Skip to contentSkip to conversation

Btrfs updates for 6.11

GIT PULL Merged David Sterba 5 participants 7 messages Jul 15, 2024 – Jul 19, 2024 Btrfs file system

lore
Share & tools
lore

David Sterba

author maintainer
#0 lore raw

Hi,

please pull the changes described below. The hilights are new logic behind background block group reclaim, automatic removal of qgroup after removing a subvolume and new 'rescue=' mount options. The rest is optimizations, cleanups and refactoring.

There's a merge conflict caused by the latency fixes from last week in extent_map.c:btrfs_scan_inode(), related commits 4e660ca3a98d931809734 and b3ebb9b7e92a928344a. Resolved in branch for-6.11-merged and that's been in linux-next for a few days.

User visible features:

- dynamic block group reclaim:
- tunable framework to avoid situations where eager data allocations prevent creating new metadata chunks due to lack of unallocated space
- reuse sysfs knob bg_reclaim_threshold (otherwise used only in zoned mode) for a fixed value threshold
- new on/off sysfs knob "dynamic_reclaim" calculating the value based on heuristics, aiming to keep spare working space for relocating chunks but not to needlessly relocate partially utilized block groups or reclaim newly allocated ones
- stats are exported in sysfs per block group type, files "reclaim_*"
- this may increase IO load at unexpected times but the corner case of no allocatable block groups is known to be worse

- automatically remove qgroup of deleted subvolumes
- adjust qgroup removal conditions, make sure all related subvolume data are already removed, or return EBUSY, also take into account setting of sysfs drop_subtree_threshold
- also works in squota mode

- mount option updates: new modes of 'rescue=' that allow to mount (as read-only) images that could have been partially converted by user space tools
- ignoremetacsums - invalid metadata checksums are ignored
- ignoresuperflags - super block flags that track conversion in progress (like UUID or checksums)

Core:

- size of struct btrfs_inode is now below 1024 (on a release config), improved memory packing and other secondary effects

- switch tracking of open inodes from rb-tree to xarray, minor performance improvement

- reduce number of empty transaction commits when there are no dirty data/metadata

- memory allocation optimizations (reduced numbers, reordering out of critical sections)

- extent map structure optimizations and refactoring, more sanity checks

- more subpage in zoned mode preparations or fixes

- general snapshot code cleanups, improvements and documentation

- tree-checker updates: more file extent ram_bytes fixes, continued

- raid-stripe-tree update (not backward compatible):
- remove extent encoding field from the structure, can be inferred from other information
- requires btrfs-progs 6.9.1 or newer

- cleanups and refactoring
- error message updates
- error handling improvements
- return type and parameter cleanups and improvements

----------------------------------------------------------------
The following changes since commit 256abd8e550ce977b728be79a74e1729438b4948:

Linux 6.10-rc7 (2024-07-07 14:23:46 -0700)

are available in the Git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git tags/for-6.11-tag

for you to fetch changes up to 8e7860543a94784d744c7ce34b78a2e11beefa5c:

btrfs: fix extent map use-after-free when adding pages to compressed bio (2024-07-11 16:32:22 +0200)

----------------------------------------------------------------
Anand Jain (7):
btrfs: drop bytenr_orig and fix comment in btrfs_scan_one_device()
btrfs: move btrfs_block_group_root() to block-group.c
btrfs: rename err to ret in btrfs_cleanup_fs_roots()
btrfs: rename ret to err in btrfs_recover_relocation()
btrfs: rename ret to ret2 in btrfs_recover_relocation()
btrfs: rename err to ret in btrfs_recover_relocation()
btrfs: rename err to ret in btrfs_drop_snapshot()

Boris Burkov (7):
btrfs: preallocate ulist memory for qgroup rsv
btrfs: report reclaim stats in sysfs
btrfs: store fs_info in space_info
btrfs: dynamic block_group reclaim threshold
btrfs: periodic block_group reclaim
btrfs: prevent pathological periodic reclaim loops
btrfs: urgent periodic reclaim pass

