Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates from Ted Ts'o:
 "Various cleanups and bug fixes in ext4's extent status tree,
  journalling, and block allocator subsystems.

  Also improve performance for parallel DIO overwrites"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (55 commits)
  ext4: avoid updating the superblock on a r/o mount if not needed
  jbd2: skip reading super block if it has been verified
  ext4: fix to check return value of freeze_bdev() in ext4_shutdown()
  ext4: refactoring to use the unified helper ext4_quotas_off()
  ext4: turn quotas off if mount failed after enabling quotas
  ext4: update doc about journal superblock description
  ext4: add journal cycled recording support
  jbd2: continue to record log between each mount
  jbd2: remove j_format_version
  jbd2: factor out journal initialization from journal_get_superblock()
  jbd2: switch to check format version in superblock directly
  jbd2: remove unused feature macros
  ext4: ext4_put_super: Remove redundant checking for 'sbi->s_journal_bdev'
  ext4: Fix reusing stale buffer heads from last failed mounting
  ext4: allow concurrent unaligned dio overwrites
  ext4: clean up mballoc criteria comments
  ext4: make ext4_zeroout_es() return void
  ext4: make ext4_es_insert_extent() return void
  ext4: make ext4_es_insert_delayed_block() return void
  ext4: make ext4_es_remove_extent() return void
  ...

16 files changed, 883 insertions, 578 deletions

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 02fa8a64dc3f..0a2d55faa095 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -128,6 +128,58 @@ enum SHIFT_DIRECTION {
 };
 
 /*
+ * For each criteria, mballoc has slightly different way of finding
+ * the required blocks nad usually, higher the criteria the slower the
+ * allocation.  We start at lower criterias and keep falling back to
+ * higher ones if we are not able to find any blocks.  Lower (earlier)
+ * criteria are faster.
+ */
+enum criteria {
+	/*
+	 * Used when number of blocks needed is a power of 2. This
+	 * doesn't trigger any disk IO except prefetch and is the
+	 * fastest criteria.
+	 */
+	CR_POWER2_ALIGNED,
+
+	/*
+	 * Tries to lookup in-memory data structures to find the most
+	 * suitable group that satisfies goal request. No disk IO
+	 * except block prefetch.
+	 */
+	CR_GOAL_LEN_FAST,
+
+        /*
+	 * Same as CR_GOAL_LEN_FAST but is allowed to reduce the goal
+         * length to the best available length for faster allocation.
+	 */
+	CR_BEST_AVAIL_LEN,
+
+	/*
+	 * Reads each block group sequentially, performing disk IO if
+	 * necessary, to find find_suitable block group. Tries to
+	 * allocate goal length but might trim the request if nothing
+	 * is found after enough tries.
+	 */
+	CR_GOAL_LEN_SLOW,
+
+	/*
+	 * Finds the first free set of blocks and allocates
+	 * those. This is only used in rare cases when
+	 * CR_GOAL_LEN_SLOW also fails to allocate anything.
+	 */
+	CR_ANY_FREE,
+
+	/*
+	 * Number of criterias defined.
+	 */
+	EXT4_MB_NUM_CRS
+};
+
+/* criteria below which we use fast block scanning and avoid unnecessary IO */
+#define CR_FAST CR_GOAL_LEN_SLOW
+
+/*
  * Flags used in mballoc's allocation_context flags field.
  *
  * Also used to show what's going on for debugging purposes when the
@@ -165,9 +217,12 @@ enum SHIFT_DIRECTION {
 /* Do strict check for free blocks while retrying block allocation */
 #define EXT4_MB_STRICT_CHECK		0x4000
 /* Large fragment size list lookup succeeded at least once for cr = 0 */
-#define EXT4_MB_CR0_OPTIMIZED		0x8000
+#define EXT4_MB_CR_POWER2_ALIGNED_OPTIMIZED		0x8000
 /* Avg fragment size rb tree lookup succeeded at least once for cr = 1 */
-#define EXT4_MB_CR1_OPTIMIZED		0x00010000
+#define EXT4_MB_CR_GOAL_LEN_FAST_OPTIMIZED		0x00010000
+/* Avg fragment size rb tree lookup succeeded at least once for cr = 1.5 */
+#define EXT4_MB_CR_BEST_AVAIL_LEN_OPTIMIZED		0x00020000
+
 struct ext4_allocation_request {
 	/* target inode for block we're allocating */
 	struct inode *inode;
@@ -1532,21 +1587,25 @@ struct ext4_sb_info {
 	unsigned long s_mb_last_start;
 	unsigned int s_mb_prefetch;
 	unsigned int s_mb_prefetch_limit;
+	unsigned int s_mb_best_avail_max_trim_order;
 
