Commit c1e82275 authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge tag 'bcachefs-2024-10-22' of https://github.com/koverstreet/bcachefs 

Pull bcachefs fixes from Kent Overstreet:
 "Lots of hotfixes:

   - transaction restart injection has been shaking out a few things

   - fix a data corruption in the buffered write path on -ENOSPC, found
     by xfstests generic/299

   - Some small show_options fixes

   - Repair mismatches in inode hash type, seed: different snapshot
     versions of an inode must have the same hash/type seed, used for
     directory entries and xattrs. We were checking the hash seed, but
     not the type, and a user contributed a filesystem where the hash
     type on one inode had somehow been flipped; these fixes allow his
     filesystem to repair.

     Additionally, the hash type flip made some directory entries
     invisible, which were then recreated by userspace; so the hash
     check code now checks for duplicate non dangling dirents, and
     renames one of them if necessary.

   - Don't use wait_event_interruptible() in recovery: this fixes some
     filesystems failing to mount with -ERESTARTSYS

   - Workaround for kvmalloc not supporting > INT_MAX allocations,
     causing an -ENOMEM when allocating the sorted array of journal
     keys: this allows a 75 TB filesystem to mount

   - Make sure bch_inode_unpacked.bi_snapshot is set in the old inode
     compat path: this alllows Marcin's filesystem (in use since before
     6.7) to repair and mount"

* tag 'bcachefs-2024-10-22' of https://github.com/koverstreet/bcachefs: (26 commits)
  bcachefs: Set bch_inode_unpacked.bi_snapshot in old inode path
  bcachefs: Mark more errors as AUTOFIX
  bcachefs: Workaround for kvmalloc() not supporting > INT_MAX allocations
  bcachefs: Don't use wait_event_interruptible() in recovery
  bcachefs: Fix __bch2_fsck_err() warning
  bcachefs: fsck: Improve hash_check_key()
  bcachefs: bch2_hash_set_or_get_in_snapshot()
  bcachefs: Repair mismatches in inode hash seed, type
  bcachefs: Add hash seed, type to inode_to_text()
  bcachefs: INODE_STR_HASH() for bch_inode_unpacked
  bcachefs: Run in-kernel offline fsck without ratelimit errors
  bcachefs: skip mount option handle for empty string.
  bcachefs: fix incorrect show_options results
  bcachefs: Fix data corruption on -ENOSPC in buffered write path
  bcachefs: bch2_folio_reservation_get_partial() is now better behaved
  bcachefs: fix disk reservation accounting in bch2_folio_reservation_get()
  bcachefS: ec: fix data type on stripe deletion
  bcachefs: Don't use commit_do() unnecessarily
  bcachefs: handle restarts in bch2_bucket_io_time_reset()
  bcachefs: fix restart handling in __bch2_resume_logged_op_finsert()
  ...
parents f009e946 a069f014

fs/bcachefs/alloc_background.c

+21 −16
Original line number Diff line number Diff line
@@ -1977,7 +1977,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work) 
					     ca->mi.bucket_size,

					     GFP_KERNEL);



		int ret = bch2_trans_do(c, NULL, NULL,

		int ret = bch2_trans_commit_do(c, NULL, NULL,

			BCH_WATERMARK_btree|

			BCH_TRANS_COMMIT_no_enospc,

			bch2_clear_bucket_needs_discard(trans, POS(ca->dev_idx, bucket)));
@@ -2137,14 +2137,15 @@ static void bch2_do_invalidates_work(struct work_struct *work) 


		struct bkey_s_c k = next_lru_key(trans, &iter, ca, &wrapped);

		ret = bkey_err(k);

		if (bch2_err_matches(ret, BCH_ERR_transaction_restart))

			continue;

		if (ret)

			break;

			goto restart_err;

		if (!k.k)

			break;



		ret = invalidate_one_bucket(trans, &iter, k, &nr_to_invalidate);

restart_err:

		if (bch2_err_matches(ret, BCH_ERR_transaction_restart))

			continue;

		if (ret)

			break;
@@ -2350,24 +2351,19 @@ int bch2_dev_remove_alloc(struct bch_fs *c, struct bch_dev *ca) 