David Sterba (24):
btrfs: qgroup: do quick checks if quotas are enabled before starting ioctls
btrfs: remove duplicate name variable declarations
btrfs: rename macro local variables that clash with other variables
btrfs: use for-local variables that shadow function variables
btrfs: remove unused define EXTENT_SIZE_PER_ITEM
btrfs: keep const when returning value from get_unaligned_le8()
btrfs: constify parameters of write_eb_member() and its users
btrfs: simplify range parameters of btrfs_wait_ordered_roots()
btrfs: constify pointer parameters where applicable
btrfs: abort transaction if we don't find extref in btrfs_del_inode_extref()
btrfs: only print error message when checking item size in print_extent_item()
btrfs: abort transaction on errors in btrfs_free_chunk()
btrfs: qgroup: preallocate memory before adding a relation
btrfs: qgroup: warn about inconsistent qgroups when relation update fails
btrfs: pass a btrfs_inode to btrfs_readdir_put_delayed_items()
btrfs: pass a btrfs_inode to btrfs_readdir_get_delayed_items()
btrfs: pass a btrfs_inode to is_data_inode()
btrfs: switch btrfs_block_group::inode to struct btrfs_inode
btrfs: pass a btrfs_inode to btrfs_ioctl_send()
btrfs: switch btrfs_pending_snapshot::dir to btrfs_inode
btrfs: switch btrfs_ordered_extent::inode to struct btrfs_inode
btrfs: pass a btrfs_inode to btrfs_compress_heuristic()
btrfs: pass a btrfs_inode to btrfs_set_prop()
btrfs: enhance compression error messages

Filipe Manana (51):
btrfs: zoned: make btrfs_get_dev_zone() static
btrfs: remove no longer used btrfs_migrate_to_delayed_refs_rsv()
btrfs: fix misspelled end IO compression callbacks
btrfs: fix function name in comment for btrfs_remove_ordered_extent()
btrfs: use an xarray to track open inodes in a root
btrfs: preallocate inodes xarray entry to avoid transaction abort
btrfs: reduce nesting and deduplicate error handling at btrfs_iget_path()
btrfs: remove inode_lock from struct btrfs_root and use xarray locks
btrfs: unify index_cnt and csum_bytes from struct btrfs_inode
btrfs: don't allocate file extent tree for non regular files
btrfs: remove location key from struct btrfs_inode
btrfs: remove objectid from struct btrfs_inode on 64 bits platforms
btrfs: rename rb_root member of extent_map_tree from map to root
btrfs: use a regular rb_root instead of cached rb_root for extent_map_tree
btrfs: make btrfs_finish_ordered_extent() return void
btrfs: use a btrfs_inode in the log context (struct btrfs_log_ctx)
btrfs: pass a btrfs_inode to btrfs_fdatawrite_range()
btrfs: pass a btrfs_inode to btrfs_wait_ordered_range()
btrfs: use a btrfs_inode local variable at btrfs_sync_file()
btrfs: qgroup: avoid start/commit empty transaction when flushing reservations
btrfs: avoid create and commit empty transaction when committing super
btrfs: send: make ensure_commit_roots_uptodate() simpler and more efficient
btrfs: send: avoid create/commit empty transaction at ensure_commit_roots_uptodate()
btrfs: scrub: avoid create/commit empty transaction at finish_extent_writes_for_zoned()
btrfs: add and use helper to commit the current transaction
btrfs: send: get rid of the label and gotos at ensure_commit_roots_uptodate()
btrfs: move fiemap code into its own file
btrfs: reduce critical section at btrfs_wait_ordered_roots()
btrfs: reduce critical section at btrfs_wait_ordered_extents()
btrfs: add comment about locking to btrfs_split_ordered_extent()
btrfs: avoid removal and re-insertion of split ordered extent
btrfs: mark ordered extent insertion failure checks as unlikely
btrfs: update panic message when splitting ordered extent
btrfs: remove pointless code when creating and deleting a subvolume
btrfs: avoid transaction commit on any fsync after subvolume creation
btrfs: remove super block argument from btrfs_iget()
btrfs: remove super block argument from btrfs_iget_path()
btrfs: remove super block argument from btrfs_iget_locked()
btrfs: do not BUG_ON() when freeing tree block after error
btrfs: use label to deduplicate error path at btrfs_force_cow_block()
btrfs: remove NULL transaction support for btrfs_lookup_extent_info()
btrfs: simplify setting the full backref flag at update_ref_for_cow()
btrfs: replace BUG_ON() with error handling at update_ref_for_cow()
btrfs: remove superfluous metadata check at btrfs_lookup_extent_info()
btrfs: reduce nesting for extent processing at btrfs_lookup_extent_info()
btrfs: don't BUG_ON() when 0 reference count at btrfs_lookup_extent_info()
btrfs: avoid allocating and running pointless delayed extent operations
btrfs: move the direct IO code into its own file
btrfs: fix data race when accessing the last_trans field of a root
btrfs: fix bitmap leak when loading free space cache on duplicate entry
btrfs: fix extent map use-after-free when adding pages to compressed bio