 	/* stats for buddy allocator */
 	atomic_t s_bal_reqs;	/* number of reqs with len > 1 */
 	atomic_t s_bal_success;	/* we found long enough chunks */
 	atomic_t s_bal_allocated;	/* in blocks */
 	atomic_t s_bal_ex_scanned;	/* total extents scanned */
+	atomic_t s_bal_cX_ex_scanned[EXT4_MB_NUM_CRS];	/* total extents scanned */
 	atomic_t s_bal_groups_scanned;	/* number of groups scanned */
 	atomic_t s_bal_goals;	/* goal hits */
+	atomic_t s_bal_len_goals;	/* len goal hits */
 	atomic_t s_bal_breaks;	/* too long searches */
 	atomic_t s_bal_2orders;	/* 2^order hits */
-	atomic_t s_bal_cr0_bad_suggestions;
-	atomic_t s_bal_cr1_bad_suggestions;
-	atomic64_t s_bal_cX_groups_considered[4];
-	atomic64_t s_bal_cX_hits[4];
-	atomic64_t s_bal_cX_failed[4];		/* cX loop didn't find blocks */
+	atomic_t s_bal_p2_aligned_bad_suggestions;
+	atomic_t s_bal_goal_fast_bad_suggestions;
+	atomic_t s_bal_best_avail_bad_suggestions;
+	atomic64_t s_bal_cX_groups_considered[EXT4_MB_NUM_CRS];
+	atomic64_t s_bal_cX_hits[EXT4_MB_NUM_CRS];
+	atomic64_t s_bal_cX_failed[EXT4_MB_NUM_CRS];		/* cX loop didn't find blocks */
 	atomic_t s_mb_buddies_generated;	/* number of buddies generated */
 	atomic64_t s_mb_generation_time;
 	atomic_t s_mb_lost_chunks;
@@ -2632,10 +2691,6 @@ extern void ext4_get_group_no_and_offset(struct super_block *sb,
 extern ext4_group_t ext4_get_group_number(struct super_block *sb,
 					  ext4_fsblk_t block);
 
-extern unsigned int ext4_block_group(struct super_block *sb,
-			ext4_fsblk_t blocknr);
-extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb,
-			ext4_fsblk_t blocknr);
 extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group);
 extern unsigned long ext4_bg_num_gdb(struct super_block *sb,
 			ext4_group_t group);
@@ -2841,8 +2896,6 @@ int ext4_fc_record_regions(struct super_block *sb, int ino,
 /* mballoc.c */
 extern const struct seq_operations ext4_mb_seq_groups_ops;
 extern const struct seq_operations ext4_mb_seq_structs_summary_ops;
-extern long ext4_mb_stats;
-extern long ext4_mb_max_to_scan;
 extern int ext4_seq_mb_stats_show(struct seq_file *seq, void *offset);
 extern int ext4_mb_init(struct super_block *);
 extern int ext4_mb_release(struct super_block *);
@@ -3481,14 +3534,8 @@ extern int ext4_try_to_write_inline_data(struct address_space *mapping,
 					 struct inode *inode,
 					 loff_t pos, unsigned len,
 					 struct page **pagep);
-extern int ext4_write_inline_data_end(struct inode *inode,
-				      loff_t pos, unsigned len,
-				      unsigned copied,
-				      struct page *page);
-extern struct buffer_head *
-ext4_journalled_write_inline_data(struct inode *inode,
-				  unsigned len,
-				  struct page *page);
+int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
+			       unsigned copied, struct folio *folio);
 extern int ext4_da_write_inline_data_begin(struct address_space *mapping,
 					   struct inode *inode,
 					   loff_t pos, unsigned len,
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 35703dce23a3..e4115d338f10 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3123,7 +3123,7 @@ void ext4_ext_release(struct super_block *sb)
 #endif
 }
 
-static int ext4_zeroout_es(struct inode *inode, struct ext4_extent *ex)
+static void ext4_zeroout_es(struct inode *inode, struct ext4_extent *ex)
 {
 	ext4_lblk_t  ee_block;
 	ext4_fsblk_t ee_pblock;
@@ -3134,10 +3134,10 @@ static int ext4_zeroout_es(struct inode *inode, struct ext4_extent *ex)
 	ee_pblock = ext4_ext_pblock(ex);
 
 	if (ee_len == 0)
-		return 0;
+		return;
 
-	return ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock,
-				     EXTENT_STATUS_WRITTEN);
+	ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock,
+			      EXTENT_STATUS_WRITTEN);
 }
 
