ALSA: hda - Fix no sound after Windows boot with ALC269
[linux-2.6.git] / block / elevator.c
index 536be74..0cdb4e7 100644 (file)
 #include <linux/delay.h>
 #include <linux/blktrace_api.h>
 #include <linux/hash.h>
+#include <linux/uaccess.h>
 
-#include <asm/uaccess.h>
+#include <trace/events/block.h>
+
+#include "blk.h"
 
 static DEFINE_SPINLOCK(elv_list_lock);
 static LIST_HEAD(elv_list);
@@ -45,10 +48,10 @@ static LIST_HEAD(elv_list);
  */
 static const int elv_hash_shift = 6;
 #define ELV_HASH_BLOCK(sec)    ((sec) >> 3)
-#define ELV_HASH_FN(sec)       (hash_long(ELV_HASH_BLOCK((sec)), elv_hash_shift))
+#define ELV_HASH_FN(sec)       \
+               (hash_long(ELV_HASH_BLOCK((sec)), elv_hash_shift))
 #define ELV_HASH_ENTRIES       (1 << elv_hash_shift)
-#define rq_hash_key(rq)                ((rq)->sector + (rq)->nr_sectors)
-#define ELV_ON_HASH(rq)                (!hlist_unhashed(&(rq)->hash))
+#define rq_hash_key(rq)                (blk_rq_pos(rq) + blk_rq_sectors(rq))
 
 /*
  * Query io scheduler to see if the current process issuing bio may be
@@ -56,8 +59,8 @@ static const int elv_hash_shift = 6;
  */
 static int elv_iosched_allow_merge(struct request *rq, struct bio *bio)
 {
-       request_queue_t *q = rq->q;
-       elevator_t *e = q->elevator;
+       struct request_queue *q = rq->q;
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_allow_merge_fn)
                return e->ops->elevator_allow_merge_fn(q, rq, bio);
@@ -68,12 +71,24 @@ static int elv_iosched_allow_merge(struct request *rq, struct bio *bio)
 /*
  * can we safely merge with this request?
  */
-inline int elv_rq_merge_ok(struct request *rq, struct bio *bio)
+int elv_rq_merge_ok(struct request *rq, struct bio *bio)
 {
        if (!rq_mergeable(rq))
                return 0;
 
        /*
+        * Don't merge file system requests and discard requests
+        */
+       if ((bio->bi_rw & REQ_DISCARD) != (rq->bio->bi_rw & REQ_DISCARD))
+               return 0;
+
+       /*
+        * Don't merge discard requests and secure discard requests
+        */
+       if ((bio->bi_rw & REQ_SECURE) != (rq->bio->bi_rw & REQ_SECURE))
+               return 0;
+
+       /*
         * different data direction or already started, don't merge
         */
        if (bio_data_dir(bio) != rq_data_dir(rq))
@@ -85,6 +100,12 @@ inline int elv_rq_merge_ok(struct request *rq, struct bio *bio)
        if (rq->rq_disk != bio->bi_bdev->bd_disk || rq->special)
                return 0;
 
+       /*
+        * only merge integrity protected bio into ditto rq
+        */
+       if (bio_integrity(bio) != blk_integrity_rq(rq))
+               return 0;
+
        if (!elv_iosched_allow_merge(rq, bio))
                return 0;
 
@@ -92,7 +113,7 @@ inline int elv_rq_merge_ok(struct request *rq, struct bio *bio)
 }
 EXPORT_SYMBOL(elv_rq_merge_ok);
 
-static inline int elv_try_merge(struct request *__rq, struct bio *bio)
+int elv_try_merge(struct request *__rq, struct bio *bio)
 {
        int ret = ELEVATOR_NO_MERGE;
 
@@ -100,9 +121,9 @@ static inline int elv_try_merge(struct request *__rq, struct bio *bio)
         * we can merge and sequence is ok, check if it's possible
         */
        if (elv_rq_merge_ok(__rq, bio)) {
-               if (__rq->sector + __rq->nr_sectors == bio->bi_sector)
+               if (blk_rq_pos(__rq) + blk_rq_sectors(__rq) == bio->bi_sector)
                        ret = ELEVATOR_BACK_MERGE;
-               else if (__rq->sector - bio_sectors(bio) == bio->bi_sector)
+               else if (blk_rq_pos(__rq) - bio_sectors(bio) == bio->bi_sector)
                        ret = ELEVATOR_FRONT_MERGE;
        }
 
@@ -112,12 +133,8 @@ static inline int elv_try_merge(struct request *__rq, struct bio *bio)
 static struct elevator_type *elevator_find(const char *name)
 {
        struct elevator_type *e;
-       struct list_head *entry;
-
-       list_for_each(entry, &elv_list) {
-
-               e = list_entry(entry, struct elevator_type, list);
 
+       list_for_each_entry(e, &elv_list, list) {
                if (!strcmp(e->elevator_name, name))
                        return e;
        }
@@ -134,23 +151,36 @@ static struct elevator_type *elevator_get(const char *name)
 {
        struct elevator_type *e;
 
-       spin_lock_irq(&elv_list_lock);
+       spin_lock(&elv_list_lock);
 
        e = elevator_find(name);
+       if (!e) {
+               char elv[ELV_NAME_MAX + strlen("-iosched")];
+
+               spin_unlock(&elv_list_lock);
+
+               snprintf(elv, sizeof(elv), "%s-iosched", name);
+
+               request_module("%s", elv);
+               spin_lock(&elv_list_lock);
+               e = elevator_find(name);
+       }
+
        if (e && !try_module_get(e->elevator_owner))
                e = NULL;
 
