ext4: add prefetch_block_bitmaps mount option

For file systems where we can afford to keep the buddy bitmaps cached,
we can speed up initial writes to large file systems by starting to
load the block allocation bitmaps as soon as the file system is
mounted.  This won't work well for _super_ large file systems, or
memory constrained systems, so we only enable this when it is
requested via a mount option.

Addresses-Google-Bug: 159488342
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>

3 files changed, 59 insertions, 23 deletions

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 26ae31a994a2..e5f0c66a6156 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1188,6 +1188,7 @@ struct ext4_inode_info {
 #define EXT4_MOUNT_JOURNAL_CHECKSUM	0x800000 /* Journal checksums */
 #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT	0x1000000 /* Journal Async Commit */
 #define EXT4_MOUNT_WARN_ON_ERROR	0x2000000 /* Trigger WARN_ON on error */
+#define EXT4_MOUNT_PREFETCH_BLOCK_BITMAPS 0x4000000
 #define EXT4_MOUNT_DELALLOC		0x8000000 /* Delalloc support */
 #define EXT4_MOUNT_DATA_ERR_ABORT	0x10000000 /* Abort on file data write */
 #define EXT4_MOUNT_BLOCK_VALIDITY	0x20000000 /* Block validity checking */
@@ -2334,9 +2335,15 @@ struct ext4_lazy_init {
 	struct mutex		li_list_mtx;
 };
 
+enum ext4_li_mode {
+	EXT4_LI_MODE_PREFETCH_BBITMAP,
+	EXT4_LI_MODE_ITABLE,
+};
+
 struct ext4_li_request {
 	struct super_block	*lr_super;
-	struct ext4_sb_info	*lr_sbi;
+	enum ext4_li_mode	lr_mode;
+	ext4_group_t		lr_first_not_zeroed;
 	ext4_group_t		lr_next_group;
 	struct list_head	lr_request;
 	unsigned long		lr_next_sched;
@@ -2676,6 +2683,12 @@ extern int ext4_mb_reserve_blocks(struct super_block *, int);
 extern void ext4_discard_preallocations(struct inode *);
 extern int __init ext4_init_mballoc(void);
 extern void ext4_exit_mballoc(void);
+extern ext4_group_t ext4_mb_prefetch(struct super_block *sb,
+				     ext4_group_t group,
+				     unsigned int nr, int *cnt);
+extern void ext4_mb_prefetch_fini(struct super_block *sb, ext4_group_t group,
+				  unsigned int nr);
+
 extern void ext4_free_blocks(handle_t *handle, struct inode *inode,
 			     struct buffer_head *bh, ext4_fsblk_t block,
 			     unsigned long count, int flags);
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 9a07da53ab7b..8ecd49dd3906 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2232,9 +2232,8 @@ out:
  * Start prefetching @nr block bitmaps starting at @group.
  * Return the next group which needs to be prefetched.
  */
-static ext4_group_t
-ext4_mb_prefetch(struct super_block *sb, ext4_group_t group,
-		 unsigned int nr, int *cnt)
+ext4_group_t ext4_mb_prefetch(struct super_block *sb, ext4_group_t group,
+			      unsigned int nr, int *cnt)
 {
 	ext4_group_t ngroups = ext4_get_groups_count(sb);
 	struct buffer_head *bh;
@@ -2284,9 +2283,8 @@ ext4_mb_prefetch(struct super_block *sb, ext4_group_t group,
  * waiting for the block allocation bitmap read to finish when
  * ext4_mb_prefetch_fini is called from ext4_mb_regular_allocator().
  */
-static void
-ext4_mb_prefetch_fini(struct super_block *sb, ext4_group_t group,
-		      unsigned int nr)
+void ext4_mb_prefetch_fini(struct super_block *sb, ext4_group_t group,
+			   unsigned int nr)
 {
 	while (nr-- > 0) {
 		struct ext4_group_desc *gdp = ext4_get_group_desc(sb, group,
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 8c00f0a09f4d..1cbe3f248faf 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1521,6 +1521,7 @@ enum {
 	Opt_dioread_nolock, Opt_dioread_lock,
 	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
 	Opt_max_dir_size_kb, Opt_nojournal_checksum, Opt_nombcache,
+	Opt_prefetch_block_bitmaps,
 };
 
 static const match_table_t tokens = {
@@ -1612,6 +1613,7 @@ static const match_table_t tokens = {
 	{Opt_test_dummy_encryption, "test_dummy_encryption"},
 	{Opt_nombcache, "nombcache"},
 	{Opt_nombcache, "no_mbcache"},	/* for backward compatibility */
+	{Opt_prefetch_block_bitmaps, "prefetch_block_bitmaps"},
 	{Opt_removed, "check=none"},	/* mount option from ext2/3 */
 	{Opt_removed, "nocheck"},	/* mount option from ext2/3 */
 	{Opt_removed, "reservation"},	/* mount option from ext2/3 */
@@ -1829,6 +1831,8 @@ static const struct mount_opts {
 	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
 	{Opt_test_dummy_encryption, 0, MOPT_STRING},
 	{Opt_nombcache, EXT4_MOUNT_NO_MBCACHE, MOPT_SET},
+	{Opt_prefetch_block_bitmaps, EXT4_MOUNT_PREFETCH_BLOCK_BITMAPS,
+	 MOPT_SET},
 	{Opt_err, 0, 0}
 };
 
