting delalloc reservations originally. The delalloc mutex didn't come with all of the other dependencies. That's what the lockdep messages were about, removing the lock isn't going to make them appear again. We _had_ to lock around this because we used to do tricks to keep from over-reserving, and if we didn't serialize delalloc reservations we'd end up with ugly accounting problems when we tried to clean things up. However with my recentish changes this isn't the case anymore. Every operation is responsible for reserving its space, and then adding it to the inode. Then cleaning up is straightforward and can't be mucked up by other users. So we no longer need the delalloc mutex to safe us from ourselves. Reviewed-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Filipe Manana authored
It is not used anymore since commit 957780eb ("Btrfs: introduce ticketed enospc infrastructure"), so just remove it. Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
In btrfs_statfs() we cache fs_info::space_info in a local variable only to use it once in a list_for_each_rcu() statement. Not only is the local variable unnecessary it even makes the code harder to follow as it's not clear which list it is iterating. Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
The on-disk format of block group item makes use of the key that stores the offset and length. This is further used in the code, although this makes thing harder to understand. The key is also packed so the offset/length is not properly aligned as u64. Add start (key.objectid) and length (key.offset) members to block group and remove the embedded key. When the item is searched or written, a local variable for key is used. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Accessors defined by BTRFS_SETGET_FUNCS take a raw extent buffer and manipulate the items there, there's no special prefix required. The block group accessors had _disk_ because previously the names were occupied by the on-stack accessors. As this has been addressed in the previous patch, we can now unify the naming. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
All accessors defined by BTRFS_SETGET_STACK_FUNCS contain _stack_ in the name, the block group ones were not following that scheme, so let's switch them. Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
The members ::used and ::flags are now in the block group cache structure, the last one is chunk_objectid, but that's set to a fixed value and otherwise unused. The item is constructed from a local variable before write, so we can remove the embedded one from block group. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
The flags are read from the item that's embedded to block group struct, but the item will be removed. Use the ::flags after read and before write. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
For unknown reasons, the member 'used' in the block group struct is stored in the b-tree item and accessed everywhere using the special accessor helper. Let's unify it and make it a regular member and only update the item before writing it to the tree. The item is still being used for flags and chunk_objectid, there's some duplication until the item is removed in following patches. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Qu Wenruo authored
The last user of btrfs_bio::flags was removed in commit 326e1dbb ("block: remove management of bi_remaining when restoring original bi_end_io"), remove it. (Tagged for stable as the structure is heavily used and space savings are desirable.) CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Qu Wenruo <wqu@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Add blake2b (with 256 bit digest) to the list of possible checksumming algorithms used by BTRFS. Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Currently all the checksum algorithms generate a fixed size digest size and we use it. The on-disk format can hold up to BTRFS_CSUM_SIZE bytes and BLAKE2b produces digest of 512 bits by default. We can't do that and will use the blake2b-256, this needs to be passed to the crypto API. Separate that from the base algorithm name and add a member to request specific driver, in this case with the digest size. The only place that uses the driver name is the crypto API setup. Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
Show the used driver for the checksum algorithm for the filesystem in sysfs file /sys/fs/btrfs/UUID/features/checksum, eg. crc32c (crc32c-generic) Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Export supported checksum algorithms via sysfs in the list of static features: /sys/fs/btrfs/features/supported_checksums Space spearated list of checksum algorithm names. Co-developed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
Add sha256 to the list of possible checksumming algorithms used by BTRFS. Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
Add xxhash64 to the list of possible checksumming algorithms used by BTRFS. Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
To remove use of extent_map::bdev we need to find a replacement, and the latest_bdev is the only one we can use here, because inode::i_bdev and superblock::s_bdev are NULL. The DIO code uses bdev in two places: * to read blocksize to perform alignment checks in do_blockdev_direct_IO, but we do them in btrfs code before any call to DIO * in the following call chain: do_direct_IO get_more_blocks sdio->get_block() <-- this is btrfs_get_blocks_direct subsequently the map_bh->b_dev member is used in clean_bdev_aliases and dio_new_bio to set the bio's bdev to that of the buffer_head. However, because we have provided a submit function dio_bio_submit calls our submission function and ignores the bdev. So it's safe to pass any valid bdev that's used within the filesystem. Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