-       spin_unlock_irq(&elv_list_lock);
+       spin_unlock(&elv_list_lock);
 
        return e;
 }
 
-static void *elevator_init_queue(request_queue_t *q, struct elevator_queue *eq)
+static void *elevator_init_queue(struct request_queue *q,
+                                struct elevator_queue *eq)
 {
        return eq->ops->elevator_init_fn(q);
 }
 
-static void elevator_attach(request_queue_t *q, struct elevator_queue *eq,
+static void elevator_attach(struct request_queue *q, struct elevator_queue *eq,
                           void *data)
 {
        q->elevator = eq;
@@ -165,10 +195,7 @@ static int __init elevator_setup(char *str)
         * Be backwards-compatible with previous kernels, so users
         * won't get the wrong elevator.
         */
-       if (!strcmp(str, "as"))
-               strcpy(chosen_elevator, "anticipatory");
-       else
-               strncpy(chosen_elevator, str, sizeof(chosen_elevator) - 1);
+       strncpy(chosen_elevator, str, sizeof(chosen_elevator) - 1);
        return 1;
 }
 
@@ -176,21 +203,19 @@ __setup("elevator=", elevator_setup);
 
 static struct kobj_type elv_ktype;
 
-static elevator_t *elevator_alloc(request_queue_t *q, struct elevator_type *e)
+static struct elevator_queue *elevator_alloc(struct request_queue *q,
+                                 struct elevator_type *e)
 {
-       elevator_t *eq;
+       struct elevator_queue *eq;
        int i;
 
-       eq = kmalloc_node(sizeof(elevator_t), GFP_KERNEL, q->node);
+       eq = kmalloc_node(sizeof(*eq), GFP_KERNEL | __GFP_ZERO, q->node);
        if (unlikely(!eq))
                goto err;
 
-       memset(eq, 0, sizeof(*eq));
        eq->ops = &e->ops;
        eq->elevator_type = e;
-       kobject_init(&eq->kobj);
-       snprintf(eq->kobj.name, KOBJ_NAME_LEN, "%s", "iosched");
-       eq->kobj.ktype = &elv_ktype;
+       kobject_init(&eq->kobj, &elv_ktype);
        mutex_init(&eq->sysfs_lock);
 
        eq->hash = kmalloc_node(sizeof(struct hlist_head) * ELV_HASH_ENTRIES,
@@ -210,34 +235,49 @@ err:
 
 static void elevator_release(struct kobject *kobj)
 {
-       elevator_t *e = container_of(kobj, elevator_t, kobj);
+       struct elevator_queue *e;
 
+       e = container_of(kobj, struct elevator_queue, kobj);
        elevator_put(e->elevator_type);
        kfree(e->hash);
        kfree(e);
 }
 
-int elevator_init(request_queue_t *q, char *name)
+int elevator_init(struct request_queue *q, char *name)
 {
        struct elevator_type *e = NULL;
        struct elevator_queue *eq;
-       int ret = 0;
        void *data;
 
+       if (unlikely(q->elevator))
+               return 0;
+
        INIT_LIST_HEAD(&q->queue_head);
        q->last_merge = NULL;
        q->end_sector = 0;
        q->boundary_rq = NULL;
 
-       if (name && !(e = elevator_get(name)))
-               return -EINVAL;
+       if (name) {
+               e = elevator_get(name);
+               if (!e)
+                       return -EINVAL;
+       }
 
-       if (!e && *chosen_elevator && !(e = elevator_get(chosen_elevator)))
-               printk("I/O scheduler %s not found\n", chosen_elevator);
+       if (!e && *chosen_elevator) {
+               e = elevator_get(chosen_elevator);
+               if (!e)
+                       printk(KERN_ERR "I/O scheduler %s not found\n",
+                                                       chosen_elevator);
+       }
 
-       if (!e && !(e = elevator_get(CONFIG_DEFAULT_IOSCHED))) {
-               printk("Default I/O scheduler not found, using no-op\n");
-               e = elevator_get("noop");
+       if (!e) {
+               e = elevator_get(CONFIG_DEFAULT_IOSCHED);
+               if (!e) {
+                       printk(KERN_ERR
+                               "Default I/O scheduler not found. " \
+                               "Using noop.\n");
+                       e = elevator_get("noop");
+               }
        }
 
        eq = elevator_alloc(q, e);
@@ -251,12 +291,11 @@ int elevator_init(request_queue_t *q, char *name)
        }
 
        elevator_attach(q, eq, data);
-       return ret;
+       return 0;
 }
-
 EXPORT_SYMBOL(elevator_init);
 
-void elevator_exit(elevator_t *e)
+void elevator_exit(struct elevator_queue *e)
 {
        mutex_lock(&e->sysfs_lock);
        if (e->ops->elevator_exit_fn)
@@ -266,7 +305,6 @@ void elevator_exit(elevator_t *e)
 
        kobject_put(&e->kobj);
 }
-
 EXPORT_SYMBOL(elevator_exit);
 
 static inline void __elv_rqhash_del(struct request *rq)
@@ -274,29 +312,29 @@ static inline void __elv_rqhash_del(struct request *rq)
        hlist_del_init(&rq->hash);
 }
 
-static void elv_rqhash_del(request_queue_t *q, struct request *rq)
+static void elv_rqhash_del(struct request_queue *q, struct request *rq)
 {
        if (ELV_ON_HASH(rq))
                __elv_rqhash_del(rq);
 }
 
-static void elv_rqhash_add(request_queue_t *q, struct request *rq)
+static void elv_rqhash_add(struct request_queue *q, struct request *rq)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
 
        BUG_ON(ELV_ON_HASH(rq));
        hlist_add_head(&rq->hash, &e->hash[ELV_HASH_FN(rq_hash_key(rq))]);
 }
 