Jeff Johnson (1):
btrfs: add MODULE_DESCRIPTION()

Johannes Thumshirn (8):
btrfs: pass struct btrfs_io_geometry into handle_ops_on_dev_replace()
btrfs: pass reloc_control to relocate_data_extent()
btrfs: pass a reloc_control to relocate_file_extent_cluster()
btrfs: pass a reloc_control to relocate_one_folio()
btrfs: don't pass fs_info to describe_relocation()
btrfs: pass a struct reloc_control to prealloc_file_extent_cluster()
btrfs: pass reloc_control to setup_relocation_extent_mapping()
btrfs: remove raid-stripe-tree encoding field from stripe_extent

Josef Bacik (15):
btrfs: don't do extra find_extent_buffer() in do_walk_down()
btrfs: remove all extra btrfs_check_eb_owner() calls
btrfs: use btrfs_read_extent_buffer() in do_walk_down()
btrfs: push lookup_info into struct walk_control
btrfs: factor out eb uptodate check from do_walk_down()
btrfs: remove local variable need_account in do_walk_down()
btrfs: unify logic to decide if we need to walk down into a node during snapshot delete
btrfs: extract the reference dropping code into it's own helper
btrfs: don't BUG_ON on ENOMEM from btrfs_lookup_extent_info() in walk_down_proc()
btrfs: handle errors from ref mods during UPDATE_BACKREF in walk_down_proc()
btrfs: replace BUG_ON with ASSERT in walk_down_proc()
btrfs: clean up our handling of refs == 0 in snapshot delete
btrfs: convert correctness BUG_ON()'s to ASSERT()'s in walk_up_proc()
btrfs: handle errors from btrfs_dec_ref() properly
btrfs: add documentation around snapshot delete

Junchao Sun (1):
btrfs: qgroup: delete a TODO about using kmem cache to allocate structures

Mark Harmstone (1):
btrfs: fix typo in error message in btrfs_validate_super()

Qu Wenruo (37):
btrfs: raid56: do extra dumping for CONFIG_BTRFS_ASSERT
btrfs: slightly loosen the requirement for qgroup removal
btrfs: automatically remove the subvolume qgroup
btrfs: rename extent_map::orig_block_len to disk_num_bytes
btrfs: export the expected file extent through can_nocow_extent()
btrfs: introduce new members for extent_map
btrfs: introduce extra sanity checks for extent maps
btrfs: remove extent_map::orig_start member
btrfs: remove extent_map::block_len member
btrfs: remove extent_map::block_start member
btrfs: cleanup duplicated parameters related to can_nocow_file_extent_args
btrfs: cleanup duplicated parameters related to btrfs_alloc_ordered_extent
btrfs: cleanup duplicated parameters related to create_io_em()
btrfs: cleanup duplicated parameters related to btrfs_create_dio_extent()
btrfs: make __extent_writepage_io() to write specified range only
btrfs: subpage: introduce helpers to handle subpage delalloc locking
btrfs: lock subpage ranges in one go for writepage_delalloc()
btrfs: do not clear page dirty inside extent_write_locked_range()
btrfs: make extent_write_locked_range() handle subpage writeback correctly
btrfs: cleanup recursive include of the same header
btrfs: do not directly include rwlock_types.h
btrfs: uapi: record temporary super flags used by btrfstune
btrfs: subpage: remove the unused error bitmap dumping
btrfs: print-tree: add generation and type dump for EXTENT_DATA_KEY
btrfs: cleanup the bytenr usage inside btrfs_extent_item_to_extent_map()
btrfs: ignore incorrect btrfs_file_extent_item::ram_bytes
btrfs: make validate_extent_map() catch ram_bytes mismatch
btrfs: fix the ram_bytes assignment for truncated ordered extents
btrfs: tree-checker: add extra ram_bytes and disk_num_bytes check
btrfs: remove unused Opt enums
btrfs: output the unrecognized super block flags as hex
btrfs: introduce new "rescue=ignoremetacsums" mount option
btrfs: introduce new "rescue=ignoresuperflags" mount option
btrfs: remove the extra_gfp parameter from btrfs_alloc_folio_array()
btrfs: rename the extra_gfp parameter of btrfs_alloc_page_array()
btrfs: move extent_range_clear_dirty_for_io() into inode.c
btrfs: remove the BUG_ON() inside extent_range_clear_dirty_for_io()