 /* FIXME!! we need to try to merge to left or right after zero-out  */
@@ -3287,7 +3287,7 @@ static int ext4_split_extent_at(handle_t *handle,
 			err = ext4_ext_dirty(handle, inode, path + path->p_depth);
 			if (!err)
 				/* update extent status tree */
-				err = ext4_zeroout_es(inode, &zero_ex);
+				ext4_zeroout_es(inode, &zero_ex);
 			/* If we failed at this point, we don't know in which
 			 * state the extent tree exactly is so don't try to fix
 			 * length of the original extent as it may do even more
@@ -3640,9 +3640,8 @@ fallback:
 out:
 	/* If we have gotten a failure, don't zero out status tree */
 	if (!err) {
-		err = ext4_zeroout_es(inode, &zero_ex1);
-		if (!err)
-			err = ext4_zeroout_es(inode, &zero_ex2);
+		ext4_zeroout_es(inode, &zero_ex1);
+		ext4_zeroout_es(inode, &zero_ex2);
 	}
 	return err ? err : allocated;
 }
@@ -4403,15 +4402,8 @@ int ext4_ext_truncate(handle_t *handle, struct inode *inode)
 
 	last_block = (inode->i_size + sb->s_blocksize - 1)
 			>> EXT4_BLOCK_SIZE_BITS(sb);
-retry:
-	err = ext4_es_remove_extent(inode, last_block,
-				    EXT_MAX_BLOCKS - last_block);
-	if (err == -ENOMEM) {
-		memalloc_retry_wait(GFP_ATOMIC);
-		goto retry;
-	}
-	if (err)
-		return err;
+	ext4_es_remove_extent(inode, last_block, EXT_MAX_BLOCKS - last_block);
+
 retry_remove_space:
 	err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1);
 	if (err == -ENOMEM) {
@@ -5363,13 +5355,7 @@ static int ext4_collapse_range(struct file *file, loff_t offset, loff_t len)
 
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ext4_discard_preallocations(inode, 0);
-
-	ret = ext4_es_remove_extent(inode, punch_start,
-				    EXT_MAX_BLOCKS - punch_start);
-	if (ret) {
-		up_write(&EXT4_I(inode)->i_data_sem);
-		goto out_stop;
-	}
+	ext4_es_remove_extent(inode, punch_start, EXT_MAX_BLOCKS - punch_start);
 
 	ret = ext4_ext_remove_space(inode, punch_start, punch_stop - 1);
 	if (ret) {
@@ -5547,12 +5533,7 @@ static int ext4_insert_range(struct file *file, loff_t offset, loff_t len)
 		ext4_free_ext_path(path);
 	}
 
-	ret = ext4_es_remove_extent(inode, offset_lblk,
-			EXT_MAX_BLOCKS - offset_lblk);
-	if (ret) {
-		up_write(&EXT4_I(inode)->i_data_sem);
-		goto out_stop;
-	}
+	ext4_es_remove_extent(inode, offset_lblk, EXT_MAX_BLOCKS - offset_lblk);
 
 	/*
 	 * if offset_lblk lies in a hole which is at start of file, use
@@ -5610,12 +5591,8 @@ ext4_swap_extents(handle_t *handle, struct inode *inode1,
 	BUG_ON(!inode_is_locked(inode1));
 	BUG_ON(!inode_is_locked(inode2));
 
-	*erp = ext4_es_remove_extent(inode1, lblk1, count);
-	if (unlikely(*erp))
-		return 0;
-	*erp = ext4_es_remove_extent(inode2, lblk2, count);
-	if (unlikely(*erp))
-		return 0;
+	ext4_es_remove_extent(inode1, lblk1, count);
+	ext4_es_remove_extent(inode2, lblk2, count);
 
 	while (count) {
 		struct ext4_extent *ex1, *ex2, tmp_ex;
diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c
index 595abb9e7d74..9b5b8951afb4 100644
--- a/fs/ext4/extents_status.c
+++ b/fs/ext4/extents_status.c
@@ -144,9 +144,11 @@
 static struct kmem_cache *ext4_es_cachep;
 static struct kmem_cache *ext4_pending_cachep;
 
-static int __es_insert_extent(struct inode *inode, struct extent_status *newes);
+static int __es_insert_extent(struct inode *inode, struct extent_status *newes,
+			      struct extent_status *prealloc);
 static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
-			      ext4_lblk_t end, int *reserved);
+			      ext4_lblk_t end, int *reserved,
+			      struct extent_status *prealloc);
 static int es_reclaim_extents(struct ext4_inode_info *ei, int *nr_to_scan);
 static int __es_shrink(struct ext4_sb_info *sbi, int nr_to_scan,
 		       struct ext4_inode_info *locked_ei);
@@ -446,22 +448,36 @@ static void ext4_es_list_del(struct inode *inode)
 	spin_unlock(&sbi->s_es_lock);
 }
 