-static void elv_rqhash_reposition(request_queue_t *q, struct request *rq)
+static void elv_rqhash_reposition(struct request_queue *q, struct request *rq)
 {
        __elv_rqhash_del(rq);
        elv_rqhash_add(q, rq);
 }
 
-static struct request *elv_rqhash_find(request_queue_t *q, sector_t offset)
+static struct request *elv_rqhash_find(struct request_queue *q, sector_t offset)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
        struct hlist_head *hash_list = &e->hash[ELV_HASH_FN(offset)];
        struct hlist_node *entry, *next;
        struct request *rq;
@@ -330,9 +368,9 @@ struct request *elv_rb_add(struct rb_root *root, struct request *rq)
                parent = *p;
                __rq = rb_entry(parent, struct request, rb_node);
 
-               if (rq->sector < __rq->sector)
+               if (blk_rq_pos(rq) < blk_rq_pos(__rq))
                        p = &(*p)->rb_left;
-               else if (rq->sector > __rq->sector)
+               else if (blk_rq_pos(rq) > blk_rq_pos(__rq))
                        p = &(*p)->rb_right;
                else
                        return __rq;
@@ -342,7 +380,6 @@ struct request *elv_rb_add(struct rb_root *root, struct request *rq)
        rb_insert_color(&rq->rb_node, root);
        return NULL;
 }
-
 EXPORT_SYMBOL(elv_rb_add);
 
 void elv_rb_del(struct rb_root *root, struct request *rq)
@@ -351,7 +388,6 @@ void elv_rb_del(struct rb_root *root, struct request *rq)
        rb_erase(&rq->rb_node, root);
        RB_CLEAR_NODE(&rq->rb_node);
 }
-
 EXPORT_SYMBOL(elv_rb_del);
 
 struct request *elv_rb_find(struct rb_root *root, sector_t sector)
@@ -362,9 +398,9 @@ struct request *elv_rb_find(struct rb_root *root, sector_t sector)
        while (n) {
                rq = rb_entry(n, struct request, rb_node);
 
-               if (sector < rq->sector)
+               if (sector < blk_rq_pos(rq))
                        n = n->rb_left;
-               else if (sector > rq->sector)
+               else if (sector > blk_rq_pos(rq))
                        n = n->rb_right;
                else
                        return rq;
@@ -372,18 +408,20 @@ struct request *elv_rb_find(struct rb_root *root, sector_t sector)
 
        return NULL;
 }
-
 EXPORT_SYMBOL(elv_rb_find);
 
 /*
  * Insert rq into dispatch queue of q.  Queue lock must be held on
- * entry.  rq is sort insted into the dispatch queue. To be used by
+ * entry.  rq is sort instead into the dispatch queue. To be used by
  * specific elevators.
  */
-void elv_dispatch_sort(request_queue_t *q, struct request *rq)
+void elv_dispatch_sort(struct request_queue *q, struct request *rq)
 {
        sector_t boundary;
        struct list_head *entry;
+       int stop_flags;
+
+       BUG_ON(rq->cmd_flags & REQ_ON_PLUG);
 
        if (q->last_merge == rq)
                q->last_merge = NULL;
@@ -393,26 +431,30 @@ void elv_dispatch_sort(request_queue_t *q, struct request *rq)
        q->nr_sorted--;
 
        boundary = q->end_sector;
-
+       stop_flags = REQ_SOFTBARRIER | REQ_STARTED;
        list_for_each_prev(entry, &q->queue_head) {
                struct request *pos = list_entry_rq(entry);
 
-               if (pos->cmd_flags & (REQ_SOFTBARRIER|REQ_HARDBARRIER|REQ_STARTED))
+               if ((rq->cmd_flags & REQ_DISCARD) !=
+                   (pos->cmd_flags & REQ_DISCARD))
+                       break;
+               if (rq_data_dir(rq) != rq_data_dir(pos))
                        break;
-               if (rq->sector >= boundary) {
-                       if (pos->sector < boundary)
+               if (pos->cmd_flags & stop_flags)
+                       break;
+               if (blk_rq_pos(rq) >= boundary) {
+                       if (blk_rq_pos(pos) < boundary)
                                continue;
                } else {
-                       if (pos->sector >= boundary)
+                       if (blk_rq_pos(pos) >= boundary)
                                break;
                }
-               if (rq->sector >= pos->sector)
+               if (blk_rq_pos(rq) >= blk_rq_pos(pos))
                        break;
        }
 
        list_add(&rq->queuelist, entry);
 }
-
 EXPORT_SYMBOL(elv_dispatch_sort);
 
 /*
@@ -433,16 +475,24 @@ void elv_dispatch_add_tail(struct request_queue *q, struct request *rq)
        q->boundary_rq = rq;
        list_add_tail(&rq->queuelist, &q->queue_head);
 }
-
 EXPORT_SYMBOL(elv_dispatch_add_tail);
 
-int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
+int elv_merge(struct request_queue *q, struct request **req, struct bio *bio)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
        struct request *__rq;
        int ret;
 
        /*
+        * Levels of merges:
+        *      nomerges:  No merges at all attempted
+        *      noxmerges: Only simple one-hit cache try
+        *      merges:    All merge tries attempted
+        */
+       if (blk_queue_nomerges(q))
+               return ELEVATOR_NO_MERGE;
+
+       /*
         * First try one-hit cache.
         */
        if (q->last_merge) {
@@ -453,6 +503,9 @@ int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
                }
        }
 
+       if (blk_queue_noxmerges(q))
+               return ELEVATOR_NO_MERGE;
+
        /*
         * See if our hash lookup can find a potential backmerge.
         */
@@ -468,9 +521,43 @@ int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
        return ELEVATOR_NO_MERGE;
 }
 