 fs/btrfs/Makefile                 |    2 +-
 fs/btrfs/accessors.h              |   15 +-
 fs/btrfs/bio.c                    |    4 +-
 fs/btrfs/block-group.c            |   53 +-
 fs/btrfs/block-group.h            |    3 +-
 fs/btrfs/btrfs_inode.h            |  156 ++--
 fs/btrfs/compression.c            |   25 +-
 fs/btrfs/compression.h            |    2 +-
 fs/btrfs/ctree.c                  |  108 +--
 fs/btrfs/ctree.h                  |   18 +-
 fs/btrfs/defrag.c                 |   18 +-
 fs/btrfs/delalloc-space.c         |    2 +-
 fs/btrfs/delalloc-space.h         |    2 +-
 fs/btrfs/delayed-inode.c          |   47 +-
 fs/btrfs/delayed-inode.h          |   10 +-
 fs/btrfs/delayed-ref.c            |   51 +-
 fs/btrfs/delayed-ref.h            |    8 +-
 fs/btrfs/dev-replace.c            |    4 +-
 fs/btrfs/dir-item.c               |    8 +-
 fs/btrfs/dir-item.h               |    6 +-
 fs/btrfs/direct-io.c              | 1052 ++++++++++++++++++++++++++
 fs/btrfs/direct-io.h              |   14 +
 fs/btrfs/disk-io.c                |  128 ++--
 fs/btrfs/disk-io.h                |   18 +-
 fs/btrfs/export.c                 |    6 +-
 fs/btrfs/extent-io-tree.c         |    4 +
 fs/btrfs/extent-tree.c            |  685 ++++++++++-------
 fs/btrfs/extent-tree.h            |    8 +-
 fs/btrfs/extent_io.c              | 1092 ++++-----------------------
 fs/btrfs/extent_io.h              |   19 +-
 fs/btrfs/extent_map.c             |  246 +++++--
 fs/btrfs/extent_map.h             |   56 +-
 fs/btrfs/fiemap.c                 |  930 +++++++++++++++++++++++
 fs/btrfs/fiemap.h                 |   11 +
 fs/btrfs/file-item.c              |   52 +-
 fs/btrfs/file.c                   |  355 +--------
 fs/btrfs/file.h                   |    4 +-
 fs/btrfs/free-space-cache.c       |   12 +-
 fs/btrfs/free-space-tree.c        |   10 +-
 fs/btrfs/fs.h                     |   17 +-
 fs/btrfs/inode-item.c             |    4 +-
 fs/btrfs/inode.c                  | 1475 ++++++++-----------------------------
 fs/btrfs/ioctl.c                  |   94 ++-
 fs/btrfs/ioctl.h                  |    2 +-
 fs/btrfs/locking.h                |    1 -
 fs/btrfs/lru_cache.h              |    1 -
 fs/btrfs/lzo.c                    |   43 +-
 fs/btrfs/messages.c               |    3 +-
 fs/btrfs/misc.h                   |    4 +-
 fs/btrfs/ordered-data.c           |  146 ++--
 fs/btrfs/ordered-data.h           |   27 +-
 fs/btrfs/print-tree.c             |   10 +-
 fs/btrfs/props.c                  |   20 +-
 fs/btrfs/props.h                  |    4 +-
 fs/btrfs/qgroup.c                 |  221 ++++--
 fs/btrfs/qgroup.h                 |   25 +-
 fs/btrfs/raid-stripe-tree.c       |   13 -
 fs/btrfs/raid-stripe-tree.h       |    3 +-
 fs/btrfs/raid56.c                 |  118 ++-
 fs/btrfs/reflink.c                |    8 +-
 fs/btrfs/relocation.c             |  157 ++--
 fs/btrfs/scrub.c                  |   13 +-
 fs/btrfs/send.c                   |   49 +-
 fs/btrfs/send.h                   |    4 +-
 fs/btrfs/space-info.c             |  265 ++++++-
 fs/btrfs/space-info.h             |   48 ++
 fs/btrfs/subpage.c                |  162 +++-
 fs/btrfs/subpage.h                |    9 +-
 fs/btrfs/super.c                  |   51 +-
 fs/btrfs/super.h                  |    2 +-
 fs/btrfs/sysfs.c                  |   85 ++-
 fs/btrfs/tests/btrfs-tests.c      |    5 +-
 fs/btrfs/tests/extent-map-tests.c |  120 +--
 fs/btrfs/tests/inode-tests.c      |  176 +++--
 fs/btrfs/transaction.c            |   31 +-
 fs/btrfs/transaction.h            |    9 +-
 fs/btrfs/tree-checker.c           |   37 +-
 fs/btrfs/tree-log.c               |   74 +-
 fs/btrfs/tree-log.h               |    6 +-
 fs/btrfs/ulist.c                  |   21 +-
 fs/btrfs/ulist.h                  |    2 +
 fs/btrfs/uuid-tree.c              |   10 +-
 fs/btrfs/uuid-tree.h              |    4 +-
 fs/btrfs/volumes.c                |   62 +-
 fs/btrfs/volumes.h                |    2 +-
 fs/btrfs/xattr.c                  |    4 +-
 fs/btrfs/xattr.h                  |    2 +-
 fs/btrfs/zlib.c                   |   56 +-
 fs/btrfs/zoned.c                  |   30 +-
 fs/btrfs/zoned.h                  |   11 +-
 fs/btrfs/zstd.c                   |   70 +-
 include/trace/events/btrfs.h      |   19 +-
 include/uapi/linux/btrfs_tree.h   |   22 +-
 93 files changed, 5186 insertions(+), 3915 deletions(-)
 create mode 100644 fs/btrfs/direct-io.c
 create mode 100644 fs/btrfs/direct-io.h
 create mode 100644 fs/btrfs/fiemap.c
 create mode 100644 fs/btrfs/fiemap.h
merge bot Merged
#1 lore raw
The pull request you sent on Mon, 15 Jul 2024 20:12:20 +0200:

