Merge tag 'for-6.12-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:

 - update fstrim loop and add more cancellation points, fix reported
   delayed or blocked suspend if there's a huge chunk queued

 - fix error handling in recent qgroup xarray conversion

 - in zoned mode, fix warning printing device path without RCU
   protection

 - again fix invalid extent xarray state (6252690), lost due to
   refactoring

* tag 'for-6.12-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: fix clear_dirty and writeback ordering in submit_one_sector()
  btrfs: zoned: fix missing RCU locking in error message when loading zone info
  btrfs: fix missing error handling when adding delayed ref with qgroups enabled
  btrfs: add cancellation points to trim loops
  btrfs: split remaining space to discard in chunks

Author: torvalds

fs/btrfs/delayed-ref.c

@@ -840,6 +840,8 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref,
  * helper function to actually insert a head node into the rbtree.
  * this does all the dirty work in terms of maintaining the correct
  * overall modification count.
+ *
+ * Returns an error pointer in case of an error.
  */
 static noinline struct btrfs_delayed_ref_head *
 add_delayed_ref_head(struct btrfs_trans_handle *trans,
@@ -862,6 +864,9 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans,
 		if (ret) {
 			/* Clean up if insertion fails or item exists. */
 			xa_release(&delayed_refs->dirty_extents, qrecord->bytenr);
+			/* Caller responsible for freeing qrecord on error. */
+			if (ret < 0)
+				return ERR_PTR(ret);
 			kfree(qrecord);
 		} else {
 			qrecord_inserted = true;
@@ -1000,27 +1005,35 @@ static int add_delayed_ref(struct btrfs_trans_handle *trans,
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_ref_node *node;
 	struct btrfs_delayed_ref_head *head_ref;
+	struct btrfs_delayed_ref_head *new_head_ref;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_qgroup_extent_record *record = NULL;
 	bool qrecord_inserted;
 	int action = generic_ref->action;
 	bool merged;
+	int ret;
 
 	node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS);
 	if (!node)
 		return -ENOMEM;
 
 	head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
-	if (!head_ref)
+	if (!head_ref) {
+		ret = -ENOMEM;
 		goto free_node;
+	}
 
 	if (btrfs_qgroup_full_accounting(fs_info) && !generic_ref->skip_qgroup) {
 		record = kzalloc(sizeof(*record), GFP_NOFS);
-		if (!record)
+		if (!record) {
+			ret = -ENOMEM;
 			goto free_head_ref;
+		}
 		if (xa_reserve(&trans->transaction->delayed_refs.dirty_extents,
-			       generic_ref->bytenr, GFP_NOFS))
+			       generic_ref->bytenr, GFP_NOFS)) {
+			ret = -ENOMEM;
 			goto free_record;
+		}
 	}
 
 	init_delayed_ref_common(fs_info, node, generic_ref);
@@ -1034,8 +1047,14 @@ static int add_delayed_ref(struct btrfs_trans_handle *trans,
 	 * insert both the head node and the new ref without dropping
 	 * the spin lock
 	 */
-	head_ref = add_delayed_ref_head(trans, head_ref, record,
-					action, &qrecord_inserted);
+	new_head_ref = add_delayed_ref_head(trans, head_ref, record,
+					    action, &qrecord_inserted);
+	if (IS_ERR(new_head_ref)) {
+		spin_unlock(&delayed_refs->lock);
+		ret = PTR_ERR(new_head_ref);
+		goto free_record;
+	}
+	head_ref = new_head_ref;
 
 	merged = insert_delayed_ref(trans, head_ref, node);
 	spin_unlock(&delayed_refs->lock);
@@ -1063,7 +1082,7 @@ static int add_delayed_ref(struct btrfs_trans_handle *trans,
 	kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref);
 free_node:
 	kmem_cache_free(btrfs_delayed_ref_node_cachep, node);
-	return -ENOMEM;
+	return ret;
 }
 
 /*
@@ -1094,6 +1113,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans,
 				struct btrfs_delayed_extent_op *extent_op)
 {
 	struct btrfs_delayed_ref_head *head_ref;
+	struct btrfs_delayed_ref_head *head_ref_ret;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_ref generic_ref = {
 		.type = BTRFS_REF_METADATA,
@@ -1113,11 +1133,15 @@ int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans,
 	delayed_refs = &trans->transaction->delayed_refs;
 	spin_lock(&delayed_refs->lock);
 
-	add_delayed_ref_head(trans, head_ref, NULL, BTRFS_UPDATE_DELAYED_HEAD,
-			     NULL);
-
+	head_ref_ret = add_delayed_ref_head(trans, head_ref, NULL,
+					    BTRFS_UPDATE_DELAYED_HEAD, NULL);
 	spin_unlock(&delayed_refs->lock);
 
+	if (IS_ERR(head_ref_ret)) {
+		kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref);
+		return PTR_ERR(head_ref_ret);
+	}
+
 	/*
 	 * Need to update the delayed_refs_rsv with any changes we may have
 	 * made.

fs/btrfs/extent-tree.c

@@ -1300,13 +1300,29 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len,
 		bytes_left = end - start;
 	}
 