-static struct extent_status *
-ext4_es_alloc_extent(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len,
-		     ext4_fsblk_t pblk)
+/*
+ * Returns true if we cannot fail to allocate memory for this extent_status
+ * entry and cannot reclaim it until its status changes.
+ */
+static inline bool ext4_es_must_keep(struct extent_status *es)
+{
+	/* fiemap, bigalloc, and seek_data/hole need to use it. */
+	if (ext4_es_is_delayed(es))
+		return true;
+
+	return false;
+}
+
+static inline struct extent_status *__es_alloc_extent(bool nofail)
+{
+	if (!nofail)
+		return kmem_cache_alloc(ext4_es_cachep, GFP_ATOMIC);
+
+	return kmem_cache_zalloc(ext4_es_cachep, GFP_KERNEL | __GFP_NOFAIL);
+}
+
+static void ext4_es_init_extent(struct inode *inode, struct extent_status *es,
+		ext4_lblk_t lblk, ext4_lblk_t len, ext4_fsblk_t pblk)
 {
-	struct extent_status *es;
-	es = kmem_cache_alloc(ext4_es_cachep, GFP_ATOMIC);
-	if (es == NULL)
-		return NULL;
 	es->es_lblk = lblk;
 	es->es_len = len;
 	es->es_pblk = pblk;
 
-	/*
-	 * We don't count delayed extent because we never try to reclaim them
-	 */
-	if (!ext4_es_is_delayed(es)) {
+	/* We never try to reclaim a must kept extent, so we don't count it. */
+	if (!ext4_es_must_keep(es)) {
 		if (!EXT4_I(inode)->i_es_shk_nr++)
 			ext4_es_list_add(inode);
 		percpu_counter_inc(&EXT4_SB(inode->i_sb)->
@@ -470,8 +486,11 @@ ext4_es_alloc_extent(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len,
 
 	EXT4_I(inode)->i_es_all_nr++;
 	percpu_counter_inc(&EXT4_SB(inode->i_sb)->s_es_stats.es_stats_all_cnt);
+}
 
-	return es;
+static inline void __es_free_extent(struct extent_status *es)
+{
+	kmem_cache_free(ext4_es_cachep, es);
 }
 
 static void ext4_es_free_extent(struct inode *inode, struct extent_status *es)
@@ -479,8 +498,8 @@ static void ext4_es_free_extent(struct inode *inode, struct extent_status *es)
 	EXT4_I(inode)->i_es_all_nr--;
 	percpu_counter_dec(&EXT4_SB(inode->i_sb)->s_es_stats.es_stats_all_cnt);
 
-	/* Decrease the shrink counter when this es is not delayed */
-	if (!ext4_es_is_delayed(es)) {
+	/* Decrease the shrink counter when we can reclaim the extent. */
+	if (!ext4_es_must_keep(es)) {
 		BUG_ON(EXT4_I(inode)->i_es_shk_nr == 0);
 		if (!--EXT4_I(inode)->i_es_shk_nr)
 			ext4_es_list_del(inode);
@@ -488,7 +507,7 @@ static void ext4_es_free_extent(struct inode *inode, struct extent_status *es)
 					s_es_stats.es_stats_shk_cnt);
 	}
 
-	kmem_cache_free(ext4_es_cachep, es);
+	__es_free_extent(es);
 }
 
 /*
@@ -749,7 +768,8 @@ static inline void ext4_es_insert_extent_check(struct inode *inode,
 }
 #endif
 
-static int __es_insert_extent(struct inode *inode, struct extent_status *newes)
+static int __es_insert_extent(struct inode *inode, struct extent_status *newes,
+			      struct extent_status *prealloc)
 {
 	struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree;
 	struct rb_node **p = &tree->root.rb_node;
@@ -789,10 +809,15 @@ static int __es_insert_extent(struct inode *inode, struct extent_status *newes)
 		}
 	}
 
-	es = ext4_es_alloc_extent(inode, newes->es_lblk, newes->es_len,
-				  newes->es_pblk);
+	if (prealloc)
+		es = prealloc;
+	else
+		es = __es_alloc_extent(false);
 	if (!es)
 		return -ENOMEM;
+	ext4_es_init_extent(inode, es, newes->es_lblk, newes->es_len,
+			    newes->es_pblk);
+
 	rb_link_node(&es->rb_node, parent, p);
 	rb_insert_color(&es->rb_node, &tree->root);
 
