ext4: move work from io_end to inode

It does not make much sense to have struct work in ext4_io_end_t
because we always use it for only one ext4_io_end_t per inode (the
first one in the i_completed_io list). So just move the structure to
inode itself.  This also allows for a small simplification in
processing io_end structures.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>

3 files changed, 13 insertions, 27 deletions

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 0ccda0c9682e..d93393eb5f2d 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -194,8 +194,7 @@ struct mpage_da_data {
  */
 #define EXT4_IO_END_UNWRITTEN   0x0001
 #define EXT4_IO_END_ERROR       0x0002
-#define EXT4_IO_END_QUEUED      0x0004
-#define EXT4_IO_END_DIRECT      0x0008
+#define EXT4_IO_END_DIRECT      0x0004
 
 struct ext4_io_page {
         struct page     *p_page;
@@ -217,7 +216,6 @@ typedef struct ext4_io_end {
         unsigned int            flag;           /* unwritten or not */
         loff_t                  offset;         /* offset in the file */
         ssize_t                 size;           /* size of the extent */
-        struct work_struct      work;           /* data work queue */
         struct kiocb            *iocb;          /* iocb struct for AIO */
         int                     result;         /* error value for AIO */
         int                     num_io_pages;   /* for writepages() */
@@ -929,6 +927,7 @@ struct ext4_inode_info {
         spinlock_t i_completed_io_lock;
         atomic_t i_ioend_count; /* Number of outstanding io_end structs */
         atomic_t i_unwritten; /* Nr. of inflight conversions pending */
+        struct work_struct i_unwritten_work;    /* deferred extent conversion */
 
         spinlock_t i_block_reservation_lock;
 
@@ -2538,6 +2537,7 @@ extern void ext4_exit_pageio(void);
 extern void ext4_ioend_wait(struct inode *);
 extern void ext4_free_io_end(ext4_io_end_t *io);
 extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags);
+extern void ext4_end_io_work(struct work_struct *work);
 extern void ext4_io_submit(struct ext4_io_submit *io);
 extern int ext4_bio_write_page(struct ext4_io_submit *io,
                                struct page *page,
diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index 05795f10e55a..a0290176ee75 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -151,16 +151,13 @@ void ext4_add_complete_io(ext4_io_end_t *io_end)
         wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq;
 
         spin_lock_irqsave(&ei->i_completed_io_lock, flags);
-        if (list_empty(&ei->i_completed_io_list)) {
-                io_end->flag |= EXT4_IO_END_QUEUED;
-                queue_work(wq, &io_end->work);
-        }
+        if (list_empty(&ei->i_completed_io_list))
+                queue_work(wq, &ei->i_unwritten_work);
         list_add_tail(&io_end->list, &ei->i_completed_io_list);
         spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
 }
 
-static int ext4_do_flush_completed_IO(struct inode *inode,
-                                      ext4_io_end_t *work_io)
+static int ext4_do_flush_completed_IO(struct inode *inode)
 {
         ext4_io_end_t *io;
         struct list_head unwritten, complete, to_free;
@@ -191,19 +188,7 @@ static int ext4_do_flush_completed_IO(struct inode *inode,
         while (!list_empty(&complete)) {
                 io = list_entry(complete.next, ext4_io_end_t, list);
                 io->flag &= ~EXT4_IO_END_UNWRITTEN;
-                /* end_io context can not be destroyed now because it still
-                 * used by queued worker. Worker thread will destroy it later */
-                if (io->flag & EXT4_IO_END_QUEUED)
-                        list_del_init(&io->list);
-                else
-                        list_move(&io->list, &to_free);
-        }
-        /* If we are called from worker context, it is time to clear queued
-         * flag, and destroy it's end_io if it was converted already */
-        if (work_io) {
-                work_io->flag &= ~EXT4_IO_END_QUEUED;
-                if (!(work_io->flag & EXT4_IO_END_UNWRITTEN))
-                        list_add_tail(&work_io->list, &to_free);
+                list_move(&io->list, &to_free);
         }
         spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
 
@@ -218,10 +203,11 @@ static int ext4_do_flush_completed_IO(struct inode *inode,
 /*
  * work on completed aio dio IO, to convert unwritten extents to extents
  */
-static void ext4_end_io_work(struct work_struct *work)
+void ext4_end_io_work(struct work_struct *work)
 {
-        ext4_io_end_t *io = container_of(work, ext4_io_end_t, work);
-        ext4_do_flush_completed_IO(io->inode, io);
+        struct ext4_inode_info *ei = container_of(work, struct ext4_inode_info,
+                                                  i_unwritten_work);
+        ext4_do_flush_completed_IO(&ei->vfs_inode);
 }
 
 int ext4_flush_unwritten_io(struct inode *inode)
@@ -229,7 +215,7 @@ int ext4_flush_unwritten_io(struct inode *inode)
         int ret;
         WARN_ON_ONCE(!mutex_is_locked(&inode->i_mutex) &&
                      !(inode->i_state & I_FREEING));
-        ret = ext4_do_flush_completed_IO(inode, NULL);
+        ret = ext4_do_flush_completed_IO(inode);
         ext4_unwritten_wait(inode);
         return ret;
 }
@@ -240,7 +226,6 @@ ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
         if (io) {
                 atomic_inc(&EXT4_I(inode)->i_ioend_count);
                 io->inode = inode;
-                INIT_WORK(&io->work, ext4_end_io_work);
                 INIT_LIST_HEAD(&io->list);
         }
         return io;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index d5d336bd2180..dc0fb7b942cf 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -960,6 +960,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
         ei->i_datasync_tid = 0;
         atomic_set(&ei->i_ioend_count, 0);
         atomic_set(&ei->i_unwritten, 0);
+        INIT_WORK(&ei->i_unwritten_work, ext4_end_io_work);
 
         return &ei->vfs_inode;
 }