Changes in Makefile

5443041b - xfs: export zone stats in /proc/*/mountstats

Sat, 30 Nov 2024 04:53:49 +0000

xfs: export zone stats in /proc/*/mountstatsAdd the per-zone life time hint and the used block distributionfor fully written zones, grouping reclaimable zones in fixed-percentagebuckets spanning 0..9%, 10..19% and full zones as 100% used as well as afew statistics about the zone allocator and open and reclaimable zonesin /proc/*/mountstats.This gives good insight into data fragmentation and data placementsuccess rate.Signed-off-by: Hans Holmberg Co-developed-by: Christoph Hellwig Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" List of files: /linux-6.15/fs/xfs/Makefile

080d01c4 - xfs: implement zoned garbage collection

Sat, 15 Feb 2025 18:37:19 +0000

xfs: implement zoned garbage collectionRT groups on a zoned file system need to be completely empty before theirspace can be reused. This means that partially empty groups need to beemptied entirely to free up space if no entirely free groups areavailable.Add a garbage collection thread that moves all data out of the least usedzone when not enough free zones are available, and which resets all zonesthat have been emptied. To find empty zone a simple set of 10 bucketsbased on the amount of space used in the zone is used. To empty zones,the rmap is walked to find the owners and the data is read and thenwritten to the new place.To automatically defragment files the rmap records are sorted by inodeand logical offset. This means defragmentation of parallel writes intoa single zone happens automatically when performing garbage collection.Because holding the iolock over the entire GC cycle would inject verynoticeable latency for other accesses to the inodes, the iolock is nottaken while performing I/O. Instead the I/O completion handler checksthat the mapping hasn't changed over the one recorded at the start ofthe GC cycle and doesn't update the mapping if it change.Co-developed-by: Hans Holmberg Signed-off-by: Hans Holmberg Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" List of files: /linux-6.15/fs/xfs/Makefile

0bb21930 - xfs: add support for zoned space reservations

Thu, 13 Feb 2025 08:16:06 +0000

xfs: add support for zoned space reservationsFor zoned file systems garbage collection (GC) has to take the iolockand mmaplock after moving data to a new place to synchronize withreaders. This means waiting for garbage collection with the iolock candeadlock.To avoid this, the worst case required blocks have to be reserved beforetaking the iolock, which is done using a new RTAVAILABLE counter thattracks blocks that are free to write into and don't require garbagecollection. The new helpers try to take these available blocks, andif there aren't enough available it wakes and waits for GC. This isdone using a list of on-stack reservations to ensure fairness.Co-developed-by: Hans Holmberg Signed-off-by: Hans Holmberg Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" List of files: /linux-6.15/fs/xfs/Makefile

4e4d5207 - xfs: add the zoned space allocator

Thu, 13 Feb 2025 04:49:17 +0000

xfs: add the zoned space allocatorFor zoned RT devices space is always allocated at the write pointer, thatis right after the last written block and only recorded on I/O completion.Because the actual allocation algorithm is very simple and just involvespicking a good zone - preferably the one used for the last write to theinode. As the number of zones that can written at the same time isusually limited by the hardware, selecting a zone is done as late aspossible from the iomap dio and buffered writeback bio submissionshelpers just before submitting the bio.Given that the writers already took a reservation before acquiring theiolock, space will always be readily available if an open zone slot isavailable. A new structure is used to track these open zones, andpointed to by the xfs_rtgroup. Because zoned file systems don't havea rsum cache the space for that pointer can be reused.Allocations are only recorded at I/O completion time. The scheme usedfor that is very similar to the reflink COW end I/O path.Co-developed-by: Hans Holmberg Signed-off-by: Hans Holmberg Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" List of files: /linux-6.15/fs/xfs/Makefile

720c2d58 - xfs: parse and validate hardware zone information

Sun, 17 Nov 2024 05:28:10 +0000

xfs: parse and validate hardware zone informationAdd support to validate and parse reported hardware zone state.Co-developed-by: Hans Holmberg Signed-off-by: Hans Holmberg Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" List of files: /linux-6.15/fs/xfs/Makefile

83ccffc4 - xfs: online repair of the realtime refcount btree

Thu, 21 Nov 2024 00:21:14 +0000

xfs: online repair of the realtime refcount btreePort the data device's refcount btree repair code to the realtimerefcount btree.Signed-off-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

c2792967 - xfs: scrub the realtime refcount btree

Thu, 21 Nov 2024 00:21:06 +0000

xfs: scrub the realtime refcount btreeAdd code to scrub realtime refcount btrees. Similar to the refcountbtree checking code for the data device, we walk the rmap btree for eachrefcount record to confirm that the reference counts are correct.Signed-off-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

9abe03a0 - xfs: introduce realtime refcount btree ondisk definitions

Thu, 21 Nov 2024 00:20:47 +0000

xfs: introduce realtime refcount btree ondisk definitionsAdd the ondisk structure definitions for realtime refcount btrees. Therealtime refcount btree will be rooted from a hidden inode so it needsto have a separate btree block magic and pointer format.Next, add everything needed to read, write and manipulate refcount btreeblocks. This prepares the way for connecting the btree operationsimplementation, though the changes to actually root the rtrefcount btreein an inode come later.Signed-off-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

6a849bd8 - xfs: online repair of the realtime rmap btree

Thu, 21 Nov 2024 00:20:42 +0000

xfs: online repair of the realtime rmap btreeRepair the realtime rmap btree while mounted. Similar to the regularrmap btree repair code, we walk the data fork mappings of every realtimefile in the filesystem to collect reverse-mapping records in an xfarray.Then we sort the xfarray, and use the btree bulk loader to create a newrtrmap btree ondisk. Finally, we swap the btree roots, and reap the oldblocks in the usual way.Signed-off-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

