/*
* Check if any of the ctx's have pending work in this hardware queue
*/
-bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
+static bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
{
- return sbitmap_any_bit_set(&hctx->ctx_map) ||
- !list_empty_careful(&hctx->dispatch) ||
+ return !list_empty_careful(&hctx->dispatch) ||
+ sbitmap_any_bit_set(&hctx->ctx_map) ||
blk_mq_sched_has_work(hctx);
}
freeze_depth = atomic_inc_return(&q->mq_freeze_depth);
if (freeze_depth == 1) {
percpu_ref_kill(&q->q_usage_counter);
- blk_mq_run_hw_queues(q, false);
+ if (q->mq_ops)
+ blk_mq_run_hw_queues(q, false);
}
}
EXPORT_SYMBOL_GPL(blk_freeze_queue_start);
queue_for_each_hw_ctx(q, hctx, i)
if (blk_mq_hw_queue_mapped(hctx))
blk_mq_tag_wakeup_all(hctx->tags, true);
-
- /*
- * If we are called because the queue has now been marked as
- * dying, we need to ensure that processes currently waiting on
- * the queue are notified as well.
- */
- wake_up_all(&q->mq_freeze_wq);
}
bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
rq->q = data->q;
rq->mq_ctx = data->ctx;
rq->cmd_flags = op;
+ if (data->flags & BLK_MQ_REQ_PREEMPT)
+ rq->rq_flags |= RQF_PREEMPT;
if (blk_queue_io_stat(data->q))
rq->rq_flags |= RQF_IO_STAT;
/* do not touch atomic flags, it needs atomic ops against the timer */
}
struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
- unsigned int flags)
+ blk_mq_req_flags_t flags)
{
struct blk_mq_alloc_data alloc_data = { .flags = flags };
struct request *rq;
int ret;
- ret = blk_queue_enter(q, flags & BLK_MQ_REQ_NOWAIT);
+ ret = blk_queue_enter(q, flags);
if (ret)
return ERR_PTR(ret);
EXPORT_SYMBOL(blk_mq_alloc_request);
struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
- unsigned int op, unsigned int flags, unsigned int hctx_idx)
+ unsigned int op, blk_mq_req_flags_t flags, unsigned int hctx_idx)
{
struct blk_mq_alloc_data alloc_data = { .flags = flags };
struct request *rq;
if (hctx_idx >= q->nr_hw_queues)
return ERR_PTR(-EIO);
- ret = blk_queue_enter(q, true);
+ ret = blk_queue_enter(q, flags);
if (ret)
return ERR_PTR(ret);
{
struct request_queue *q = rq->q;
+ blk_mq_put_driver_tag(rq);
+
trace_block_rq_requeue(q, rq);
wbt_requeue(q->rq_wb, &rq->issue_stat);
blk_mq_sched_requeue_request(rq);
/*
* We abuse this flag that is otherwise used by the I/O scheduler to
- * request head insertation from the workqueue.
+ * request head insertion from the workqueue.
*/
BUG_ON(rq->rq_flags & RQF_SOFTBARRIER);
return rq->tag != -1;
}
-static void __blk_mq_put_driver_tag(struct blk_mq_hw_ctx *hctx,
- struct request *rq)
-{
- blk_mq_put_tag(hctx, hctx->tags, rq->mq_ctx, rq->tag);
- rq->tag = -1;
-
- if (rq->rq_flags & RQF_MQ_INFLIGHT) {
- rq->rq_flags &= ~RQF_MQ_INFLIGHT;
- atomic_dec(&hctx->nr_active);
- }
-}
-
-static void blk_mq_put_driver_tag_hctx(struct blk_mq_hw_ctx *hctx,
- struct request *rq)
-{
- if (rq->tag == -1 || rq->internal_tag == -1)
- return;
-
- __blk_mq_put_driver_tag(hctx, rq);
-}
-
-static void blk_mq_put_driver_tag(struct request *rq)
+static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode,
+ int flags, void *key)
{
struct blk_mq_hw_ctx *hctx;
- if (rq->tag == -1 || rq->internal_tag == -1)
- return;
+ hctx = container_of(wait, struct blk_mq_hw_ctx, dispatch_wait);
- hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu);
- __blk_mq_put_driver_tag(hctx, rq);
+ list_del_init(&wait->entry);
+ blk_mq_run_hw_queue(hctx, true);
+ return 1;
}
/*
- * If we fail getting a driver tag because all the driver tags are already
- * assigned and on the dispatch list, BUT the first entry does not have a
- * tag, then we could deadlock. For that case, move entries with assigned
- * driver tags to the front, leaving the set of tagged requests in the
- * same order, and the untagged set in the same order.
