vfs: re-implement writeback_inodes_sb(_nr)_if_idle() and rename them

fs/fs-writeback.c void writeback_inodes_sb_nr(struct super_block *sb, · line 1317

Quoted 7 lines · Miao Xie

fs/fs-writeback.c +0 -1

@@ -1314,7 +1314,6 @@ void writeback_inodes_sb_nr(struct super_block *sb,
 	bdi_queue_work(sb->s_bdi, &work);
 	wait_for_completion(&done);
 }
-EXPORT_SYMBOL(writeback_inodes_sb_nr);

Why do you remove the export? writeback_inodes_sb is exported as well.

Originally the _nr variant has been introduced for btrfs (3259f8bed2f0f57c2fdcdac1b510c3fa319ef97e) and there are no other users now, so from that point it would make sense. From the other side, the change is not strictly necessary for this patch and keeps the writeback API a bit more flexible. I vote for keeping it.

Otherwise (for the btrfs part),

Tested-by David Sterba <[email protected]>

writeback_inodes_sb(_nr)_if_idle() is re-implemented by replacing down_read() with down_read_trylock() because
- If ->s_umount is write locked, then the sb is not idle. That is writeback_inodes_sb(_nr)_if_idle() needn't wait for the lock.
- writeback_inodes_sb(_nr)_if_idle() grabs s_umount lock when it want to start writeback, it may bring us deadlock problem when doing umount. In order to fix the problem, ext4 and btrfs implemented their own writeback functions instead of writeback_inodes_sb(_nr)_if_idle(), but it introduced the redundant code, it is better to implement a new writeback_inodes_sb(_nr)_if_idle().

The name of these two functions is cumbersome, so rename them to try_to_writeback_inodes_sb(_nr).

This idea came from Christoph Hellwig.
Some code is from the patch of Kamal Mostafa.

Signed-off-by Miao Xie <[email protected]>

---
Changelog v1 -> v2:
- do not remove EXPORT_SYMBOL of writeback_inodes_sb_br()

---
 fs/btrfs/extent-tree.c    | 20 +++-----------------
 fs/ext4/inode.c           |  8 ++------
 fs/fs-writeback.c         | 44 ++++++++++++++++++++------------------------
 include/linux/writeback.h |  6 +++---
 4 files changed, 28 insertions(+), 50 deletions(-)

fs/btrfs/extent-tree.c +3 -17

index 521e9d4..f31abb1 100644
@@ -3689,20 +3689,6 @@ static int can_overcommit(struct btrfs_root *root,
 	return 0;
 }
 
-static int writeback_inodes_sb_nr_if_idle_safe(struct super_block *sb,
-					       unsigned long nr_pages,
-					       enum wb_reason reason)
-{
-	if (!writeback_in_progress(sb->s_bdi) &&
-	    down_read_trylock(&sb->s_umount)) {
-		writeback_inodes_sb_nr(sb, nr_pages, reason);
-		up_read(&sb->s_umount);
-		return 1;
-	}
-
-	return 0;
-}
-
 /*
  * shrink metadata reservation for delalloc
  */

@@ -3735,9 +3721,9 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 	while (delalloc_bytes && loops < 3) {
 		max_reclaim = min(delalloc_bytes, to_reclaim);
 		nr_pages = max_reclaim >> PAGE_CACHE_SHIFT;
-		writeback_inodes_sb_nr_if_idle_safe(root->fs_info->sb,
-						    nr_pages,
-						    WB_REASON_FS_FREE_SPACE);
+		try_to_writeback_inodes_sb_nr(root->fs_info->sb,
+					      nr_pages,
+					      WB_REASON_FS_FREE_SPACE);
 
 		/*
 		 * We need to wait for the async pages to actually start before

fs/ext4/inode.c +2 -6

index cbfe13b..5f6eef7 100644
@@ -2512,12 +2512,8 @@ static int ext4_nonda_switch(struct super_block *sb)
 	/*
 	 * Start pushing delalloc when 1/2 of free blocks are dirty.
 	 */
-	if (dirty_blocks && (free_blocks < 2 * dirty_blocks) &&
-	    !writeback_in_progress(sb->s_bdi) &&
-	    down_read_trylock(&sb->s_umount)) {
-		writeback_inodes_sb(sb, WB_REASON_FS_FREE_SPACE);
-		up_read(&sb->s_umount);
-	}
+	if (dirty_blocks && (free_blocks < 2 * dirty_blocks))
+		try_to_writeback_inodes_sb(sb, WB_REASON_FS_FREE_SPACE);
 
 	if (2 * free_blocks < 3 * dirty_blocks ||
 		free_blocks < (dirty_blocks + EXT4_FREECLUSTERS_WATERMARK)) {

fs/fs-writeback.c +20 -24

index 310972b..ad3cc46 100644
@@ -1332,47 +1332,43 @@ void writeback_inodes_sb(struct super_block *sb, enum wb_reason reason)
 EXPORT_SYMBOL(writeback_inodes_sb);
 