9a6cc4f6 - xfs: scrub the realtime rmapbt

Thu, 21 Nov 2024 00:20:34 +0000

xfs: scrub the realtime rmapbtCheck the realtime reverse mapping btree against the rtbitmap, andmodify the rtbitmap scrub to check against the rtrmapbt.Signed-off-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

fc6856c6 - xfs: introduce realtime rmap btree ondisk definitions

Thu, 21 Nov 2024 00:20:21 +0000

xfs: introduce realtime rmap btree ondisk definitionsAdd the ondisk structure definitions for realtime rmap btrees. Therealtime rmap btree will be rooted from a hidden inode so it needs tohave a separate btree block magic and pointer format.Next, add everything needed to read, write and manipulate rmap btreeblocks. This prepares the way for connecting the btree operationsimplementation, though embedding the rtrmap btree root in the inodecomes later in the series.Signed-off-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

3f1bdf50 - xfs: scrub the realtime group superblock

Mon, 04 Nov 2024 04:19:30 +0000

xfs: scrub the realtime group superblockEnable scrubbing of realtime group superblocks.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

87fe4c34 - xfs: create incore realtime group structures

Mon, 04 Nov 2024 04:19:04 +0000

xfs: create incore realtime group structuresCreate an incore object that will contain information about a realtimeallocation group. This will eventually enable us to shard the realtimesection in a similar manner to how we shard the data section, but fornow just a single object for the entire RT subvolume is created.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

b3c03efa - xfs: check metadata directory file path connectivity

Mon, 04 Nov 2024 04:19:02 +0000

xfs: check metadata directory file path connectivityCreate a new scrubber type that checks that well known metadatadirectory paths are connected to the metadata inode that the incorestructures think is in use. For example, check that "/quota/user" inthe metadata directory tree actually points tomp->m_quotainfo->qi_uquotaip->i_ino.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

5d9b54a4 - xfs: read and write metadata inode directory tree

Mon, 04 Nov 2024 04:18:52 +0000

xfs: read and write metadata inode directory treePlumb in the bits we need to load metadata inodes from a named entry ina metadir directory, create (or hardlink) inodes into a metadirdirectory, create metadir directories, and flag inodes as being metadatafiles.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

e9c4d8bf - xfs: factor out a generic xfs_group structure

Mon, 04 Nov 2024 04:18:38 +0000

xfs: factor out a generic xfs_group structureSplit the lookup and refcount handling of struct xfs_perag into anembedded xfs_group structure that can be reused for the upcomingrealtime groups.It will be extended with more features later.Note that he xg_type field will only need a single bit even withrealtime group support. For now it fills a hole, but it might beworth to fold it into another field if we can use this space better.Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong List of files: /linux-6.15/fs/xfs/Makefile

b7c477be - xfs: hoist inode flag conversion functions to libxfs

Tue, 02 Jul 2024 18:22:32 +0000

xfs: hoist inode flag conversion functions to libxfsHoist the inode flag conversion functions into libxfs so that we cankeep them in sync. Do this by creating a new xfs_inode_util.c file inlibxfs.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

b1992c37 - kbuild: use $(src) instead of $(srctree)/$(src) for source directory

Sat, 27 Apr 2024 14:55:02 +0000

kbuild: use $(src) instead of $(srctree)/$(src) for source directoryKbuild conventionally uses $(obj)/ for generated files, and $(src)/ forchecked-in source files. It is merely a convention without any functionaldifference. In fact, $(obj) and $(src) are exactly the same, as definedin scripts/Makefile.build: src := $(obj)When the kernel is built in a separate output directory, $(src) doesnot accurately reflect the source directory location. While Kbuildresolves this discrepancy by specifying VPATH=$(srctree) to search forsource files, it does not cover all cases. For example, when adding aheader search path for local headers, -I$(srctree)/$(src) is typicallypassed to the compiler.This introduces inconsistency between upstream and downstream Makefilesbecause $(src) is used instead of $(srctree)/$(src) for the latter.To address this inconsistency, this commit changes the semantics of$(src) so that it always points to the directory in the source tree.Going forward, the variables used in Makefiles will have the followingmeanings: $(obj) - directory in the object tree $(src) - directory in the source tree (changed by this commit) $(objtree) - the top of the kernel object tree $(srctree) - the top of the kernel source treeConsequently, $(srctree)/$(src) in upstream Makefiles need to be replacedwith $(src).Signed-off-by: Masahiro Yamada Reviewed-by: Nicolas Schier List of files: /linux-6.15/fs/xfs/Makefile

3f31406a - xfs: fix corruptions in the directory tree

Mon, 22 Apr 2024 16:48:22 +0000

xfs: fix corruptions in the directory treeRepair corruptions in the directory tree itself. Cycles are broken byremoving an incoming parent->child link. Multiply-owned directories arefixed by pruning the extra parent -> child links Disconnected subtreesare reconnected to the lost and found.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile

928b721a - xfs: teach online scrub to find directory tree structure problems

Mon, 22 Apr 2024 16:48:20 +0000

xfs: teach online scrub to find directory tree structure problemsCreate a new scrubber that detects corruptions within the directory treestructure itself. It can detect directories with multiple parents;loops within the directory tree; and directory loops not accessible fromthe root.Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig List of files: /linux-6.15/fs/xfs/Makefile