Hi Chandan, Please pull this branch with changes for xfs for 6.9-rc1. As usual, I did a test-merge with the main upstream branch as of a few minutes ago, and didn't see any conflicts. Please let me know if you encounter any problems. --D The following changes since commit 7e1b84b24d257700e417bc9cd724c1efdff653d7: xfs: hook live rmap operations during a repair operation (2024-02-22 12:43:40 -0800) are available in the Git repository at: https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git tags/repair-refcount-scalability-6.9_2024-02-23 for you to fetch changes up to 7fbaab57a80f1639add1c7d02adeb9d17bd50206: xfs: port refcount repair to the new refcount bag structure (2024-02-22 12:43:42 -0800) ---------------------------------------------------------------- xfs: reduce refcount repair memory usage [v29.3 14/18] The refcountbt repair code has serious memory usage problems when the block sharing factor of the filesystem is very high. This can happen if a deduplication tool has been run against the filesystem, or if the fs stores reflinked VM images that have been aging for a long time. Recall that the original reference counting algorithm walks the reverse mapping records of the filesystem to generate reference counts. For any given block in the AG, the rmap bag structure contains the all rmap records that cover that block; the refcount is the size of that bag. For online repair, the bag doesn't need the owner, offset, or state flag information, so it discards those. This halves the record size, but the bag structure still stores one excerpted record for each reverse mapping. If the sharing count is high, this will use a LOT of memory storing redundant records. In the extreme case, 100k mappings to the same piece of space will consume 100k*16 bytes = 1.6M of memory. For offline repair, the bag stores the owner values so that we know which inodes need to be marked as being reflink inodes. If a deduplication tool has been run and there are many blocks within a file pointing to the same physical space, this will stll use a lot of memory to store redundant records. The solution to this problem is to deduplicate the bag records when possible by adding a reference count to the bag record, and changing the bag add function to detect an existing record to bump the refcount. In the above example, the 100k mappings will now use 24 bytes of memory. These lookups can be done efficiently with a btree, so we create a new refcount bag btree type (inside of online repair). This is why we refactored the btree code in the previous patchset. The btree conversion also dramatically reduces the runtime of the refcount generation algorithm, because the code to delete all bag records that end at a given agblock now only has to delete one record instead of (using the example above) 100k records. As an added benefit, record deletion now gives back the unused xfile space, which it did not do previously. This has been running on the djcloud for months with no problems. Enjoy! Signed-off-by: Darrick J. Wong <djwong@xxxxxxxxxx> ---------------------------------------------------------------- Darrick J. Wong (3): xfs: define an in-memory btree for storing refcount bag info during repairs xfs: create refcount bag structure for btree repairs xfs: port refcount repair to the new refcount bag structure fs/xfs/Makefile | 2 + fs/xfs/scrub/rcbag.c | 307 ++++++++++++++++++++++++++++++++++ fs/xfs/scrub/rcbag.h | 28 ++++ fs/xfs/scrub/rcbag_btree.c | 370 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/rcbag_btree.h | 81 +++++++++ fs/xfs/scrub/refcount.c | 12 ++ fs/xfs/scrub/refcount_repair.c | 164 +++++++----------- fs/xfs/scrub/repair.h | 2 + fs/xfs/xfs_stats.c | 3 +- fs/xfs/xfs_stats.h | 1 + fs/xfs/xfs_super.c | 10 +- 11 files changed, 872 insertions(+), 108 deletions(-) create mode 100644 fs/xfs/scrub/rcbag.c create mode 100644 fs/xfs/scrub/rcbag.h create mode 100644 fs/xfs/scrub/rcbag_btree.c create mode 100644 fs/xfs/scrub/rcbag_btree.h
