summaryrefslogtreecommitdiffstats
path: root/fs/ext4
diff options
context:
space:
mode:
authorChristoph Hellwig <hch@infradead.org>2013-09-04 15:04:39 +0200
committerAl Viro <viro@zeniv.linux.org.uk>2013-09-04 09:23:46 -0400
commit7b7a8665edd8db733980389b098530f9e4f630b2 (patch)
tree968d570a9f0c4d861226aefed2f5f97a131c8d53 /fs/ext4
parent4b6ccca701ef5977d0ffbc2c932430dea88b38b6 (diff)
downloadlinux-7b7a8665edd8db733980389b098530f9e4f630b2.tar.bz2
direct-io: Implement generic deferred AIO completions
Add support to the core direct-io code to defer AIO completions to user context using a workqueue. This replaces opencoded and less efficient code in XFS and ext4 (we save a memory allocation for each direct IO) and will be needed to properly support O_(D)SYNC for AIO. The communication between the filesystem and the direct I/O code requires a new buffer head flag, which is a bit ugly but not avoidable until the direct I/O code stops abusing the buffer_head structure for communicating with the filesystems. Currently this creates a per-superblock unbound workqueue for these completions, which is taken from an earlier patch by Jan Kara. I'm not really convinced about this use and would prefer a "normal" global workqueue with a high concurrency limit, but this needs further discussion. JK: Fixed ext4 part, dynamic allocation of the workqueue. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Diffstat (limited to 'fs/ext4')
-rw-r--r--fs/ext4/ext4.h11
-rw-r--r--fs/ext4/inode.c28
-rw-r--r--fs/ext4/page-io.c30
-rw-r--r--fs/ext4/super.c16
4 files changed, 14 insertions, 71 deletions
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 0ab26fbf3380..b247fbbed99c 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -180,7 +180,6 @@ struct ext4_map_blocks {
* Flags for ext4_io_end->flags
*/
#define EXT4_IO_END_UNWRITTEN 0x0001
-#define EXT4_IO_END_DIRECT 0x0002
/*
* For converting uninitialized extents on a work queue. 'handle' is used for
@@ -196,8 +195,6 @@ 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 kiocb *iocb; /* iocb struct for AIO */
- int result; /* error value for AIO */
atomic_t count; /* reference counter */
} ext4_io_end_t;
@@ -900,11 +897,9 @@ struct ext4_inode_info {
* Completed IOs that need unwritten extents handling and don't have
* transaction reserved
*/
- struct list_head i_unrsv_conversion_list;
atomic_t i_ioend_count; /* Number of outstanding io_end structs */
atomic_t i_unwritten; /* Nr. of inflight conversions pending */
struct work_struct i_rsv_conversion_work;
- struct work_struct i_unrsv_conversion_work;
spinlock_t i_block_reservation_lock;
@@ -1276,8 +1271,6 @@ struct ext4_sb_info {
struct flex_groups *s_flex_groups;
ext4_group_t s_flex_groups_allocated;
- /* workqueue for unreserved extent convertions (dio) */
- struct workqueue_struct *unrsv_conversion_wq;
/* workqueue for reserved extent conversions (buffered io) */
struct workqueue_struct *rsv_conversion_wq;
@@ -1340,9 +1333,6 @@ static inline void ext4_set_io_unwritten_flag(struct inode *inode,
struct ext4_io_end *io_end)
{
if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) {
- /* Writeback has to have coversion transaction reserved */
- WARN_ON(EXT4_SB(inode->i_sb)->s_journal && !io_end->handle &&
- !(io_end->flag & EXT4_IO_END_DIRECT));
io_end->flag |= EXT4_IO_END_UNWRITTEN;
atomic_inc(&EXT4_I(inode)->i_unwritten);
}
@@ -2716,7 +2706,6 @@ extern void ext4_put_io_end_defer(ext4_io_end_t *io_end);