@@ -3201,15 +3205,34 @@ static void print_daily_error_info(struct timer_list *t)
 static int ext4_run_li_request(struct ext4_li_request *elr)
 {
 	struct ext4_group_desc *gdp = NULL;
-	ext4_group_t group, ngroups;
-	struct super_block *sb;
+	struct super_block *sb = elr->lr_super;
+	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
+	ext4_group_t group = elr->lr_next_group;
 	unsigned long timeout = 0;
+	unsigned int prefetch_ios = 0;
 	int ret = 0;
 
-	sb = elr->lr_super;
-	ngroups = EXT4_SB(sb)->s_groups_count;
+	if (elr->lr_mode == EXT4_LI_MODE_PREFETCH_BBITMAP) {
+		elr->lr_next_group = ext4_mb_prefetch(sb, group,
+				EXT4_SB(sb)->s_mb_prefetch, &prefetch_ios);
+		if (prefetch_ios)
+			ext4_mb_prefetch_fini(sb, elr->lr_next_group,
+					      prefetch_ios);
+		trace_ext4_prefetch_bitmaps(sb, group, elr->lr_next_group,
+					    prefetch_ios);
+		if (group >= elr->lr_next_group) {
+			ret = 1;
+			if (elr->lr_first_not_zeroed != ngroups &&
+			    !sb_rdonly(sb) && test_opt(sb, INIT_INODE_TABLE)) {
+				elr->lr_next_group = elr->lr_first_not_zeroed;
+				elr->lr_mode = EXT4_LI_MODE_ITABLE;
+				ret = 0;
+			}
+		}
+		return ret;
+	}
 
-	for (group = elr->lr_next_group; group < ngroups; group++) {
+	for (; group < ngroups; group++) {
 		gdp = ext4_get_group_desc(sb, group, NULL);
 		if (!gdp) {
 			ret = 1;
@@ -3227,9 +3250,10 @@ static int ext4_run_li_request(struct ext4_li_request *elr)
 		timeout = jiffies;
 		ret = ext4_init_inode_table(sb, group,
 					    elr->lr_timeout ? 0 : 1);
+		trace_ext4_lazy_itable_init(sb, group);
 		if (elr->lr_timeout == 0) {
 			timeout = (jiffies - timeout) *
-				  elr->lr_sbi->s_li_wait_mult;
+				EXT4_SB(elr->lr_super)->s_li_wait_mult;
 			elr->lr_timeout = timeout;
 		}
 		elr->lr_next_sched = jiffies + elr->lr_timeout;
@@ -3244,15 +3268,11 @@ static int ext4_run_li_request(struct ext4_li_request *elr)
  */
 static void ext4_remove_li_request(struct ext4_li_request *elr)
 {
-	struct ext4_sb_info *sbi;
-
 	if (!elr)
 		return;
 
-	sbi = elr->lr_sbi;
-
 	list_del(&elr->lr_request);
-	sbi->s_li_request = NULL;
+	EXT4_SB(elr->lr_super)->s_li_request = NULL;
 	kfree(elr);
 }
 
@@ -3461,7 +3481,6 @@ static int ext4_li_info_new(void)
 static struct ext4_li_request *ext4_li_request_new(struct super_block *sb,
 					    ext4_group_t start)
 {
-	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_li_request *elr;
 
 	elr = kzalloc(sizeof(*elr), GFP_KERNEL);
@@ -3469,8 +3488,13 @@ static struct ext4_li_request *ext4_li_request_new(struct super_block *sb,
 		return NULL;
 
 	elr->lr_super = sb;
-	elr->lr_sbi = sbi;
-	elr->lr_next_group = start;
+	elr->lr_first_not_zeroed = start;
+	if (test_opt(sb, PREFETCH_BLOCK_BITMAPS))
+		elr->lr_mode = EXT4_LI_MODE_PREFETCH_BBITMAP;
+	else {
+		elr->lr_mode = EXT4_LI_MODE_ITABLE;
+		elr->lr_next_group = start;
+	}
 
 	/*
 	 * Randomize first schedule time of the request to
@@ -3500,8 +3524,9 @@ int ext4_register_li_request(struct super_block *sb,
 		goto out;
 	}
 
-	if (first_not_zeroed == ngroups || sb_rdonly(sb) ||
-	    !test_opt(sb, INIT_INODE_TABLE))
+	if (!test_opt(sb, PREFETCH_BLOCK_BITMAPS) &&
+	    (first_not_zeroed == ngroups || sb_rdonly(sb) ||
+	     !test_opt(sb, INIT_INODE_TABLE)))
 		goto out;
 
 	elr = ext4_li_request_new(sb, first_not_zeroed);