@@ -804,26 +829,27 @@ out:
 /*
  * ext4_es_insert_extent() adds information to an inode's extent
  * status tree.
- *
- * Return 0 on success, error code on failure.
  */
-int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
-			  ext4_lblk_t len, ext4_fsblk_t pblk,
-			  unsigned int status)
+void ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
+			   ext4_lblk_t len, ext4_fsblk_t pblk,
+			   unsigned int status)
 {
 	struct extent_status newes;
 	ext4_lblk_t end = lblk + len - 1;
-	int err = 0;
+	int err1 = 0;
+	int err2 = 0;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+	struct extent_status *es1 = NULL;
+	struct extent_status *es2 = NULL;
 
 	if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY)
-		return 0;
+		return;
 
 	es_debug("add [%u/%u) %llu %x to extent status tree of inode %lu\n",
 		 lblk, len, pblk, status, inode->i_ino);
 
 	if (!len)
-		return 0;
+		return;
 
 	BUG_ON(end < lblk);
 
@@ -842,29 +868,40 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
 
 	ext4_es_insert_extent_check(inode, &newes);
 
+retry:
+	if (err1 && !es1)
+		es1 = __es_alloc_extent(true);
+	if ((err1 || err2) && !es2)
+		es2 = __es_alloc_extent(true);
 	write_lock(&EXT4_I(inode)->i_es_lock);
-	err = __es_remove_extent(inode, lblk, end, NULL);
-	if (err != 0)
+
+	err1 = __es_remove_extent(inode, lblk, end, NULL, es1);
+	if (err1 != 0)
+		goto error;
+
+	err2 = __es_insert_extent(inode, &newes, es2);
+	if (err2 == -ENOMEM && !ext4_es_must_keep(&newes))
+		err2 = 0;
+	if (err2 != 0)
 		goto error;
-retry:
-	err = __es_insert_extent(inode, &newes);
-	if (err == -ENOMEM && __es_shrink(EXT4_SB(inode->i_sb),
-					  128, EXT4_I(inode)))
-		goto retry;
-	if (err == -ENOMEM && !ext4_es_is_delayed(&newes))
-		err = 0;
 
 	if (sbi->s_cluster_ratio > 1 && test_opt(inode->i_sb, DELALLOC) &&
 	    (status & EXTENT_STATUS_WRITTEN ||
 	     status & EXTENT_STATUS_UNWRITTEN))
 		__revise_pending(inode, lblk, len);
 
+	/* es is pre-allocated but not used, free it. */
+	if (es1 && !es1->es_len)
+		__es_free_extent(es1);
+	if (es2 && !es2->es_len)
+		__es_free_extent(es2);
 error:
 	write_unlock(&EXT4_I(inode)->i_es_lock);
+	if (err1 || err2)
+		goto retry;
 
 	ext4_es_print_tree(inode);
-
-	return err;
+	return;
 }
 
 /*
@@ -897,7 +934,7 @@ void ext4_es_cache_extent(struct inode *inode, ext4_lblk_t lblk,
 
 	es = __es_tree_search(&EXT4_I(inode)->i_es_tree.root, lblk);
 	if (!es || es->es_lblk > end)
-		__es_insert_extent(inode, &newes);
+		__es_insert_extent(inode, &newes, NULL);
 	write_unlock(&EXT4_I(inode)->i_es_lock);
 }
 
@@ -1287,6 +1324,7 @@ static unsigned int get_rsvd(struct inode *inode, ext4_lblk_t end,
  * @lblk - first block in range
  * @end - last block in range
  * @reserved - number of cluster reservations released
+ * @prealloc - pre-allocated es to avoid memory allocation failures
  *
  * If @reserved is not NULL and delayed allocation is enabled, counts
  * block/cluster reservations freed by removing range and if bigalloc
@@ -1294,7 +1332,8 @@ static unsigned int get_rsvd(struct inode *inode, ext4_lblk_t end,
  * error code on failure.
  */
 static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
-			      ext4_lblk_t end, int *reserved)
+			      ext4_lblk_t end, int *reserved,
+			      struct extent_status *prealloc)
 {
 	struct ext4_es_tree *tree = &EXT4_I(inode)->i_es_tree;
 	struct rb_node *node;
@@ -1302,14 +1341,12 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 	struct extent_status orig_es;
 	ext4_lblk_t len1, len2;
 	ext4_fsblk_t block;
-	int err;
+	int err = 0;
 	bool count_reserved = true;
 	struct rsvd_count rc;
 