extern void ext4_io_submit_init(struct ext4_io_submit *io,
struct writeback_control *wbc);
extern void ext4_end_io_rsv_work(struct work_struct *work);
-extern void ext4_end_io_unrsv_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/inode.c b/fs/ext4/inode.c
index c2ca04e67a4f..123bd81692d1 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -727,8 +727,12 @@ static int _ext4_get_block(struct inode *inode, sector_t iblock,
ret = ext4_map_blocks(handle, inode, &map, flags);
if (ret > 0) {
+ ext4_io_end_t *io_end = ext4_inode_aio(inode);
+
map_bh(bh, inode->i_sb, map.m_pblk);
bh->b_state = (bh->b_state & ~EXT4_MAP_FLAGS) | map.m_flags;
+ if (io_end && io_end->flag & EXT4_IO_END_UNWRITTEN)
+ set_buffer_defer_completion(bh);
bh->b_size = inode->i_sb->s_blocksize * map.m_len;
ret = 0;
}
@@ -2991,19 +2995,13 @@ static int ext4_get_block_write_nolock(struct inode *inode, sector_t iblock,
}
static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
- ssize_t size, void *private, int ret,
- bool is_async)
+ ssize_t size, void *private)
{
- struct inode *inode = file_inode(iocb->ki_filp);
ext4_io_end_t *io_end = iocb->private;
/* if not async direct IO just return */
- if (!io_end) {
- inode_dio_done(inode);
- if (is_async)
- aio_complete(iocb, ret, 0);
+ if (!io_end)
return;
- }
ext_debug("ext4_end_io_dio(): io_end 0x%p "
"for inode %lu, iocb 0x%p, offset %llu, size %zd\n",
@@ -3013,11 +3011,7 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
iocb->private = NULL;
io_end->offset = offset;
io_end->size = size;
- if (is_async) {
- io_end->iocb = iocb;
- io_end->result = ret;
- }
- ext4_put_io_end_defer(io_end);
+ ext4_put_io_end(io_end);
}
/*
@@ -3102,7 +3096,6 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
ret = -ENOMEM;
goto retake_lock;
}
- io_end->flag |= EXT4_IO_END_DIRECT;
/*
* Grab reference for DIO. Will be dropped in ext4_end_io_dio()
*/
@@ -3147,13 +3140,6 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
if (ret <= 0 && ret != -EIOCBQUEUED && iocb->private) {
WARN_ON(iocb->private != io_end);
WARN_ON(io_end->flag & EXT4_IO_END_UNWRITTEN);
- WARN_ON(io_end->iocb);
- /*
- * Generic code already did inode_dio_done() so we
- * have to clear EXT4_IO_END_DIRECT to not do it for
- * the second time.
- */
- io_end->flag = 0;
ext4_put_io_end(io_end);
iocb->private = NULL;
}
diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index 6625d210fb45..d7d0c7b46ed4 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -123,10 +123,6 @@ static void ext4_release_io_end(ext4_io_end_t *io_end)
ext4_finish_bio(bio);
bio_put(bio);
}
- if (io_end->flag & EXT4_IO_END_DIRECT)
- inode_dio_done(io_end->inode);
- if (io_end->iocb)
- aio_complete(io_end->iocb, io_end->result, 0);
kmem_cache_free(io_end_cachep, io_end);
}
@@ -204,19 +200,14 @@ static void ext4_add_complete_io(ext4_io_end_t *io_end)
struct workqueue_struct *wq;
unsigned long flags;
- BUG_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
+ /* Only reserved conversions from writeback should enter here */
+ WARN_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
+ WARN_ON(!io_end->handle);
spin_lock_irqsave(&ei->i_completed_io_lock, flags);
- if (io_end->handle) {
- wq = EXT4_SB(io_end->inode->i_sb)->rsv_conversion_wq;
- if (list_empty(&ei->i_rsv_conversion_list))
- queue_work(wq, &ei->i_rsv_conversion_work);
- list_add_tail(&io_end->list, &ei->i_rsv_conversion_list);
- } else {