 /**
- * writeback_inodes_sb_if_idle	-	start writeback if none underway
+ * try_to_writeback_inodes_sb_nr - try to start writeback if none underway
  * @sb: the superblock
- * @reason: reason why some writeback work was initiated
+ * @nr: the number of pages to write
+ * @reason: the reason of writeback
  *
- * Invoke writeback_inodes_sb if no writeback is currently underway.
+ * Invoke writeback_inodes_sb_nr if no writeback is currently underway.
  * Returns 1 if writeback was started, 0 if not.
  */
-int writeback_inodes_sb_if_idle(struct super_block *sb, enum wb_reason reason)
+int try_to_writeback_inodes_sb_nr(struct super_block *sb,
+				  unsigned long nr,
+				  enum wb_reason reason)
 {
-	if (!writeback_in_progress(sb->s_bdi)) {
-		down_read(&sb->s_umount);
-		writeback_inodes_sb(sb, reason);
-		up_read(&sb->s_umount);
+	if (writeback_in_progress(sb->s_bdi))
 		return 1;
-	} else
+
+	if (!down_read_trylock(&sb->s_umount))
 		return 0;
+
+	writeback_inodes_sb_nr(sb, nr, reason);
+	up_read(&sb->s_umount);
+	return 1;
 }
-EXPORT_SYMBOL(writeback_inodes_sb_if_idle);
+EXPORT_SYMBOL(try_to_writeback_inodes_sb_nr);
 
 /**
- * writeback_inodes_sb_nr_if_idle	-	start writeback if none underway
+ * try_to_writeback_inodes_sb - try to start writeback if none underway
  * @sb: the superblock
- * @nr: the number of pages to write
  * @reason: reason why some writeback work was initiated
  *
- * Invoke writeback_inodes_sb if no writeback is currently underway.
+ * Implement by try_to_writeback_inodes_sb_nr()
  * Returns 1 if writeback was started, 0 if not.
  */
-int writeback_inodes_sb_nr_if_idle(struct super_block *sb,
-				   unsigned long nr,
-				   enum wb_reason reason)
+int try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason)
 {
-	if (!writeback_in_progress(sb->s_bdi)) {
-		down_read(&sb->s_umount);
-		writeback_inodes_sb_nr(sb, nr, reason);
-		up_read(&sb->s_umount);
-		return 1;
-	} else
-		return 0;
+	return try_to_writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason);
 }
-EXPORT_SYMBOL(writeback_inodes_sb_nr_if_idle);
+EXPORT_SYMBOL(try_to_writeback_inodes_sb);
 
 /**
  * sync_inodes_sb	-	sync sb inode pages

include/linux/writeback.h +3 -3

index b82a83a..9a9367c 100644
@@ -87,9 +87,9 @@ int inode_wait(void *);
 void writeback_inodes_sb(struct super_block *, enum wb_reason reason);
 void writeback_inodes_sb_nr(struct super_block *, unsigned long nr,
 							enum wb_reason reason);
-int writeback_inodes_sb_if_idle(struct super_block *, enum wb_reason reason);
-int writeback_inodes_sb_nr_if_idle(struct super_block *, unsigned long nr,
-							enum wb_reason reason);
+int try_to_writeback_inodes_sb(struct super_block *, enum wb_reason reason);
+int try_to_writeback_inodes_sb_nr(struct super_block *, unsigned long nr,
+				  enum wb_reason reason);
 void sync_inodes_sb(struct super_block *);
 long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages,
 				enum wb_reason reason);

Signature · 7 lines

1.7.11.7

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Quoted 13 lines · Miao Xie

writeback_inodes_sb(_nr)_if_idle() is re-implemented by replacing down_read() with down_read_trylock() because
- If ->s_umount is write locked, then the sb is not idle. That is writeback_inodes_sb(_nr)_if_idle() needn't wait for the lock.
- writeback_inodes_sb(_nr)_if_idle() grabs s_umount lock when it want to start writeback, it may bring us deadlock problem when doing umount. In order to fix the problem, ext4 and btrfs implemented their own writeback functions instead of writeback_inodes_sb(_nr)_if_idle(), but it introduced the redundant code, it is better to implement a new writeback_inodes_sb(_nr)_if_idle().

The name of these two functions is cumbersome, so rename them to try_to_writeback_inodes_sb(_nr).

Quoted 2 lines

This idea came from Christoph Hellwig.
Some code is from the patch of Kamal Mostafa.

The patch looks good. Thanks! You can add:

Reviewed-by Jan Kara <[email protected]>

BTW, since changes to filesystems are minimal, maybe Fengguang can take
the patch through his writeback tree? Fengguang?