-void elv_merged_request(request_queue_t *q, struct request *rq, int type)
+/*
+ * Attempt to do an insertion back merge. Only check for the case where
+ * we can append 'rq' to an existing request, so we can throw 'rq' away
+ * afterwards.
+ *
+ * Returns true if we merged, false otherwise
+ */
+static bool elv_attempt_insert_merge(struct request_queue *q,
+                                    struct request *rq)
 {
-       elevator_t *e = q->elevator;
+       struct request *__rq;
+
+       if (blk_queue_nomerges(q))
+               return false;
+
+       /*
+        * First try one-hit cache.
+        */
+       if (q->last_merge && blk_attempt_req_merge(q, q->last_merge, rq))
+               return true;
+
+       if (blk_queue_noxmerges(q))
+               return false;
+
+       /*
+        * See if our hash lookup can find a potential backmerge.
+        */
+       __rq = elv_rqhash_find(q, blk_rq_pos(rq));
+       if (__rq && blk_attempt_req_merge(q, __rq, rq))
+               return true;
+
+       return false;
+}
+
+void elv_merged_request(struct request_queue *q, struct request *rq, int type)
+{
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_merged_fn)
                e->ops->elevator_merged_fn(q, rq, type);
@@ -481,41 +568,52 @@ void elv_merged_request(request_queue_t *q, struct request *rq, int type)
        q->last_merge = rq;
 }
 
-void elv_merge_requests(request_queue_t *q, struct request *rq,
+void elv_merge_requests(struct request_queue *q, struct request *rq,
                             struct request *next)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
+       const int next_sorted = next->cmd_flags & REQ_SORTED;
 
-       if (e->ops->elevator_merge_req_fn)
+       if (next_sorted && e->ops->elevator_merge_req_fn)
                e->ops->elevator_merge_req_fn(q, rq, next);
 
        elv_rqhash_reposition(q, rq);
-       elv_rqhash_del(q, next);
 
-       q->nr_sorted--;
+       if (next_sorted) {
+               elv_rqhash_del(q, next);
+               q->nr_sorted--;
+       }
+
        q->last_merge = rq;
 }
 
-void elv_requeue_request(request_queue_t *q, struct request *rq)
+void elv_bio_merged(struct request_queue *q, struct request *rq,
+                       struct bio *bio)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
+
+       if (e->ops->elevator_bio_merged_fn)
+               e->ops->elevator_bio_merged_fn(q, rq, bio);
+}
 
+void elv_requeue_request(struct request_queue *q, struct request *rq)
+{
        /*
         * it already went through dequeue, we need to decrement the
         * in_flight count again
         */
        if (blk_account_rq(rq)) {
-               q->in_flight--;
-               if (blk_sorted_rq(rq) && e->ops->elevator_deactivate_req_fn)
-                       e->ops->elevator_deactivate_req_fn(q, rq);
+               q->in_flight[rq_is_sync(rq)]--;
+               if (rq->cmd_flags & REQ_SORTED)
+                       elv_deactivate_rq(q, rq);
        }
 
        rq->cmd_flags &= ~REQ_STARTED;
 
-       elv_insert(q, rq, ELEVATOR_INSERT_REQUEUE);
+       __elv_add_request(q, rq, ELEVATOR_INSERT_REQUEUE);
 }
 
-static void elv_drain_elevator(request_queue_t *q)
+void elv_drain_elevator(struct request_queue *q)
 {
        static int printed;
        while (q->elevator->ops->elevator_dispatch_fn(q, 1))
@@ -529,20 +627,57 @@ static void elv_drain_elevator(request_queue_t *q)
        }
 }
 
-void elv_insert(request_queue_t *q, struct request *rq, int where)
+/*
+ * Call with queue lock held, interrupts disabled
+ */
+void elv_quiesce_start(struct request_queue *q)
 {
-       struct list_head *pos;
-       unsigned ordseq;
-       int unplug_it = 1;
+       if (!q->elevator)
+               return;
 
-       blk_add_trace_rq(q, rq, BLK_TA_INSERT);
+       queue_flag_set(QUEUE_FLAG_ELVSWITCH, q);
+
+       /*
+        * make sure we don't have any requests in flight
+        */
+       elv_drain_elevator(q);
+       while (q->rq.elvpriv) {
+               __blk_run_queue(q, false);
+               spin_unlock_irq(q->queue_lock);
+               msleep(10);
+               spin_lock_irq(q->queue_lock);
+               elv_drain_elevator(q);
+       }
+}
+
+void elv_quiesce_end(struct request_queue *q)
+{
+       queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q);
+}
+
+void __elv_add_request(struct request_queue *q, struct request *rq, int where)
+{
+       trace_block_rq_insert(q, rq);
 
        rq->q = q;
 
+       BUG_ON(rq->cmd_flags & REQ_ON_PLUG);
+
+       if (rq->cmd_flags & REQ_SOFTBARRIER) {
+               /* barriers are scheduling boundary, update end_sector */
+               if (rq->cmd_type == REQ_TYPE_FS ||
+                   (rq->cmd_flags & REQ_DISCARD)) {
+                       q->end_sector = rq_end_sector(rq);
+                       q->boundary_rq = rq;
+               }
+       } else if (!(rq->cmd_flags & REQ_ELVPRIV) &&
+                   where == ELEVATOR_INSERT_SORT)
+               where = ELEVATOR_INSERT_BACK;
+
        switch (where) {
+       case ELEVATOR_INSERT_REQUEUE:
        case ELEVATOR_INSERT_FRONT:
                rq->cmd_flags |= REQ_SOFTBARRIER;
-
                list_add(&rq->queuelist, &q->queue_head);
                break;
 