 	if (reserved == NULL || !test_opt(inode->i_sb, DELALLOC))
 		count_reserved = false;
-retry:
-	err = 0;
 
 	es = __es_tree_search(&tree->root, lblk);
 	if (!es)
@@ -1343,14 +1380,13 @@ retry:
 					orig_es.es_len - len2;
 			ext4_es_store_pblock_status(&newes, block,
 						    ext4_es_status(&orig_es));
-			err = __es_insert_extent(inode, &newes);
+			err = __es_insert_extent(inode, &newes, prealloc);
 			if (err) {
+				if (!ext4_es_must_keep(&newes))
+					return 0;
+
 				es->es_lblk = orig_es.es_lblk;
 				es->es_len = orig_es.es_len;
-				if ((err == -ENOMEM) &&
-				    __es_shrink(EXT4_SB(inode->i_sb),
-							128, EXT4_I(inode)))
-					goto retry;
 				goto out;
 			}
 		} else {
@@ -1422,39 +1458,48 @@ out:
  * @len - number of blocks to remove
  *
  * Reduces block/cluster reservation count and for bigalloc cancels pending
- * reservations as needed. Returns 0 on success, error code on failure.
+ * reservations as needed.
  */
-int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
-			  ext4_lblk_t len)
+void ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
+			   ext4_lblk_t len)
 {
 	ext4_lblk_t end;
 	int err = 0;
 	int reserved = 0;
+	struct extent_status *es = NULL;
 
 	if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY)
-		return 0;
+		return;
 
 	trace_ext4_es_remove_extent(inode, lblk, len);
 	es_debug("remove [%u/%u) from extent status tree of inode %lu\n",
 		 lblk, len, inode->i_ino);
 
 	if (!len)
-		return err;
+		return;
 
 	end = lblk + len - 1;
 	BUG_ON(end < lblk);
 
+retry:
+	if (err && !es)
+		es = __es_alloc_extent(true);
 	/*
 	 * ext4_clear_inode() depends on us taking i_es_lock unconditionally
 	 * so that we are sure __es_shrink() is done with the inode before it
 	 * is reclaimed.
 	 */
 	write_lock(&EXT4_I(inode)->i_es_lock);
-	err = __es_remove_extent(inode, lblk, end, &reserved);
+	err = __es_remove_extent(inode, lblk, end, &reserved, es);
+	if (es && !es->es_len)
+		__es_free_extent(es);
 	write_unlock(&EXT4_I(inode)->i_es_lock);
+	if (err)
+		goto retry;
+
 	ext4_es_print_tree(inode);
 	ext4_da_release_space(inode, reserved);
-	return err;
+	return;
 }
 
 static int __es_shrink(struct ext4_sb_info *sbi, int nr_to_scan,
@@ -1702,11 +1747,8 @@ static int es_do_reclaim_extents(struct ext4_inode_info *ei, ext4_lblk_t end,
 
 		(*nr_to_scan)--;
 		node = rb_next(&es->rb_node);
-		/*
-		 * We can't reclaim delayed extent from status tree because
-		 * fiemap, bigallic, and seek_data/hole need to use it.
-		 */
-		if (ext4_es_is_delayed(es))
+
+		if (ext4_es_must_keep(es))
 			goto next;
 		if (ext4_es_is_referenced(es)) {
 			ext4_es_clear_referenced(es);
@@ -1770,7 +1812,7 @@ void ext4_clear_inode_es(struct inode *inode)
 	while (node) {
 		es = rb_entry(node, struct extent_status, rb_node);
 		node = rb_next(node);
-		if (!ext4_es_is_delayed(es)) {
+		if (!ext4_es_must_keep(es)) {
 			rb_erase(&es->rb_node, &tree->root);
 			ext4_es_free_extent(inode, es);
 		}
@@ -1972,17 +2014,18 @@ bool ext4_is_pending(struct inode *inode, ext4_lblk_t lblk)
  * @lblk - logical block to be added
  * @allocated - indicates whether a physical cluster has been allocated for
  *              the logical cluster that contains the block
- *
- * Returns 0 on success, negative error code on failure.
  */
-int ext4_es_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk,
-				 bool allocated)
+void ext4_es_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk,
+				  bool allocated)
 {
 	struct extent_status newes;
-	int err = 0;
+	int err1 = 0;
+	int err2 = 0;
+	struct extent_status *es1 = NULL;
+	struct extent_status *es2 = NULL;
 
 	if (EXT4_SB(inode->i_sb)->s_mount_state & EXT4_FC_REPLAY)
-		return 0;
+		return;
 
 	es_debug("add [%u/1) delayed to extent status tree of inode %lu\n",
 		 lblk, inode->i_ino);
@@ -1994,29 +2037,37 @@ int ext4_es_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk,
 
 	ext4_es_insert_extent_check(inode, &newes);
 