- wq = EXT4_SB(io_end->inode->i_sb)->unrsv_conversion_wq;
- if (list_empty(&ei->i_unrsv_conversion_list))
- queue_work(wq, &ei->i_unrsv_conversion_work);
- list_add_tail(&io_end->list, &ei->i_unrsv_conversion_list);
- }
+ wq = EXT4_SB(io_end->inode->i_sb)->rsv_conversion_wq;
+ if (list_empty(&ei->i_rsv_conversion_list))
+ queue_work(wq, &ei->i_rsv_conversion_work);
+ list_add_tail(&io_end->list, &ei->i_rsv_conversion_list);
spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
}
@@ -256,13 +247,6 @@ void ext4_end_io_rsv_work(struct work_struct *work)
ext4_do_flush_completed_IO(&ei->vfs_inode, &ei->i_rsv_conversion_list);
}
-void ext4_end_io_unrsv_work(struct work_struct *work)
-{
- struct ext4_inode_info *ei = container_of(work, struct ext4_inode_info,
- i_unrsv_conversion_work);
- ext4_do_flush_completed_IO(&ei->vfs_inode, &ei->i_unrsv_conversion_list);
-}
-
ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
{
ext4_io_end_t *io = kmem_cache_zalloc(io_end_cachep, flags);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index b59373b625e9..5db4f0df8174 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -762,9 +762,7 @@ static void ext4_put_super(struct super_block *sb)
ext4_unregister_li_request(sb);
dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);
- flush_workqueue(sbi->unrsv_conversion_wq);
flush_workqueue(sbi->rsv_conversion_wq);
- destroy_workqueue(sbi->unrsv_conversion_wq);
destroy_workqueue(sbi->rsv_conversion_wq);
if (sbi->s_journal) {
@@ -875,14 +873,12 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
#endif
ei->jinode = NULL;
INIT_LIST_HEAD(&ei->i_rsv_conversion_list);
- INIT_LIST_HEAD(&ei->i_unrsv_conversion_list);
spin_lock_init(&ei->i_completed_io_lock);
ei->i_sync_tid = 0;
ei->i_datasync_tid = 0;
atomic_set(&ei->i_ioend_count, 0);
atomic_set(&ei->i_unwritten, 0);
INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work);
- INIT_WORK(&ei->i_unrsv_conversion_work, ext4_end_io_unrsv_work);
return &ei->vfs_inode;
}
@@ -3954,14 +3950,6 @@ no_journal:
goto failed_mount4;
}
- EXT4_SB(sb)->unrsv_conversion_wq =
- alloc_workqueue("ext4-unrsv-conversion", WQ_MEM_RECLAIM | WQ_UNBOUND, 1);
- if (!EXT4_SB(sb)->unrsv_conversion_wq) {
- printk(KERN_ERR "EXT4-fs: failed to create workqueue\n");
- ret = -ENOMEM;
- goto failed_mount4;
- }
-
/*
* The jbd2_journal_load will have done any necessary log recovery,
* so we can safely mount the rest of the filesystem now.
@@ -4115,8 +4103,6 @@ failed_mount4:
ext4_msg(sb, KERN_ERR, "mount failed");
if (EXT4_SB(sb)->rsv_conversion_wq)
destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
- if (EXT4_SB(sb)->unrsv_conversion_wq)
- destroy_workqueue(EXT4_SB(sb)->unrsv_conversion_wq);
failed_mount_wq:
if (sbi->s_journal) {
jbd2_journal_destroy(sbi->s_journal);
@@ -4564,7 +4550,6 @@ static int ext4_sync_fs(struct super_block *sb, int wait)
trace_ext4_sync_fs(sb, wait);
flush_workqueue(sbi->rsv_conversion_wq);
- flush_workqueue(sbi->unrsv_conversion_wq);
/*
* Writeback quota in non-journalled quota case - journalled quota has
* no dirty dquots
@@ -4600,7 +4585,6 @@ static int ext4_sync_fs_nojournal(struct super_block *sb, int wait)
trace_ext4_sync_fs(sb, wait);
flush_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
- flush_workqueue(EXT4_SB(sb)->unrsv_conversion_wq);
dquot_writeback_dquots(sb, -1);
if (wait && test_opt(sb, BARRIER))
ret = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);