/* Bucket IO clocks: */



int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev,

static int __bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev,

				size_t bucket_nr, int rw)

{

	struct bch_fs *c = trans->c;

	struct btree_iter iter;

	struct bkey_i_alloc_v4 *a;

	u64 now;

	int ret = 0;



	if (bch2_trans_relock(trans))

		bch2_trans_begin(trans);



	a = bch2_trans_start_alloc_update_noupdate(trans, &iter, POS(dev, bucket_nr));

	ret = PTR_ERR_OR_ZERO(a);

	struct btree_iter iter;

	struct bkey_i_alloc_v4 *a =

		bch2_trans_start_alloc_update_noupdate(trans, &iter, POS(dev, bucket_nr));

	int ret = PTR_ERR_OR_ZERO(a);

	if (ret)

		return ret;



	now = bch2_current_io_time(c, rw);

	u64 now = bch2_current_io_time(c, rw);

	if (a->v.io_time[rw] == now)

		goto out;
@@ -2380,6 +2376,15 @@ int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev, 
	return ret;

}



int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev,

			      size_t bucket_nr, int rw)

{

	if (bch2_trans_relock(trans))

		bch2_trans_begin(trans);



	return nested_lockrestart_do(trans, __bch2_bucket_io_time_reset(trans, dev, bucket_nr, rw));

}



/* Startup/shutdown (ro/rw): */



void bch2_recalc_capacity(struct bch_fs *c)

fs/bcachefs/alloc_foreground.c

+1 −1
Original line number Diff line number Diff line
@@ -684,7 +684,7 @@ struct open_bucket *bch2_bucket_alloc(struct bch_fs *c, struct bch_dev *ca, 
	struct bch_dev_usage usage;

	struct open_bucket *ob;



	bch2_trans_do(c, NULL, NULL, 0,

	bch2_trans_do(c,

		      PTR_ERR_OR_ZERO(ob = bch2_bucket_alloc_trans(trans, ca, watermark,

							data_type, cl, false, &usage)));

	return ob;

fs/bcachefs/btree_gc.c

+11 −1
Original line number Diff line number Diff line
@@ -820,12 +820,22 @@ static int bch2_alloc_write_key(struct btree_trans *trans, 
	 * fix that here:

	 */

	alloc_data_type_set(&gc, gc.data_type);



	if (gc.data_type != old_gc.data_type ||

	    gc.dirty_sectors != old_gc.dirty_sectors) {

		ret = bch2_alloc_key_to_dev_counters(trans, ca, &old_gc, &gc, BTREE_TRIGGER_gc);

		if (ret)

			return ret;



		/*

		 * Ugly: alloc_key_to_dev_counters(..., BTREE_TRIGGER_gc) is not

		 * safe w.r.t. transaction restarts, so fixup the gc_bucket so

		 * we don't run it twice:

		 */

		percpu_down_read(&c->mark_lock);

		struct bucket *gc_m = gc_bucket(ca, iter->pos.offset);

		gc_m->data_type = gc.data_type;

		gc_m->dirty_sectors = gc.dirty_sectors;

		percpu_up_read(&c->mark_lock);

	}



	if (fsck_err_on(new.data_type != gc.data_type,

fs/bcachefs/btree_io.c

+1 −1
Original line number Diff line number Diff line
@@ -1871,7 +1871,7 @@ static void btree_node_write_work(struct work_struct *work) 


		}

	} else {

		ret = bch2_trans_do(c, NULL, NULL, 0,

		ret = bch2_trans_do(c,

			bch2_btree_node_update_key_get_iter(trans, b, &wbio->key,

					BCH_WATERMARK_interior_updates|

					BCH_TRANS_COMMIT_journal_reclaim|

fs/bcachefs/btree_iter.h

+2 −0
Original line number Diff line number Diff line
@@ -912,6 +912,8 @@ struct bkey_s_c bch2_btree_iter_peek_and_restart_outlined(struct btree_iter *); 
	_ret;								\

})



#define bch2_trans_do(_c, _do)	bch2_trans_run(_c, lockrestart_do(trans, _do))



struct btree_trans *__bch2_trans_get(struct bch_fs *, unsigned);

void bch2_trans_put(struct btree_trans *);