+retry:
+	if (err1 && !es1)
+		es1 = __es_alloc_extent(true);
+	if ((err1 || err2) && !es2)
+		es2 = __es_alloc_extent(true);
 	write_lock(&EXT4_I(inode)->i_es_lock);
 
-	err = __es_remove_extent(inode, lblk, lblk, NULL);
-	if (err != 0)
+	err1 = __es_remove_extent(inode, lblk, lblk, NULL, es1);
+	if (err1 != 0)
 		goto error;
-retry:
-	err = __es_insert_extent(inode, &newes);
-	if (err == -ENOMEM && __es_shrink(EXT4_SB(inode->i_sb),
-					  128, EXT4_I(inode)))
-		goto retry;
-	if (err != 0)
+
+	err2 = __es_insert_extent(inode, &newes, es2);
+	if (err2 != 0)
 		goto error;
 
 	if (allocated)
 		__insert_pending(inode, lblk);
 
+	/* es is pre-allocated but not used, free it. */
+	if (es1 && !es1->es_len)
+		__es_free_extent(es1);
+	if (es2 && !es2->es_len)
+		__es_free_extent(es2);
 error:
 	write_unlock(&EXT4_I(inode)->i_es_lock);
+	if (err1 || err2)
+		goto retry;
 
 	ext4_es_print_tree(inode);
 	ext4_print_pending_tree(inode);
-
-	return err;
+	return;
 }
 
 /*
diff --git a/fs/ext4/extents_status.h b/fs/ext4/extents_status.h
index 4ec30a798260..d9847a4a25db 100644
--- a/fs/ext4/extents_status.h
+++ b/fs/ext4/extents_status.h
@@ -127,14 +127,14 @@ extern int __init ext4_init_es(void);
 extern void ext4_exit_es(void);
 extern void ext4_es_init_tree(struct ext4_es_tree *tree);
 
-extern int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
-				 ext4_lblk_t len, ext4_fsblk_t pblk,
-				 unsigned int status);
+extern void ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
+				  ext4_lblk_t len, ext4_fsblk_t pblk,
+				  unsigned int status);
 extern void ext4_es_cache_extent(struct inode *inode, ext4_lblk_t lblk,
 				 ext4_lblk_t len, ext4_fsblk_t pblk,
 				 unsigned int status);
-extern int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
-				 ext4_lblk_t len);
+extern void ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
+				  ext4_lblk_t len);
 extern void ext4_es_find_extent_range(struct inode *inode,
 				      int (*match_fn)(struct extent_status *es),
 				      ext4_lblk_t lblk, ext4_lblk_t end,
@@ -249,8 +249,8 @@ extern void ext4_exit_pending(void);
 extern void ext4_init_pending_tree(struct ext4_pending_tree *tree);
 extern void ext4_remove_pending(struct inode *inode, ext4_lblk_t lblk);
 extern bool ext4_is_pending(struct inode *inode, ext4_lblk_t lblk);
-extern int ext4_es_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk,
-					bool allocated);
+extern void ext4_es_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk,
+					 bool allocated);
 extern unsigned int ext4_es_delayed_clu(struct inode *inode, ext4_lblk_t lblk,
 					ext4_lblk_t len);
 extern void ext4_clear_inode_es(struct inode *inode);
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 6a16d07965f9..c457c8517f0f 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -450,13 +450,14 @@ static const struct iomap_dio_ops ext4_dio_write_ops = {
  */
 static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from,
 				     bool *ilock_shared, bool *extend,
-				     bool *unwritten)
+				     bool *unwritten, int *dio_flags)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
 	loff_t offset;
 	size_t count;
 	ssize_t ret;
+	bool overwrite, unaligned_io;
 
 restart:
 	ret = ext4_generic_write_checks(iocb, from);
@@ -465,16 +466,20 @@ restart:
 