+ * Mark us waiting for a tag. For shared tags, this involves hooking us into
+ * the tag wakeups. For non-shared tags, we can simply mark us nedeing a
+ * restart. For both caes, take care to check the condition again after
+ * marking us as waiting.
*/
-static bool reorder_tags_to_front(struct list_head *list)
+static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx,
+ struct request *rq)
{
- struct request *rq, *tmp, *first = NULL;
-
- list_for_each_entry_safe_reverse(rq, tmp, list, queuelist) {
- if (rq == first)
- break;
- if (rq->tag != -1) {
- list_move(&rq->queuelist, list);
- if (!first)
- first = rq;
- }
- }
-
- return first != NULL;
-}
-
-static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode, int flags,
- void *key)
-{
- struct blk_mq_hw_ctx *hctx;
+ struct blk_mq_hw_ctx *this_hctx = *hctx;
+ bool shared_tags = (this_hctx->flags & BLK_MQ_F_TAG_SHARED) != 0;
+ struct sbq_wait_state *ws;
+ wait_queue_entry_t *wait;
+ bool ret;
- hctx = container_of(wait, struct blk_mq_hw_ctx, dispatch_wait);
+ if (!shared_tags) {
+ if (!test_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state))
+ set_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state);
+ } else {
+ wait = &this_hctx->dispatch_wait;
+ if (!list_empty_careful(&wait->entry))
+ return false;
- list_del(&wait->entry);
- clear_bit_unlock(BLK_MQ_S_TAG_WAITING, &hctx->state);
- blk_mq_run_hw_queue(hctx, true);
- return 1;
-}
+ spin_lock(&this_hctx->lock);
+ if (!list_empty(&wait->entry)) {
+ spin_unlock(&this_hctx->lock);
+ return false;
+ }
-static bool blk_mq_dispatch_wait_add(struct blk_mq_hw_ctx *hctx)
-{
- struct sbq_wait_state *ws;
+ ws = bt_wait_ptr(&this_hctx->tags->bitmap_tags, this_hctx);
+ add_wait_queue(&ws->wait, wait);
+ }
/*
- * The TAG_WAITING bit serves as a lock protecting hctx->dispatch_wait.
- * The thread which wins the race to grab this bit adds the hardware
- * queue to the wait queue.
+ * It's possible that a tag was freed in the window between the
+ * allocation failure and adding the hardware queue to the wait
+ * queue.
*/
- if (test_bit(BLK_MQ_S_TAG_WAITING, &hctx->state) ||
- test_and_set_bit_lock(BLK_MQ_S_TAG_WAITING, &hctx->state))
- return false;
+ ret = blk_mq_get_driver_tag(rq, hctx, false);
- init_waitqueue_func_entry(&hctx->dispatch_wait, blk_mq_dispatch_wake);
- ws = bt_wait_ptr(&hctx->tags->bitmap_tags, hctx);
+ if (!shared_tags) {
+ /*
+ * Don't clear RESTART here, someone else could have set it.
+ * At most this will cost an extra queue run.