@@ -560,12 +695,20 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
                 *   with anything.  There's no point in delaying queue
                 *   processing.
                 */
-               blk_remove_plug(q);
-               q->request_fn(q);
+               __blk_run_queue(q, false);
                break;
 
+       case ELEVATOR_INSERT_SORT_MERGE:
+               /*
+                * If we succeed in merging this request with one in the
+                * queue already, we are done - rq has now been freed,
+                * so no need to do anything further.
+                */
+               if (elv_attempt_insert_merge(q, rq))
+                       break;
        case ELEVATOR_INSERT_SORT:
-               BUG_ON(!blk_fs_request(rq));
+               BUG_ON(rq->cmd_type != REQ_TYPE_FS &&
+                      !(rq->cmd_flags & REQ_DISCARD));
                rq->cmd_flags |= REQ_SORTED;
                q->nr_sorted++;
                if (rq_mergeable(rq)) {
@@ -582,258 +725,68 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
                q->elevator->ops->elevator_add_req_fn(q, rq);
                break;
 
-       case ELEVATOR_INSERT_REQUEUE:
-               /*
-                * If ordered flush isn't in progress, we do front
-                * insertion; otherwise, requests should be requeued
-                * in ordseq order.
-                */
+       case ELEVATOR_INSERT_FLUSH:
                rq->cmd_flags |= REQ_SOFTBARRIER;
-
-               if (q->ordseq == 0) {
-                       list_add(&rq->queuelist, &q->queue_head);
-                       break;
-               }
-
-               ordseq = blk_ordered_req_seq(rq);
-
-               list_for_each(pos, &q->queue_head) {
-                       struct request *pos_rq = list_entry_rq(pos);
-                       if (ordseq <= blk_ordered_req_seq(pos_rq))
-                               break;
-               }
-
-               list_add_tail(&rq->queuelist, pos);
-               /*
-                * most requeues happen because of a busy condition, don't
-                * force unplug of the queue for that case.
-                */
-               unplug_it = 0;
+               blk_insert_flush(rq);
                break;
-
        default:
                printk(KERN_ERR "%s: bad insertion point %d\n",
-                      __FUNCTION__, where);
+                      __func__, where);
                BUG();
        }
-
-       if (unplug_it && blk_queue_plugged(q)) {
-               int nrq = q->rq.count[READ] + q->rq.count[WRITE]
-                       - q->in_flight;
-
-               if (nrq >= q->unplug_thresh)
-                       __generic_unplug_device(q);
-       }
-}
-
-void __elv_add_request(request_queue_t *q, struct request *rq, int where,
-                      int plug)
-{
-       if (q->ordcolor)
-               rq->cmd_flags |= REQ_ORDERED_COLOR;
-
-       if (rq->cmd_flags & (REQ_SOFTBARRIER | REQ_HARDBARRIER)) {
-               /*
-                * toggle ordered color
-                */
-               if (blk_barrier_rq(rq))
-                       q->ordcolor ^= 1;
-
-               /*
-                * barriers implicitly indicate back insertion
-                */
-               if (where == ELEVATOR_INSERT_SORT)
-                       where = ELEVATOR_INSERT_BACK;
-
-               /*
-                * this request is scheduling boundary, update
-                * end_sector
-                */
-               if (blk_fs_request(rq)) {
-                       q->end_sector = rq_end_sector(rq);
-                       q->boundary_rq = rq;
-               }
-       } else if (!(rq->cmd_flags & REQ_ELVPRIV) && where == ELEVATOR_INSERT_SORT)
-               where = ELEVATOR_INSERT_BACK;
-
-       if (plug)
-               blk_plug_device(q);
-
-       elv_insert(q, rq, where);
 }
-
 EXPORT_SYMBOL(__elv_add_request);
 
-void elv_add_request(request_queue_t *q, struct request *rq, int where,
-                    int plug)
+void elv_add_request(struct request_queue *q, struct request *rq, int where)
 {
        unsigned long flags;
 
        spin_lock_irqsave(q->queue_lock, flags);
-       __elv_add_request(q, rq, where, plug);
+       __elv_add_request(q, rq, where);
        spin_unlock_irqrestore(q->queue_lock, flags);
 }
-
 EXPORT_SYMBOL(elv_add_request);
 
