[PATCH] switchable and modular io schedulers

This patch modularizes the io schedulers completely, allowing them to be
modular.  Additionally it enables online switching of io schedulers.  See
also http://lwn.net/Articles/102593/ .


There's a scheduler file in the sysfs directory for the block device
queue:

axboe@router:/sys/block/hda/queue> ls
iosched            max_sectors_kb  read_ahead_kb
max_hw_sectors_kb  nr_requests     scheduler

If you list the contents of the file, it will show available schedulers
and the active one:

axboe@router:/sys/block/hda/queue> cat scheduler
[cfq]

Lets load a few more.

router:/sys/block/hda/queue # modprobe deadline-iosched
router:/sys/block/hda/queue # modprobe as-iosched
router:/sys/block/hda/queue # cat scheduler
[cfq] deadline anticipatory

Changing is done with

router:/sys/block/hda/queue # echo deadline > scheduler
router:/sys/block/hda/queue # cat scheduler
cfq [deadline] anticipatory

deadline is now the new active io scheduler for hda.

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

2 files changed, 40 insertions, 25 deletions

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4efe45d1af7e..5e4a6ab84ecb 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -19,8 +19,8 @@
 
 struct request_queue;
 typedef struct request_queue request_queue_t;
-struct elevator_s;
-typedef struct elevator_s elevator_t;
+struct elevator_queue;
+typedef struct elevator_queue elevator_t;
 struct request_pm_state;
 
 #define BLKDEV_MIN_RQ	4
@@ -80,6 +80,7 @@ struct request_list {
 	int count[2];
 	mempool_t *rq_pool;
 	wait_queue_head_t wait[2];
+	wait_queue_head_t drain;
 };
 
 #define BLK_MAX_CDB	16
@@ -279,7 +280,7 @@ struct request_queue
 	 */
 	struct list_head	queue_head;
 	struct request		*last_merge;
-	elevator_t		elevator;
+	elevator_t		*elevator;
 
 	/*
 	 * the queue request freelist, one for reads and one for writes
@@ -381,6 +382,7 @@ struct request_queue
 #define QUEUE_FLAG_REENTER	6	/* Re-entrancy avoidance */
 #define QUEUE_FLAG_PLUGGED	7	/* queue is plugged */
 #define QUEUE_FLAG_ORDERED	8	/* supports ordered writes */
+#define QUEUE_FLAG_DRAIN	9	/* draining queue for sched switch */
 
 #define blk_queue_plugged(q)	test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
 #define blk_queue_tagged(q)	test_bit(QUEUE_FLAG_QUEUED, &(q)->queue_flags)
@@ -617,6 +619,8 @@ extern void blk_dump_rq_flags(struct request *, char *);
 extern void generic_unplug_device(request_queue_t *);
 extern void __generic_unplug_device(request_queue_t *);
 extern long nr_blockdev_pages(void);
+extern void blk_wait_queue_drained(request_queue_t *);
+extern void blk_finish_queue_drain(request_queue_t *);
 
 int blk_get_queue(request_queue_t *);
 request_queue_t *blk_alloc_queue(int);
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 27e8183f4776..95cdfb5bb790 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -22,9 +22,9 @@ typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, int);
 typedef void (elevator_put_req_fn) (request_queue_t *, struct request *);
 
 typedef int (elevator_init_fn) (request_queue_t *, elevator_t *);
-typedef void (elevator_exit_fn) (request_queue_t *, elevator_t *);
+typedef void (elevator_exit_fn) (elevator_t *);
 
-struct elevator_s
+struct elevator_ops
 {
 	elevator_merge_fn *elevator_merge_fn;
 	elevator_merged_fn *elevator_merged_fn;
@@ -48,12 +48,32 @@ struct elevator_s
 
 	elevator_init_fn *elevator_init_fn;
 	elevator_exit_fn *elevator_exit_fn;
+};
 
-	void *elevator_data;
+#define ELV_NAME_MAX	(16)
 
-	struct kobject kobj;
+/*
+ * identifies an elevator type, such as AS or deadline
+ */
+struct elevator_type
+{
+	struct list_head list;
+	struct elevator_ops ops;
+	struct elevator_type *elevator_type;
 	struct kobj_type *elevator_ktype;
-	const char *elevator_name;
+	char elevator_name[ELV_NAME_MAX];
+	struct module *elevator_owner;
+};
+
+/*
+ * each queue has an elevator_queue assoicated with it
+ */
+struct elevator_queue
+{
+	struct elevator_ops *ops;
+	void *elevator_data;
+	struct kobject kobj;
+	struct elevator_type *elevator_type;
 };
 
 /*
@@ -79,28 +99,19 @@ extern int elv_set_request(request_queue_t *, struct request *, int);
 extern void elv_put_request(request_queue_t *, struct request *);
 
 /*
- * noop I/O scheduler. always merges, always inserts new request at tail
- */
-extern elevator_t elevator_noop;
-
-/*
- * deadline i/o scheduler. uses request time outs to prevent indefinite
- * starvation
- */
-extern elevator_t iosched_deadline;
-
-/*
- * anticipatory I/O scheduler
+ * io scheduler registration
  */
-extern elevator_t iosched_as;
+extern int elv_register(struct elevator_type *);
+extern void elv_unregister(struct elevator_type *);
 
 /*
- * completely fair queueing I/O scheduler
+ * io scheduler sysfs switching
  */
-extern elevator_t iosched_cfq;
+extern ssize_t elv_iosched_show(request_queue_t *, char *);
+extern ssize_t elv_iosched_store(request_queue_t *, const char *, size_t);
 
-extern int elevator_init(request_queue_t *, elevator_t *);
-extern void elevator_exit(request_queue_t *);
+extern int elevator_init(request_queue_t *, char *);
+extern void elevator_exit(elevator_t *);
 extern int elv_rq_merge_ok(struct request *, struct bio *);
 extern int elv_try_merge(struct request *, struct bio *);
 extern int elv_try_last_merge(request_queue_t *, struct bio *);