+ */
+ return ret;
+ } else {
+ if (!ret) {
+ spin_unlock(&this_hctx->lock);
+ return false;
+ }
- /*
- * As soon as this returns, it's no longer safe to fiddle with
- * hctx->dispatch_wait, since a completion can wake up the wait queue
- * and unlock the bit.
- */
- add_wait_queue(&ws->wait, &hctx->dispatch_wait);
- return true;
+ /*
+ * We got a tag, remove ourselves from the wait queue to ensure
+ * someone else gets the wakeup.
+ */
+ spin_lock_irq(&ws->wait.lock);
+ list_del_init(&wait->entry);
+ spin_unlock_irq(&ws->wait.lock);
+ spin_unlock(&this_hctx->lock);
+ return true;
+ }
}
bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list,
- bool got_budget)
+ bool got_budget)
{
struct blk_mq_hw_ctx *hctx;
struct request *rq, *nxt;
+ bool no_tag = false;
int errors, queued;
if (list_empty(list))
rq = list_first_entry(list, struct request, queuelist);
if (!blk_mq_get_driver_tag(rq, &hctx, false)) {
- if (!queued && reorder_tags_to_front(list))
- continue;
-
/*
* The initial allocation attempt failed, so we need to
- * rerun the hardware queue when a tag is freed.
- */
- if (!blk_mq_dispatch_wait_add(hctx)) {
- if (got_budget)
- blk_mq_put_dispatch_budget(hctx);
- break;
- }
-
- /*
- * It's possible that a tag was freed in the window
- * between the allocation failure and adding the
- * hardware queue to the wait queue.
+ * rerun the hardware queue when a tag is freed. The
+ * waitqueue takes care of that. If the queue is run
+ * before we add this entry back on the dispatch list,
+ * we'll re-run it below.
*/
- if (!blk_mq_get_driver_tag(rq, &hctx, false)) {
+ if (!blk_mq_mark_tag_wait(&hctx, rq)) {
if (got_budget)
blk_mq_put_dispatch_budget(hctx);
+ /*
+ * For non-shared tags, the RESTART check
+ * will suffice.
+ */
+ if (hctx->flags & BLK_MQ_F_TAG_SHARED)
+ no_tag = true;
break;
}
}
- if (!got_budget && !blk_mq_get_dispatch_budget(hctx))
+ if (!got_budget && !blk_mq_get_dispatch_budget(hctx)) {
+ blk_mq_put_driver_tag(rq);
break;
+ }
list_del_init(&rq->queuelist);
if (ret == BLK_STS_RESOURCE) {
/*
* If an I/O scheduler has been configured and we got a
- * driver tag for the next request already, free it again.
+ * driver tag for the next request already, free it
+ * again.
*/
if (!list_empty(list)) {
nxt = list_first_entry(list, struct request, queuelist);
blk_mq_put_driver_tag(nxt);
}
- blk_mq_put_driver_tag_hctx(hctx, rq);
list_add(&rq->queuelist, list);
__blk_mq_requeue_request(rq);
break;
* it is no longer set that means that it was cleared by another
* thread and hence that a queue rerun is needed.
*
- * If TAG_WAITING is set that means that an I/O scheduler has
- * been configured and another thread is waiting for a driver
- * tag. To guarantee fairness, do not rerun this hardware queue
- * but let the other thread grab the driver tag.
+ * If 'no_tag' is set, that means that we failed getting
+ * a driver tag with an I/O scheduler attached. If our dispatch
+ * waitqueue is no longer active, ensure that we run the queue
+ * AFTER adding our entries back to the list.
*
* If no I/O scheduler has been configured it is possible that
* the hardware queue got stopped and restarted before requests
* returning BLK_STS_RESOURCE. Two exceptions are scsi-mq
* and dm-rq.