-static inline struct request *__elv_next_request(request_queue_t *q)
-{
-       struct request *rq;
-
-       while (1) {
-               while (!list_empty(&q->queue_head)) {
-                       rq = list_entry_rq(q->queue_head.next);
-                       if (blk_do_ordered(q, &rq))
-                               return rq;
-               }
-
-               if (!q->elevator->ops->elevator_dispatch_fn(q, 0))
-                       return NULL;
-       }
-}
-
-struct request *elv_next_request(request_queue_t *q)
+struct request *elv_latter_request(struct request_queue *q, struct request *rq)
 {
-       struct request *rq;
-       int ret;
-
-       while ((rq = __elv_next_request(q)) != NULL) {
-               if (!(rq->cmd_flags & REQ_STARTED)) {
-                       elevator_t *e = q->elevator;
-
-                       /*
-                        * This is the first time the device driver
-                        * sees this request (possibly after
-                        * requeueing).  Notify IO scheduler.
-                        */
-                       if (blk_sorted_rq(rq) &&
-                           e->ops->elevator_activate_req_fn)
-                               e->ops->elevator_activate_req_fn(q, rq);
-
-                       /*
-                        * just mark as started even if we don't start
-                        * it, a request that has been delayed should
-                        * not be passed by new incoming requests
-                        */
-                       rq->cmd_flags |= REQ_STARTED;
-                       blk_add_trace_rq(q, rq, BLK_TA_ISSUE);
-               }
-
-               if (!q->boundary_rq || q->boundary_rq == rq) {
-                       q->end_sector = rq_end_sector(rq);
-                       q->boundary_rq = NULL;
-               }
-
-               if ((rq->cmd_flags & REQ_DONTPREP) || !q->prep_rq_fn)
-                       break;
-
-               ret = q->prep_rq_fn(q, rq);
-               if (ret == BLKPREP_OK) {
-                       break;
-               } else if (ret == BLKPREP_DEFER) {
-                       /*
-                        * the request may have been (partially) prepped.
-                        * we need to keep this request in the front to
-                        * avoid resource deadlock.  REQ_STARTED will
-                        * prevent other fs requests from passing this one.
-                        */
-                       rq = NULL;
-                       break;
-               } else if (ret == BLKPREP_KILL) {
-                       int nr_bytes = rq->hard_nr_sectors << 9;
-
-                       if (!nr_bytes)
-                               nr_bytes = rq->data_len;
-
-                       blkdev_dequeue_request(rq);
-                       rq->cmd_flags |= REQ_QUIET;
-                       end_that_request_chunk(rq, 0, nr_bytes);
-                       end_that_request_last(rq, 0);
-               } else {
-                       printk(KERN_ERR "%s: bad return=%d\n", __FUNCTION__,
-                                                               ret);
-                       break;
-               }
-       }
-
-       return rq;
-}
-
-EXPORT_SYMBOL(elv_next_request);
-
-void elv_dequeue_request(request_queue_t *q, struct request *rq)
-{
-       BUG_ON(list_empty(&rq->queuelist));
-       BUG_ON(ELV_ON_HASH(rq));
-
-       list_del_init(&rq->queuelist);
-
-       /*
-        * the time frame between a request being removed from the lists
-        * and to it is freed is accounted as io that is in progress at
-        * the driver side.
-        */
-       if (blk_account_rq(rq))
-               q->in_flight++;
-}
-
-EXPORT_SYMBOL(elv_dequeue_request);
-
-int elv_queue_empty(request_queue_t *q)
-{
-       elevator_t *e = q->elevator;
-
-       if (!list_empty(&q->queue_head))
-               return 0;
-
-       if (e->ops->elevator_queue_empty_fn)
-               return e->ops->elevator_queue_empty_fn(q);
-
-       return 1;
-}
-
-EXPORT_SYMBOL(elv_queue_empty);
-
-struct request *elv_latter_request(request_queue_t *q, struct request *rq)
-{
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_latter_req_fn)
                return e->ops->elevator_latter_req_fn(q, rq);
        return NULL;
 }
 
-struct request *elv_former_request(request_queue_t *q, struct request *rq)
+struct request *elv_former_request(struct request_queue *q, struct request *rq)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_former_req_fn)
                return e->ops->elevator_former_req_fn(q, rq);
        return NULL;
 }
 
-int elv_set_request(request_queue_t *q, struct request *rq, gfp_t gfp_mask)
+int elv_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_set_req_fn)
                return e->ops->elevator_set_req_fn(q, rq, gfp_mask);
 
-       rq->elevator_private = NULL;
+       rq->elevator_private[0] = NULL;
        return 0;
 }
 
-void elv_put_request(request_queue_t *q, struct request *rq)
+void elv_put_request(struct request_queue *q, struct request *rq)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_put_req_fn)
                e->ops->elevator_put_req_fn(rq);
 }
 
-int elv_may_queue(request_queue_t *q, int rw)
+int elv_may_queue(struct request_queue *q, int rw)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
 
        if (e->ops->elevator_may_queue_fn)
                return e->ops->elevator_may_queue_fn(q, rw);
@@ -841,32 +794,39 @@ int elv_may_queue(request_queue_t *q, int rw)
        return ELV_MQUEUE_MAY;
 }
 
-void elv_completed_request(request_queue_t *q, struct request *rq)
+void elv_abort_queue(struct request_queue *q)
 {
-       elevator_t *e = q->elevator;
+       struct request *rq;
+
+       blk_abort_flushes(q);
+
+       while (!list_empty(&q->queue_head)) {
+               rq = list_entry_rq(q->queue_head.next);
+               rq->cmd_flags |= REQ_QUIET;
+               trace_block_rq_abort(q, rq);
+               /*
+                * Mark this request as started so we don't trigger
+                * any debug logic in the end I/O path.
+                */
+               blk_start_request(rq);
+               __blk_end_request_all(rq, -EIO);
+       }
+}
+EXPORT_SYMBOL(elv_abort_queue);
+
+void elv_completed_request(struct request_queue *q, struct request *rq)
+{
+       struct elevator_queue *e = q->elevator;
 
        /*
         * request is released from the driver, io must be done
         */
        if (blk_account_rq(rq)) {
-               q->in_flight--;
-               if (blk_sorted_rq(rq) && e->ops->elevator_completed_req_fn)
+               q->in_flight[rq_is_sync(rq)]--;
+               if ((rq->cmd_flags & REQ_SORTED) &&
+                   e->ops->elevator_completed_req_fn)
                        e->ops->elevator_completed_req_fn(q, rq);
        }
-
-       /*
-        * Check if the queue is waiting for fs requests to be
-        * drained for flush sequence.
-        */
-       if (unlikely(q->ordseq)) {
-               struct request *first_rq = list_entry_rq(q->queue_head.next);
-               if (q->in_flight == 0 &&
-                   blk_ordered_cur_seq(q) == QUEUE_ORDSEQ_DRAIN &&
-                   blk_ordered_req_seq(first_rq) > QUEUE_ORDSEQ_DRAIN) {
-                       blk_ordered_complete_seq(q, QUEUE_ORDSEQ_DRAIN, 0);
-                       q->request_fn(q);
-               }
-       }
 }
 