This is a preparatory patch for removing extent_map::bdev. There's some history behind the code so this is only precaution to catch if things break before the actual removal happens. Logically, comparing a raw low-level block device (bdev) does not make sense for extent maps (high-level objects). This had no effect in practice but was quite confusing in the code. The lookup_map is set iff EXTENT_FLAG_FS_MAPPING is set. The two pointers were stored in the same bytes and used potentially in two meanings. Now they're split, so the asserts are in place to check that the condition will not change. The lookup map pointer misused bdev, this has been changed in commit 95617d69 ("btrfs: cleanup, stop casting for extent_map->lookup everywhere") to the explicit type. But the semantics hasn't changed and bdev was not actually used to decide if maps are mergeable. Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
Instead of checking if we've read a BTRFS_CHUNK_ITEM_KEY from disk and then process it we could just bail out early if the read disk key wasn't a BTRFS_CHUNK_ITEM_KEY. This removes a level of indentation and makes the code nicer to read. Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
In btrfs_may_alloc_data_chunk() we're checking if the chunk type is of type BTRFS_BLOCK_GROUP_DATA and if it is we process it. Instead of checking if the chunk type is a BTRFS_BLOCK_GROUP_DATA chunk we can negate the check and bail out early if it isn't. This makes the code a bit more readable. Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
In lock_stripe_add() we're caching the bucket for the stripe hash table just for a single call to dereference the stripe hash. If we just directly call rbio_bucket() we can safe the pointless local variable. Also move the dereferencing of the stripe hash outside of the variable declaration block to not break over the 80 characters limit. Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Johannes Thumshirn authored
In lock_stripe_add() we're traversing the stripe hash list and check if the current list element's raid_map equals is equal to the raid bio's raid_map. If both are equal we continue processing. If we'd check for inequality instead of equality we can reduce one level of indentation. Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
We don't modify the data passed to tracepoints, some of the declarations are already const, add it to the rest. Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Remove typecasts from trace printk, adjust types and move typecast to the assignment if necessary. When assigning, the types are more obvious compared to matching the variables to the format strings. Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
Instead of using an input pointer parameter as the return value and have an int as the return type of find_desired_extent, rework the function to directly return the found offset. Doing that the 'ret' variable in btrfs_llseek_file can be removed. Additional (subjective) benefit is that btrfs' llseek function now resemebles those of the other major filesystems. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
Handle SEEK_END/SEEK_CUR in a single 'default' case by directly returning from generic_file_llseek. This makes the 'out' label redundant. Finally return directly the vale from vfs_setpos. No semantic changes. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
Nikolay Borisov authored
Modifying the file position is done on a per-file basis. This renders holding the inode lock for writing useless and makes the performance of concurrent llseek's abysmal. Fix this by holding the inode for read. This provides protection against concurrent truncates and find_desired_extent already includes proper extent locking for the range which ensures proper locking against concurrent writes. SEEK_CUR and SEEK_END can be done lockessly. The former is synchronized by file::f_lock spinlock. SEEK_END is not synchronized but atomic, but that's OK since there is not guarantee that SEEK_END will always be at the end of the file in the face of tail modifications. This change brings ~82% performance improvement when doing a lot of parallel fseeks. The workload essentially does: for (d=0; d<num_seek_read; d++) { /* offset %= 16777216; */ fseek (f, 256 * d % 16777216, SEEK_SET); fread (buffer, 64, 1, f); } Without patch: num workprocesses = 16 num fseek/fread = 8000000 step = 256 fork 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 real 0m41.412s user 0m28.777s sys 2m16.510s With patch: num workprocesses = 16 num fseek/fread = 8000000 step = 256 fork 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 real 0m11.479s user 0m27.629s sys 0m21.040s Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
We can infer the ops from the type that is now passed to all functions that would need it, this makes workspace_manager::ops redundant and can be removed. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Replace indirect calls to free_workspace by switch and calls to the specific callbacks. This is mainly to get rid of the indirection due to spectre vulnerability mitigations. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
We can infer the workspace_manager from type and the type will be used in the following patch to call a common helper for free_workspace. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Replace indirect calls to alloc_workspace by switch and calls to the specific callbacks. This is mainly to get rid of the indirection due to spectre vulnerability mitigations. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
We can infer the workspace_manager from type and the type will be used in the following patch to call a common helper for alloc_workspace. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-
David Sterba authored
Similar to get_workspace, majority of the callbacks is trivial, we don't gain anything by the indirection, so replace them by a switch function. Trivial callback implementations use the helper. Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
-