The pull request you sent on Mon, 15 Jul 2024 20:12:20 +0200:

Quoted 1 line

git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git tags/for-6.11-tag

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/a1b547f0f217cfb06af7eb4ce8488b02d83a0370

Thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html
#2 lore raw
In reply to David Sterba
Quoted 5 lines · David Sterba

There's a merge conflict caused by the latency fixes from last week in extent_map.c:btrfs_scan_inode(), related commits 4e660ca3a98d931809734 and b3ebb9b7e92a928344a. Resolved in branch for-6.11-merged and that's been in linux-next for a few days.

Oh, and I notice that my resolution is slightly different, but looks like the actual code is equivalent.

I kept the "q" logic that had been introduced by commit 4e660ca3a98d ("btrfs: use a regular rb_root instead of cached rb_root for extent_map_tree").

I don't know how you prefer the code, but it kept the two "while (node)" loops in that file looking similar.

Of course, they were very different in other respects (ie drop_all_extent_maps_fast() does all extents unconditionally with that retry logic, while btrfs_scan_inode() has that whole "bail out on any contention" model, so whatever.

Anyway, it all *looks* ok to me, but please go and double-check that I didn't mess something up.

Linus

#3 lore raw
Quoted 10 lines · Linus Torvalds
Quoted 5 lines · David Sterba

There's a merge conflict caused by the latency fixes from last week in extent_map.c:btrfs_scan_inode(), related commits 4e660ca3a98d931809734 and b3ebb9b7e92a928344a. Resolved in branch for-6.11-merged and that's been in linux-next for a few days.

Oh, and I notice that my resolution is slightly different, but looks like the actual code is equivalent.

Quoted 6 lines

I kept the "q" logic that had been introduced by commit 4e660ca3a98d ("btrfs: use a regular rb_root instead of cached rb_root for extent_map_tree").

I don't know how you prefer the code, but it kept the two "while (node)" loops in that file looking similar.

That is fine, keeping the same style makes sense.

Quoted 7 lines

Of course, they were very different in other respects (ie drop_all_extent_maps_fast() does all extents unconditionally with that retry logic, while btrfs_scan_inode() has that whole "bail out on any contention" model, so whatever.

Anyway, it all *looks* ok to me, but please go and double-check that I didn't mess something up.

Looks correct to me as well.

#4 lore raw
In reply to David Sterba
Quoted 9 lines · David Sterba

please pull the changes described below. The hilights are new logic behind background block group reclaim, automatic removal of qgroup after removing a subvolume and new 'rescue=' mount options. The rest is optimizations, cleanups and refactoring.