 #define to_elv(atr) container_of((atr), struct elv_fs_entry, attr)
@@ -874,13 +834,14 @@ void elv_completed_request(request_queue_t *q, struct request *rq)
 static ssize_t
 elv_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
 {
-       elevator_t *e = container_of(kobj, elevator_t, kobj);
        struct elv_fs_entry *entry = to_elv(attr);
+       struct elevator_queue *e;
        ssize_t error;
 
        if (!entry->show)
                return -EIO;
 
+       e = container_of(kobj, struct elevator_queue, kobj);
        mutex_lock(&e->sysfs_lock);
        error = e->ops ? entry->show(e, page) : -ENOENT;
        mutex_unlock(&e->sysfs_lock);
@@ -891,20 +852,21 @@ static ssize_t
 elv_attr_store(struct kobject *kobj, struct attribute *attr,
               const char *page, size_t length)
 {
-       elevator_t *e = container_of(kobj, elevator_t, kobj);
        struct elv_fs_entry *entry = to_elv(attr);
+       struct elevator_queue *e;
        ssize_t error;
 
        if (!entry->store)
                return -EIO;
 
+       e = container_of(kobj, struct elevator_queue, kobj);
        mutex_lock(&e->sysfs_lock);
        error = e->ops ? entry->store(e, page, length) : -ENOENT;
        mutex_unlock(&e->sysfs_lock);
        return error;
 }
 
-static struct sysfs_ops elv_sysfs_ops = {
+static const struct sysfs_ops elv_sysfs_ops = {
        .show   = elv_attr_show,
        .store  = elv_attr_store,
 };
@@ -916,12 +878,10 @@ static struct kobj_type elv_ktype = {
 
 int elv_register_queue(struct request_queue *q)
 {
-       elevator_t *e = q->elevator;
+       struct elevator_queue *e = q->elevator;
        int error;
 
-       e->kobj.parent = &q->kobj;
-
-       error = kobject_add(&e->kobj);
+       error = kobject_add(&e->kobj, &q->kobj, "%s", "iosched");
        if (!error) {
                struct elv_fs_entry *attr = e->elevator_type->elevator_attrs;
                if (attr) {
@@ -932,14 +892,17 @@ int elv_register_queue(struct request_queue *q)
                        }
                }
                kobject_uevent(&e->kobj, KOBJ_ADD);
+               e->registered = 1;
        }
        return error;
 }
+EXPORT_SYMBOL(elv_register_queue);
 
-static void __elv_unregister_queue(elevator_t *e)
+static void __elv_unregister_queue(struct elevator_queue *e)
 {
        kobject_uevent(&e->kobj, KOBJ_REMOVE);
        kobject_del(&e->kobj);
+       e->registered = 0;
 }
 
 void elv_unregister_queue(struct request_queue *q)
@@ -947,21 +910,24 @@ void elv_unregister_queue(struct request_queue *q)
        if (q)
                __elv_unregister_queue(q->elevator);
 }
+EXPORT_SYMBOL(elv_unregister_queue);
 
-int elv_register(struct elevator_type *e)
+void elv_register(struct elevator_type *e)
 {
-       spin_lock_irq(&elv_list_lock);
+       char *def = "";
+
+       spin_lock(&elv_list_lock);
        BUG_ON(elevator_find(e->elevator_name));
        list_add_tail(&e->list, &elv_list);
-       spin_unlock_irq(&elv_list_lock);
+       spin_unlock(&elv_list_lock);
 
-       printk(KERN_INFO "io scheduler %s registered", e->elevator_name);
        if (!strcmp(e->elevator_name, chosen_elevator) ||
                        (!*chosen_elevator &&
                         !strcmp(e->elevator_name, CONFIG_DEFAULT_IOSCHED)))
-                               printk(" (default)");
-       printk("\n");
-       return 0;
+                               def = " (default)";
+
+       printk(KERN_INFO "io scheduler %s registered%s\n", e->elevator_name,
+                                                               def);
 }
 EXPORT_SYMBOL_GPL(elv_register);
 
@@ -983,9 +949,9 @@ void elv_unregister(struct elevator_type *e)
                read_unlock(&tasklist_lock);
        }
 
-       spin_lock_irq(&elv_list_lock);
+       spin_lock(&elv_list_lock);
        list_del_init(&e->list);
-       spin_unlock_irq(&elv_list_lock);
+       spin_unlock(&elv_list_lock);
 }
 EXPORT_SYMBOL_GPL(elv_unregister);
 
@@ -995,41 +961,30 @@ EXPORT_SYMBOL_GPL(elv_unregister);
  * need for the new one. this way we have a chance of going back to the old
  * one, if the new one fails init for some reason.
  */
-static int elevator_switch(request_queue_t *q, struct elevator_type *new_e)
+static int elevator_switch(struct request_queue *q, struct elevator_type *new_e)
 {
-       elevator_t *old_elevator, *e;
+       struct elevator_queue *old_elevator, *e;
        void *data;
+       int err;
 
        /*
         * Allocate new elevator
         */
        e = elevator_alloc(q, new_e);
        if (!e)
-               return 0;
+               return -ENOMEM;
 
        data = elevator_init_queue(q, e);
        if (!data) {
                kobject_put(&e->kobj);
-               return 0;
+               return -ENOMEM;
        }
 
        /*
         * Turn on BYPASS and drain all requests w/ elevator private data
         */
        spin_lock_irq(q->queue_lock);
-
-       set_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
-
-       elv_drain_elevator(q);
-
-       while (q->rq.elvpriv) {
-               blk_remove_plug(q);
-               q->request_fn(q);
-               spin_unlock_irq(q->queue_lock);
-               msleep(10);
-               spin_lock_irq(q->queue_lock);
-               elv_drain_elevator(q);
-       }
+       elv_quiesce_start(q);
 