 	offset = iocb->ki_pos;
 	count = ret;
-	if (ext4_extending_io(inode, offset, count))
-		*extend = true;
+
+	unaligned_io = ext4_unaligned_io(inode, from, offset);
+	*extend = ext4_extending_io(inode, offset, count);
+	overwrite = ext4_overwrite_io(inode, offset, count, unwritten);
+
 	/*
-	 * Determine whether the IO operation will overwrite allocated
-	 * and initialized blocks.
-	 * We need exclusive i_rwsem for changing security info
-	 * in file_modified().
+	 * Determine whether we need to upgrade to an exclusive lock. This is
+	 * required to change security info in file_modified(), for extending
+	 * I/O, any form of non-overwrite I/O, and unaligned I/O to unwritten
+	 * extents (as partial block zeroing may be required).
 	 */
-	if (*ilock_shared && (!IS_NOSEC(inode) || *extend ||
-	     !ext4_overwrite_io(inode, offset, count, unwritten))) {
+	if (*ilock_shared &&
+	    ((!IS_NOSEC(inode) || *extend || !overwrite ||
+	     (unaligned_io && *unwritten)))) {
 		if (iocb->ki_flags & IOCB_NOWAIT) {
 			ret = -EAGAIN;
 			goto out;
@@ -485,6 +490,32 @@ restart:
 		goto restart;
 	}
 
+	/*
+	 * Now that locking is settled, determine dio flags and exclusivity
+	 * requirements. Unaligned writes are allowed under shared lock so long
+	 * as they are pure overwrites. Set the iomap overwrite only flag as an
+	 * added precaution in this case. Even though this is unnecessary, we
+	 * can detect and warn on unexpected -EAGAIN if an unsafe unaligned
+	 * write is ever submitted.
+	 *
+	 * Otherwise, concurrent unaligned writes risk data corruption due to
+	 * partial block zeroing in the dio layer, and so the I/O must occur
+	 * exclusively. The inode lock is already held exclusive if the write is
+	 * non-overwrite or extending, so drain all outstanding dio and set the
+	 * force wait dio flag.
+	 */
+	if (*ilock_shared && unaligned_io) {
+		*dio_flags = IOMAP_DIO_OVERWRITE_ONLY;
+	} else if (!*ilock_shared && (unaligned_io || *extend)) {
+		if (iocb->ki_flags & IOCB_NOWAIT) {
+			ret = -EAGAIN;
+			goto out;
+		}
+		if (unaligned_io && (!overwrite || *unwritten))
+			inode_dio_wait(inode);
+		*dio_flags = IOMAP_DIO_FORCE_WAIT;
+	}
+
 	ret = file_modified(file);
 	if (ret < 0)
 		goto out;
@@ -506,18 +537,11 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	loff_t offset = iocb->ki_pos;
 	size_t count = iov_iter_count(from);
 	const struct iomap_ops *iomap_ops = &ext4_iomap_ops;
-	bool extend = false, unaligned_io = false, unwritten = false;
+	bool extend = false, unwritten = false;
 	bool ilock_shared = true;
+	int dio_flags = 0;
 
 	/*
-	 * We initially start with shared inode lock unless it is
-	 * unaligned IO which needs exclusive lock anyways.
-	 */
-	if (ext4_unaligned_io(inode, from, offset)) {
-		unaligned_io = true;
-		ilock_shared = false;
-	}
-	/*
 	 * Quick check here without any i_rwsem lock to see if it is extending
 	 * IO. A more reliable check is done in ext4_dio_write_checks() with
 	 * proper locking in place.
@@ -549,16 +573,11 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		return ext4_buffered_write_iter(iocb, from);
 	}
 
-	ret = ext4_dio_write_checks(iocb, from,
-				    &ilock_shared, &extend, &unwritten);
+	ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend,
+				    &unwritten, &dio_flags);
 	if (ret <= 0)
 		return ret;
 
-	/* if we're going to block and IOCB_NOWAIT is set, return -EAGAIN */
-	if ((iocb->ki_flags & IOCB_NOWAIT) && (unaligned_io || extend)) {
-		ret = -EAGAIN;
-		goto out;
-	}
 	/*
 	 * Make sure inline data cannot be created anymore since we are going
 	 * to allocate blocks for DIO. We know the inode does not have any
@@ -569,19 +588,6 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	offset = iocb->ki_pos;
 	count = ret;
 
-	/*
-	 * Unaligned direct IO must be serialized among each other as zeroing
-	 * of partial blocks of two competing unaligned IOs can result in data
-	 * corruption.
-	 *
-	 * So we make sure we don't allow any unaligned IO in flight.
-	 * For IOs where we need not wait (like unaligned non-AIO DIO),
-	 * below inode_dio_wait() may anyway become a no-op, since we start
-	 * with exclusive lock.
-	 */
-	if (unaligned_io)
-		inode_dio_wait(inode);
-
 	if (extend) {
 		handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
 		if (IS_ERR(handle)) {
@@ -601,8 +607,8 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (ilock_shared && !unwritten)
 		iomap_ops = &ext4_iomap_overwrite_ops;
 	ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops,
-			   (unaligned_io || extend) ? IOMAP_DIO_FORCE_WAIT : 0,
-