There's a merge conflict caused by the latency fixes from last week in extent_map.c:btrfs_scan_inode(), related commits 4e660ca3a98d931809734 and b3ebb9b7e92a928344a. Resolved in branch for-6.11-merged and that's been in linux-next for a few days.

FTR, this is broken on 32-bit (doesn't build, good ;-) and on big-endian (compiler warnings, no idea how it behaves :-(, so you better don't trust your data to it in the latter case...

Gr{oetje,eeting}s,

Geert

Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
#5 lore raw

Hi David,

Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · David Sterba

please pull the changes described below. The hilights are new logic behind background block group reclaim, automatic removal of qgroup after removing a subvolume and new 'rescue=' mount options. The rest is optimizations, cleanups and refactoring.

Quoted 8 lines
Quoted 4 lines

There's a merge conflict caused by the latency fixes from last week in extent_map.c:btrfs_scan_inode(), related commits 4e660ca3a98d931809734 and b3ebb9b7e92a928344a. Resolved in branch for-6.11-merged and that's been in linux-next for a few days.

FTR, this is broken on 32-bit (doesn't build, good ;-) and on big-endian (compiler warnings, no idea how it behaves :-(, so you better don't trust your data to it in the latter case...

I cannot find any other report of this, and don't know yet where it was introduced, but the bots started reporting this last May:

3 fs/btrfs/inode.c:5711:5: warning: ‘location.type’ may be used
uninitialized in this function [-Wmaybe-uninitialized]
3 fs/btrfs/inode.c:5640:9: warning: ‘location.objectid’ may be
used uninitialized in this function [-Wmaybe-uninitialized]

https://lore.kernel.org/all/[email protected]/

and I'm seeing failures in e.g. my m68k allmodconfig builds with gcc 9.5 due to CONFIG_WERROR=y.

I suspect the big-endian accessors in fs/btrfs/accessors.h lack some initializations?

Gr{oetje,eeting}s,

Geert

Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
#6 lore raw
Quoted 17 lines · Geert Uytterhoeven

Hi David,

Quoted 14 lines · Geert Uytterhoeven
Quoted 9 lines · David Sterba

please pull the changes described below. The hilights are new logic behind background block group reclaim, automatic removal of qgroup after removing a subvolume and new 'rescue=' mount options. The rest is optimizations, cleanups and refactoring.

There's a merge conflict caused by the latency fixes from last week in extent_map.c:btrfs_scan_inode(), related commits 4e660ca3a98d931809734 and b3ebb9b7e92a928344a. Resolved in branch for-6.11-merged and that's been in linux-next for a few days.

FTR, this is broken on 32-bit (doesn't build, good ;-) and on big-endian (compiler warnings, no idea how it behaves :-(, so you better don't trust your data to it in the latter case...

Internet forums are full of such quick and wrong conclusions, you don't need to write more.

Quoted 8 lines

I cannot find any other report of this, and don't know yet where it was introduced, but the bots started reporting this last May:

3 fs/btrfs/inode.c:5711:5: warning: ‘location.type’ may be used
uninitialized in this function [-Wmaybe-uninitialized]
3 fs/btrfs/inode.c:5640:9: warning: ‘location.objectid’ may be
used uninitialized in this function [-Wmaybe-uninitialized]

Quoted 4 lines

https://lore.kernel.org/all/[email protected]/

and I'm seeing failures in e.g. my m68k allmodconfig builds with gcc 9.5 due to CONFIG_WERROR=y.

Older compilers like 9.5 could not be able to reason about variable validity in case it's passed by address, as is in this case:

5707 ret = btrfs_inode_by_name(BTRFS_I(dir), dentry, &location, &di_type);
5708 if (ret < 0)
5709 return ERR_PTR(ret);

and btrfs_inode_by_name() returns either a valid 'location' or an error that the caller handles and does not use the variable.

Quoted 2 lines

I suspect the big-endian accessors in fs/btrfs/accessors.h lack some initializations?

There are no special accessors on big endian hosts, same code, same bytes in memory only a different order.

We fix warnings caused -Wmaybe-uninitialized even if it's because of old compilers, but it's hard to notice reports if they're burried in some mailinglist.

I do read your build reports after each -rcN but there are only some modpost warnings in 6.10-rc7

https://lore.kernel.org/all/[email protected]/