        /*
         * Remember old elevator.
@@ -1043,17 +998,25 @@ static int elevator_switch(request_queue_t *q, struct elevator_type *new_e)
 
        spin_unlock_irq(q->queue_lock);
 
-       __elv_unregister_queue(old_elevator);
+       if (old_elevator->registered) {
+               __elv_unregister_queue(old_elevator);
 
-       if (elv_register_queue(q))
-               goto fail_register;
+               err = elv_register_queue(q);
+               if (err)
+                       goto fail_register;
+       }
 
        /*
         * finally exit old elevator and turn off BYPASS.
         */
        elevator_exit(old_elevator);
-       clear_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
-       return 1;
+       spin_lock_irq(q->queue_lock);
+       elv_quiesce_end(q);
+       spin_unlock_irq(q->queue_lock);
+
+       blk_add_trace_msg(q, "elv switch: %s", e->elevator_type->elevator_name);
+
+       return 0;
 
 fail_register:
        /*
@@ -1063,24 +1026,27 @@ fail_register:
        elevator_exit(e);
        q->elevator = old_elevator;
        elv_register_queue(q);
-       clear_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
-       return 0;
+
+       spin_lock_irq(q->queue_lock);
+       queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q);
+       spin_unlock_irq(q->queue_lock);
+
+       return err;
 }
 
-ssize_t elv_iosched_store(request_queue_t *q, const char *name, size_t count)
+/*
+ * Switch this queue to the given IO scheduler.
+ */
+int elevator_change(struct request_queue *q, const char *name)
 {
        char elevator_name[ELV_NAME_MAX];
-       size_t len;
        struct elevator_type *e;
 
-       elevator_name[sizeof(elevator_name) - 1] = '\0';
-       strncpy(elevator_name, name, sizeof(elevator_name) - 1);
-       len = strlen(elevator_name);
+       if (!q->elevator)
+               return -ENXIO;
 
-       if (len && elevator_name[len - 1] == '\n')
-               elevator_name[len - 1] = '\0';
-
-       e = elevator_get(elevator_name);
+       strlcpy(elevator_name, name, sizeof(elevator_name));
+       e = elevator_get(strstrip(elevator_name));
        if (!e) {
                printk(KERN_ERR "elevator: type %s not found\n", elevator_name);
                return -EINVAL;
@@ -1088,38 +1054,56 @@ ssize_t elv_iosched_store(request_queue_t *q, const char *name, size_t count)
 
        if (!strcmp(elevator_name, q->elevator->elevator_type->elevator_name)) {
                elevator_put(e);
-               return count;
+               return 0;
        }
 
-       if (!elevator_switch(q, e))
-               printk(KERN_ERR "elevator: switch to %s failed\n",elevator_name);
-       return count;
+       return elevator_switch(q, e);
 }
+EXPORT_SYMBOL(elevator_change);
 
-ssize_t elv_iosched_show(request_queue_t *q, char *name)
+ssize_t elv_iosched_store(struct request_queue *q, const char *name,
+                         size_t count)
 {
-       elevator_t *e = q->elevator;
-       struct elevator_type *elv = e->elevator_type;
-       struct list_head *entry;
+       int ret;
+
+       if (!q->elevator)
+               return count;
+
+       ret = elevator_change(q, name);
+       if (!ret)
+               return count;
+
+       printk(KERN_ERR "elevator: switch to %s failed\n", name);
+       return ret;
+}
+
+ssize_t elv_iosched_show(struct request_queue *q, char *name)
+{
+       struct elevator_queue *e = q->elevator;
+       struct elevator_type *elv;
+       struct elevator_type *__e;
        int len = 0;
 
-       spin_lock_irq(&elv_list_lock);
-       list_for_each(entry, &elv_list) {
-               struct elevator_type *__e;
+       if (!q->elevator || !blk_queue_stackable(q))
+               return sprintf(name, "none\n");
 
-               __e = list_entry(entry, struct elevator_type, list);
+       elv = e->elevator_type;
+
+       spin_lock(&elv_list_lock);
+       list_for_each_entry(__e, &elv_list, list) {
                if (!strcmp(elv->elevator_name, __e->elevator_name))
                        len += sprintf(name+len, "[%s] ", elv->elevator_name);
                else
                        len += sprintf(name+len, "%s ", __e->elevator_name);
        }
-       spin_unlock_irq(&elv_list_lock);
+       spin_unlock(&elv_list_lock);
 
        len += sprintf(len+name, "\n");
        return len;
 }
 
-struct request *elv_rb_former_request(request_queue_t *q, struct request *rq)
+struct request *elv_rb_former_request(struct request_queue *q,
+                                     struct request *rq)
 {
        struct rb_node *rbprev = rb_prev(&rq->rb_node);
 
@@ -1128,10 +1112,10 @@ struct request *elv_rb_former_request(request_queue_t *q, struct request *rq)
 
        return NULL;
 }
-
 EXPORT_SYMBOL(elv_rb_former_request);
 
-struct request *elv_rb_latter_request(request_queue_t *q, struct request *rq)
+struct request *elv_rb_latter_request(struct request_queue *q,
+                                     struct request *rq)
 {
        struct rb_node *rbnext = rb_next(&rq->rb_node);
 
@@ -1140,5 +1124,4 @@ struct request *elv_rb_latter_request(request_queue_t *q, struct request *rq)
 
        return NULL;
 }
-
 EXPORT_SYMBOL(elv_rb_latter_request);