-	if (bytes_left) {
+	while (bytes_left) {
+		u64 bytes_to_discard = min(BTRFS_MAX_DISCARD_CHUNK_SIZE, bytes_left);
+
 		ret = blkdev_issue_discard(bdev, start >> SECTOR_SHIFT,
-					   bytes_left >> SECTOR_SHIFT,
+					   bytes_to_discard >> SECTOR_SHIFT,
 					   GFP_NOFS);
-		if (!ret)
-			*discarded_bytes += bytes_left;
+
+		if (ret) {
+			if (ret != -EOPNOTSUPP)
+				break;
+			continue;
+		}
+
+		start += bytes_to_discard;
+		bytes_left -= bytes_to_discard;
+		*discarded_bytes += bytes_to_discard;
+
+		if (btrfs_trim_interrupted()) {
+			ret = -ERESTARTSYS;
+			break;
+		}
 	}
+
 	return ret;
 }
 
@@ -6459,7 +6475,7 @@ static int btrfs_trim_free_extents(struct btrfs_device *device, u64 *trimmed)
 		start += len;
 		*trimmed += bytes;
 
-		if (fatal_signal_pending(current)) {
+		if (btrfs_trim_interrupted()) {
 			ret = -ERESTARTSYS;
 			break;
 		}

fs/btrfs/extent_io.c

@@ -1306,6 +1306,13 @@ static int submit_one_sector(struct btrfs_inode *inode,
 	free_extent_map(em);
 	em = NULL;
 
+	/*
+	 * Although the PageDirty bit is cleared before entering this
+	 * function, subpage dirty bit is not cleared.
+	 * So clear subpage dirty bit here so next time we won't submit
+	 * a folio for a range already written to disk.
+	 */
+	btrfs_folio_clear_dirty(fs_info, folio, filepos, sectorsize);
 	btrfs_set_range_writeback(inode, filepos, filepos + sectorsize - 1);
 	/*
 	 * Above call should set the whole folio with writeback flag, even
@@ -1315,13 +1322,6 @@ static int submit_one_sector(struct btrfs_inode *inode,
 	 */
 	ASSERT(folio_test_writeback(folio));
 
-	/*
-	 * Although the PageDirty bit is cleared before entering this
-	 * function, subpage dirty bit is not cleared.
-	 * So clear subpage dirty bit here so next time we won't submit
-	 * folio for range already written to disk.
-	 */
-	btrfs_folio_clear_dirty(fs_info, folio, filepos, sectorsize);
 	submit_extent_folio(bio_ctrl, disk_bytenr, folio,
 			    sectorsize, filepos - folio_pos(folio));
 	return 0;

fs/btrfs/free-space-cache.c

@@ -3809,7 +3809,7 @@ static int trim_no_bitmap(struct btrfs_block_group *block_group,
 		if (async && *total_trimmed)
 			break;
 
-		if (fatal_signal_pending(current)) {
+		if (btrfs_trim_interrupted()) {
 			ret = -ERESTARTSYS;
 			break;
 		}
@@ -4000,7 +4000,7 @@ static int trim_bitmaps(struct btrfs_block_group *block_group,
 		}
 		block_group->discard_cursor = start;
 
-		if (fatal_signal_pending(current)) {
+		if (btrfs_trim_interrupted()) {
 			if (start != offset)
 				reset_trimming_bitmap(ctl, offset);
 			ret = -ERESTARTSYS;

fs/btrfs/free-space-cache.h

@@ -10,6 +10,7 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/mutex.h>
+#include <linux/freezer.h>
 #include "fs.h"
 
 struct inode;
@@ -56,6 +57,11 @@ static inline bool btrfs_free_space_trimming_bitmap(
 	return (info->trim_state == BTRFS_TRIM_STATE_TRIMMING);
 }
 
+static inline bool btrfs_trim_interrupted(void)
+{
+	return fatal_signal_pending(current) || freezing(current);
+}
+
 /*
  * Deltas are an effective way to populate global statistics.  Give macro names
  * to make it clear what we're doing.  An example is discard_extents in

fs/btrfs/volumes.h

@@ -30,6 +30,12 @@ struct btrfs_zoned_device_info;
 
 #define BTRFS_MAX_DATA_CHUNK_SIZE	(10ULL * SZ_1G)
 
+/*
+ * Arbitratry maximum size of one discard request to limit potentially long time
+ * spent in blkdev_issue_discard().
+ */
+#define BTRFS_MAX_DISCARD_CHUNK_SIZE	(SZ_1G)
+
 extern struct mutex uuid_mutex;
 
 #define BTRFS_STRIPE_LEN		SZ_64K

fs/btrfs/zoned.c

@@ -1340,7 +1340,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 	switch (zone.cond) {
 	case BLK_ZONE_COND_OFFLINE:
 	case BLK_ZONE_COND_READONLY:
-		btrfs_err(fs_info,
+		btrfs_err_in_rcu(fs_info,
 		"zoned: offline/readonly zone %llu on device %s (devid %llu)",
 			  (info->physical >> device->zone_info->zone_size_shift),
 			  rcu_str_deref(device->name), device->devid);