#include <linux/delay.h>
#include <linux/blktrace_api.h>
#include <linux/hash.h>
+#include <linux/uaccess.h>
-#include <asm/uaccess.h>
+#include "blk.h"
static DEFINE_SPINLOCK(elv_list_lock);
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))
*/
static int elv_iosched_allow_merge(struct request *rq, struct bio *bio)
{
- request_queue_t *q = rq->q;
+ struct request_queue *q = rq->q;
elevator_t *e = q->elevator;
if (e->ops->elevator_allow_merge_fn)
/*
* 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_discard(bio) != bio_discard(rq->bio))
+ return 0;
+
+ /*
* different data direction or already started, don't merge
*/
if (bio_data_dir(bio) != rq_data_dir(rq))
/*
* must be same device and not a special request
*/
- if (rq->rq_disk != bio->bi_bdev->bd_disk || !rq->special)
+ 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))
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;
}
{
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);
+
+ if (!strcmp(name, "anticipatory"))
+ sprintf(elv, "as-iosched");
+ else
+ sprintf(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;
static struct kobj_type elv_ktype;
-static elevator_t *elevator_alloc(request_queue_t *q, struct elevator_type *e)
+static elevator_t *elevator_alloc(struct request_queue *q,
+ struct elevator_type *e)
{
elevator_t *eq;
int i;
- eq = kmalloc_node(sizeof(elevator_t), GFP_KERNEL, q->node);
+ eq = kmalloc_node(sizeof(elevator_t), 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,
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;
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);
elevator_attach(q, eq, data);
return ret;
}
-
EXPORT_SYMBOL(elevator_init);
void elevator_exit(elevator_t *e)
kobject_put(&e->kobj);
}
-
EXPORT_SYMBOL(elevator_exit);
+static void elv_activate_rq(struct request_queue *q, struct request *rq)
+{
+ elevator_t *e = q->elevator;
+
+ if (e->ops->elevator_activate_req_fn)
+ e->ops->elevator_activate_req_fn(q, rq);
+}
+
+static void elv_deactivate_rq(struct request_queue *q, struct request *rq)
+{
+ elevator_t *e = q->elevator;
+
+ if (e->ops->elevator_deactivate_req_fn)
+ e->ops->elevator_deactivate_req_fn(q, rq);
+}
+
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;
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 hlist_head *hash_list = &e->hash[ELV_HASH_FN(offset)];
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)
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)
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;
if (q->last_merge == rq)
q->last_merge = NULL;
q->nr_sorted--;
boundary = q->end_sector;
-
+ stop_flags = REQ_SOFTBARRIER | REQ_HARDBARRIER | 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 (blk_discard_rq(rq) != blk_discard_rq(pos))
+ break;
+ if (rq_data_dir(rq) != rq_data_dir(pos))
+ break;
+ if (pos->cmd_flags & stop_flags)
break;
if (rq->sector >= boundary) {
if (pos->sector < boundary)
list_add(&rq->queuelist, entry);
}
-
EXPORT_SYMBOL(elv_dispatch_sort);
/*
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 request *__rq;
}
}
+ if (blk_queue_nomerges(q))
+ return ELEVATOR_NO_MERGE;
+
/*
* See if our hash lookup can find a potential backmerge.
*/
return ELEVATOR_NO_MERGE;
}
-void elv_merged_request(request_queue_t *q, struct request *rq, int type)
+void elv_merged_request(struct request_queue *q, struct request *rq, int type)
{
elevator_t *e = q->elevator;
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;
q->last_merge = rq;
}
-void elv_requeue_request(request_queue_t *q, struct request *rq)
+void elv_requeue_request(struct request_queue *q, struct request *rq)
{
- elevator_t *e = q->elevator;
-
/*
* 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);
+ if (blk_sorted_rq(rq))
+ elv_deactivate_rq(q, rq);
}
rq->cmd_flags &= ~REQ_STARTED;
elv_insert(q, rq, ELEVATOR_INSERT_REQUEUE);
}
-static void elv_drain_elevator(request_queue_t *q)
+static void elv_drain_elevator(struct request_queue *q)
{
static int printed;
while (q->elevator->ops->elevator_dispatch_fn(q, 1))
}
}
-void elv_insert(request_queue_t *q, struct request *rq, int where)
+void elv_insert(struct request_queue *q, struct request *rq, int where)
{
struct list_head *pos;
unsigned ordseq;
break;
case ELEVATOR_INSERT_SORT:
- BUG_ON(!blk_fs_request(rq));
+ BUG_ON(!blk_fs_request(rq) && !blk_discard_rq(rq));
rq->cmd_flags |= REQ_SORTED;
q->nr_sorted++;
if (rq_mergeable(rq)) {
*/
rq->cmd_flags |= REQ_SOFTBARRIER;
+ /*
+ * Most requeues happen because of a busy condition,
+ * don't force unplug of the queue for that case.