*/
- if (!blk_mq_sched_needs_restart(hctx) &&
- !test_bit(BLK_MQ_S_TAG_WAITING, &hctx->state))
+ if (!blk_mq_sched_needs_restart(hctx) ||
+ (no_tag && list_empty_careful(&hctx->dispatch_wait.entry)))
blk_mq_run_hw_queue(hctx, true);
}
}
EXPORT_SYMBOL(blk_mq_delay_run_hw_queue);
-void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
+bool blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
{
- __blk_mq_delay_run_hw_queue(hctx, async, 0);
+ if (blk_mq_hctx_has_pending(hctx)) {
+ __blk_mq_delay_run_hw_queue(hctx, async, 0);
+ return true;
+ }
+
+ return false;
}
EXPORT_SYMBOL(blk_mq_run_hw_queue);
int i;
queue_for_each_hw_ctx(q, hctx, i) {
- if (!blk_mq_hctx_has_pending(hctx) ||
- blk_mq_hctx_stopped(hctx))
+ if (blk_mq_hctx_stopped(hctx))
continue;
blk_mq_run_hw_queue(hctx, async);
* Should only be used carefully, when the caller knows we want to
* bypass a potential IO scheduler on the target device.
*/
-void blk_mq_request_bypass_insert(struct request *rq)
+void blk_mq_request_bypass_insert(struct request *rq, bool run_queue)
{
struct blk_mq_ctx *ctx = rq->mq_ctx;
struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(rq->q, ctx->cpu);
list_add_tail(&rq->queuelist, &hctx->dispatch);
spin_unlock(&hctx->lock);
- blk_mq_run_hw_queue(hctx, false);
+ if (run_queue)
+ blk_mq_run_hw_queue(hctx, false);
}
void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx,
if (unlikely(is_flush_fua)) {
blk_mq_put_ctx(data.ctx);
blk_mq_bio_to_request(rq, bio);
- if (q->elevator) {
- blk_mq_sched_insert_request(rq, false, true, true,
- true);
- } else {
- blk_insert_flush(rq);
- blk_mq_run_hw_queue(data.hctx, true);
- }
+
+ /* bypass scheduler for flush rq */
+ blk_insert_flush(rq);
+ blk_mq_run_hw_queue(data.hctx, true);
} else if (plug && q->nr_hw_queues == 1) {
struct request *last = NULL;
* Allocate space for all possible cpus to avoid allocation at
* runtime
*/
- hctx->ctxs = kmalloc_node(nr_cpu_ids * sizeof(void *),
+ hctx->ctxs = kmalloc_array_node(nr_cpu_ids, sizeof(void *),
GFP_KERNEL, node);
if (!hctx->ctxs)
goto unregister_cpu_notifier;
hctx->nr_ctx = 0;
+ init_waitqueue_func_entry(&hctx->dispatch_wait, blk_mq_dispatch_wake);
+ INIT_LIST_HEAD(&hctx->dispatch_wait.entry);
+
if (set->ops->init_hctx &&
set->ops->init_hctx(hctx, set->driver_data, hctx_idx))
goto free_bitmap;
mutex_lock(&set->tag_list_lock);
- /* Check to see if we're transitioning to shared (from 1 to 2 queues). */
- if (!list_empty(&set->tag_list) && !(set->flags & BLK_MQ_F_TAG_SHARED)) {
+ /*
+ * Check to see if we're transitioning to shared (from 1 to 2 queues).
+ */
+ if (!list_empty(&set->tag_list) &&
+ !(set->flags & BLK_MQ_F_TAG_SHARED)) {
set->flags |= BLK_MQ_F_TAG_SHARED;
/* update existing queue */
blk_mq_update_tag_set_depth(set, true);
/*
* redo blk_mq_init_cpu_queues and blk_mq_init_hw_queues. FIXME: maybe
- * we should change hctx numa_node according to new topology (this
- * involves free and re-allocate memory, worthy doing?)
+ * we should change hctx numa_node according to the new topology (this
+ * involves freeing and re-allocating memory, worth doing?)
*/
-
blk_mq_map_swqueue(q);
blk_mq_sysfs_register(q);
projects / linux-2.6-block.git / blobdiff