+ */
+ unplug_it = 0;
+
if (q->ordseq == 0) {
list_add(&rq->queuelist, &q->queue_head);
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;
break;
default:
printk(KERN_ERR "%s: bad insertion point %d\n",
- __FUNCTION__, where);
+ __func__, where);
BUG();
}
}
}
-void __elv_add_request(request_queue_t *q, struct request *rq, int where,
+void __elv_add_request(struct request_queue *q, struct request *rq, int where,
int plug)
{
if (q->ordcolor)
* this request is scheduling boundary, update
* end_sector
*/
- if (blk_fs_request(rq)) {
+ if (blk_fs_request(rq) || blk_discard_rq(rq)) {
q->end_sector = rq_end_sector(rq);
q->boundary_rq = rq;
}
- } else if (!(rq->cmd_flags & REQ_ELVPRIV) && where == ELEVATOR_INSERT_SORT)
+ } else if (!(rq->cmd_flags & REQ_ELVPRIV) &&
+ where == ELEVATOR_INSERT_SORT)
where = ELEVATOR_INSERT_BACK;
if (plug)
elv_insert(q, rq, where);
}
-
EXPORT_SYMBOL(__elv_add_request);
-void elv_add_request(request_queue_t *q, struct request *rq, int where,
+void elv_add_request(struct request_queue *q, struct request *rq, int where,
int plug)
{
unsigned long flags;
__elv_add_request(q, rq, where, plug);
spin_unlock_irqrestore(q->queue_lock, flags);
}
-
EXPORT_SYMBOL(elv_add_request);
-static inline struct request *__elv_next_request(request_queue_t *q)
+static inline struct request *__elv_next_request(struct request_queue *q)
{
struct request *rq;
}
}
-struct request *elv_next_request(request_queue_t *q)
+struct request *elv_next_request(struct request_queue *q)
{
struct request *rq;
int ret;
while ((rq = __elv_next_request(q)) != NULL) {
+ /*
+ * Kill the empty barrier place holder, the driver must
+ * not ever see it.
+ */
+ if (blk_empty_barrier(rq)) {
+ __blk_end_request(rq, 0, blk_rq_bytes(rq));
+ continue;
+ }
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);
+ if (blk_sorted_rq(rq))
+ elv_activate_rq(q, rq);
/*
* just mark as started even if we don't start
*/
rq->cmd_flags |= REQ_STARTED;
blk_add_trace_rq(q, rq, BLK_TA_ISSUE);
+
+ /*
+ * We are now handing the request to the hardware,
+ * add the timeout handler
+ */
+ blk_add_timer(rq);
}
if (!q->boundary_rq || q->boundary_rq == rq) {
q->boundary_rq = NULL;
}
- if ((rq->cmd_flags & REQ_DONTPREP) || !q->prep_rq_fn)
+ if (rq->cmd_flags & REQ_DONTPREP)
+ break;
+
+ if (q->dma_drain_size && rq->data_len) {
+ /*
+ * make sure space for the drain appears we
+ * know we can do this because max_hw_segments
+ * has been adjusted to be one fewer than the
+ * device can handle
+ */
+ rq->nr_phys_segments++;
+ }
+
+ if (!q->prep_rq_fn)
break;
ret = q->prep_rq_fn(q, rq);
* avoid resource deadlock. REQ_STARTED will
* prevent other fs requests from passing this one.
*/
+ if (q->dma_drain_size && rq->data_len &&
+ !(rq->cmd_flags & REQ_DONTPREP)) {
+ /*
+ * remove the space for the drain we added
+ * so that we don't add it again
+ */
+ --rq->nr_phys_segments;
+ }
+
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);
+ __blk_end_request(rq, -EIO, blk_rq_bytes(rq));
} else {
- printk(KERN_ERR "%s: bad return=%d\n", __FUNCTION__,
- ret);
+ printk(KERN_ERR "%s: bad return=%d\n", __func__, ret);
break;
}
}
return rq;
}
-
EXPORT_SYMBOL(elv_next_request);
-void elv_dequeue_request(request_queue_t *q, struct request *rq)
+void elv_dequeue_request(struct request_queue *q, struct request *rq)
{
BUG_ON(list_empty(&rq->queuelist));
BUG_ON(ELV_ON_HASH(rq));
if (blk_account_rq(rq))
q->in_flight++;
}
-
EXPORT_SYMBOL(elv_dequeue_request);
-int elv_queue_empty(request_queue_t *q)
+int elv_queue_empty(struct request_queue *q)
{
elevator_t *e = q->elevator;
return 1;
}
-
EXPORT_SYMBOL(elv_queue_empty);
-struct request *elv_latter_request(request_queue_t *q, struct request *rq)
+struct request *elv_latter_request(struct request_queue *q, struct request *rq)
{
elevator_t *e = q->elevator;
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;
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;
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;
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;
return ELV_MQUEUE_MAY;
}
-void elv_completed_request(request_queue_t *q, struct request *rq)
+void elv_abort_queue(struct request_queue *q)
+{
+ struct request *rq;
+
+ while (!list_empty(&q->queue_head)) {
+ rq = list_entry_rq(q->queue_head.next);
+ rq->cmd_flags |= REQ_QUIET;
+ blk_add_trace_rq(q, rq, BLK_TA_ABORT);
+ __blk_end_request(rq, -EIO, blk_rq_bytes(rq));
+ }
+}
+EXPORT_SYMBOL(elv_abort_queue);
+
+void elv_completed_request(struct request_queue *q, struct request *rq)
{
elevator_t *e = q->elevator;
elevator_t *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) {
__elv_unregister_queue(q->elevator);
}
-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);
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);
* 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;
void *data;
*/
spin_lock_irq(q->queue_lock);
- set_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
+ queue_flag_set(QUEUE_FLAG_ELVSWITCH, q);
elv_drain_elevator(q);
* finally exit old elevator and turn off BYPASS.
*/
elevator_exit(old_elevator);
- clear_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
+ spin_lock_irq(q->queue_lock);
+ queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q);
+ spin_unlock_irq(q->queue_lock);
+
+ blk_add_trace_msg(q, "elv switch: %s", e->elevator_type->elevator_name);
+
return 1;
fail_register:
elevator_exit(e);
q->elevator = old_elevator;
elv_register_queue(q);
- clear_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
+
+ spin_lock_irq(q->queue_lock);
+ queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q);
+ spin_unlock_irq(q->queue_lock);
+
return 0;
}
-ssize_t elv_iosched_store(request_queue_t *q, const char *name, size_t count)
+ssize_t elv_iosched_store(struct request_queue *q, const char *name,
+ size_t count)
{
char elevator_name[ELV_NAME_MAX];
size_t len;
}
if (!elevator_switch(q, e))
- printk(KERN_ERR "elevator: switch to %s failed\n",elevator_name);
+ printk(KERN_ERR "elevator: switch to %s failed\n",
+ elevator_name);
return count;
}
-ssize_t elv_iosched_show(request_queue_t *q, char *name)
+ssize_t elv_iosched_show(struct request_queue *q, char *name)
{
elevator_t *e = q->elevator;
struct elevator_type *elv = e->elevator_type;
- struct list_head *entry;
+ struct elevator_type *__e;
int len = 0;
- spin_lock_irq(&elv_list_lock);
- list_for_each(entry, &elv_list) {
- struct elevator_type *__e;
-
- __e = list_entry(entry, struct elevator_type, list);
+ 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);
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);
return NULL;
}
-
EXPORT_SYMBOL(elv_rb_latter_request);
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff