#include <linux/sched/sysctl.h>
#include <linux/delay.h>
#include <linux/crash_dump.h>
+#include <linux/prefetch.h>
#include <trace/events/block.h>
static DEFINE_MUTEX(all_q_mutex);
static LIST_HEAD(all_q_list);
-static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx);
-
/*
* Check if any of the ctx's have pending work in this hardware queue
*/
static bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
{
- unsigned int i;
-
- for (i = 0; i < hctx->ctx_map.size; i++)
- if (hctx->ctx_map.map[i].word)
- return true;
-
- return false;
-}
-
-static inline struct blk_align_bitmap *get_bm(struct blk_mq_hw_ctx *hctx,
- struct blk_mq_ctx *ctx)
-{
- return &hctx->ctx_map.map[ctx->index_hw / hctx->ctx_map.bits_per_word];
+ return sbitmap_any_bit_set(&hctx->ctx_map);
}
-#define CTX_TO_BIT(hctx, ctx) \
- ((ctx)->index_hw & ((hctx)->ctx_map.bits_per_word - 1))
-
/*
* Mark this ctx as having pending work in this hardware queue
*/
static void blk_mq_hctx_mark_pending(struct blk_mq_hw_ctx *hctx,
struct blk_mq_ctx *ctx)
{
- struct blk_align_bitmap *bm = get_bm(hctx, ctx);
-
- if (!test_bit(CTX_TO_BIT(hctx, ctx), &bm->word))
- set_bit(CTX_TO_BIT(hctx, ctx), &bm->word);
+ if (!sbitmap_test_bit(&hctx->ctx_map, ctx->index_hw))
+ sbitmap_set_bit(&hctx->ctx_map, ctx->index_hw);
}
static void blk_mq_hctx_clear_pending(struct blk_mq_hw_ctx *hctx,
struct blk_mq_ctx *ctx)
{
- struct blk_align_bitmap *bm = get_bm(hctx, ctx);
-
- clear_bit(CTX_TO_BIT(hctx, ctx), &bm->word);
+ sbitmap_clear_bit(&hctx->ctx_map, ctx->index_hw);
}
void blk_mq_freeze_queue_start(struct request_queue *q)
return ERR_PTR(ret);
ctx = blk_mq_get_ctx(q);
- hctx = q->mq_ops->map_queue(q, ctx->cpu);
+ hctx = blk_mq_map_queue(q, ctx->cpu);
blk_mq_set_alloc_data(&alloc_data, q, flags, ctx, hctx);
-
rq = __blk_mq_alloc_request(&alloc_data, rw, 0);
- if (!rq && !(flags & BLK_MQ_REQ_NOWAIT)) {
- __blk_mq_run_hw_queue(hctx);
- blk_mq_put_ctx(ctx);
-
- ctx = blk_mq_get_ctx(q);
- hctx = q->mq_ops->map_queue(q, ctx->cpu);
- blk_mq_set_alloc_data(&alloc_data, q, flags, ctx, hctx);
- rq = __blk_mq_alloc_request(&alloc_data, rw, 0);
- ctx = alloc_data.ctx;
- }
blk_mq_put_ctx(ctx);
+
if (!rq) {
blk_queue_exit(q);
return ERR_PTR(-EWOULDBLOCK);
rq->cmd_flags = 0;
clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
- blk_mq_put_tag(hctx, tag, &ctx->last_tag);
+ blk_mq_put_tag(hctx, ctx, tag);
blk_queue_exit(q);
}
void blk_mq_free_request(struct request *rq)
{
- struct blk_mq_hw_ctx *hctx;
- struct request_queue *q = rq->q;
-
- hctx = q->mq_ops->map_queue(q, rq->mq_ctx->cpu);
- blk_mq_free_hctx_request(hctx, rq);
+ blk_mq_free_hctx_request(blk_mq_map_queue(rq->q, rq->mq_ctx->cpu), rq);
}
EXPORT_SYMBOL_GPL(blk_mq_free_request);
static void blk_mq_requeue_work(struct work_struct *work)
{
struct request_queue *q =
- container_of(work, struct request_queue, requeue_work);
+ container_of(work, struct request_queue, requeue_work.work);
LIST_HEAD(rq_list);
struct request *rq, *next;
unsigned long flags;
void blk_mq_cancel_requeue_work(struct request_queue *q)
{
- cancel_work_sync(&q->requeue_work);
+ cancel_delayed_work_sync(&q->requeue_work);
}
EXPORT_SYMBOL_GPL(blk_mq_cancel_requeue_work);
void blk_mq_kick_requeue_list(struct request_queue *q)
{
- kblockd_schedule_work(&q->requeue_work);
+ kblockd_schedule_delayed_work(&q->requeue_work, 0);
}
EXPORT_SYMBOL(blk_mq_kick_requeue_list);
+void blk_mq_delay_kick_requeue_list(struct request_queue *q,
+ unsigned long msecs)
+{
+ kblockd_schedule_delayed_work(&q->requeue_work,
+ msecs_to_jiffies(msecs));
+}
+EXPORT_SYMBOL(blk_mq_delay_kick_requeue_list);
+
void blk_mq_abort_requeue_list(struct request_queue *q)
{
unsigned long flags;
struct request *blk_mq_tag_to_rq(struct blk_mq_tags *tags, unsigned int tag)
{
- if (tag < tags->nr_tags)
+ if (tag < tags->nr_tags) {
+ prefetch(tags->rqs[tag]);
return tags->rqs[tag];
+ }
return NULL;
}
return false;
}
+struct flush_busy_ctx_data {
+ struct blk_mq_hw_ctx *hctx;
+ struct list_head *list;
+};
+
+static bool flush_busy_ctx(struct sbitmap *sb, unsigned int bitnr, void *data)
+{
+ struct flush_busy_ctx_data *flush_data = data;
+ struct blk_mq_hw_ctx *hctx = flush_data->hctx;
+ struct blk_mq_ctx *ctx = hctx->ctxs[bitnr];
+
+ sbitmap_clear_bit(sb, bitnr);
+ spin_lock(&ctx->lock);
+ list_splice_tail_init(&ctx->rq_list, flush_data->list);
+ spin_unlock(&ctx->lock);
+ return true;
+}
+
/*
* Process software queues that have been marked busy, splicing them
* to the for-dispatch
*/
static void flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list)
{
- struct blk_mq_ctx *ctx;
- int i;
-
- for (i = 0; i < hctx->ctx_map.size; i++) {
- struct blk_align_bitmap *bm = &hctx->ctx_map.map[i];
- unsigned int off, bit;
-
- if (!bm->word)
- continue;
+ struct flush_busy_ctx_data data = {
+ .hctx = hctx,
+ .list = list,
+ };
- bit = 0;
- off = i * hctx->ctx_map.bits_per_word;
- do {
- bit = find_next_bit(&bm->word, bm->depth, bit);
- if (bit >= bm->depth)
- break;
+ sbitmap_for_each_set(&hctx->ctx_map, flush_busy_ctx, &data);
+}
- ctx = hctx->ctxs[bit + off];
- clear_bit(bit, &bm->word);
- spin_lock(&ctx->lock);
- list_splice_tail_init(&ctx->rq_list, list);
- spin_unlock(&ctx->lock);
+static inline unsigned int queued_to_index(unsigned int queued)
+{
+ if (!queued)
+ return 0;
- bit++;
- } while (1);
- }
+ return min(BLK_MQ_MAX_DISPATCH_ORDER - 1, ilog2(queued) + 1);
}
/*
dptr = &driver_list;
}
- if (!queued)
- hctx->dispatched[0]++;
- else if (queued < (1 << (BLK_MQ_MAX_DISPATCH_ORDER - 1)))
- hctx->dispatched[ilog2(queued) + 1]++;
+ hctx->dispatched[queued_to_index(queued)]++;
/*
* Any items that need requeuing? Stuff them into hctx->dispatch,
!blk_mq_hw_queue_mapped(hctx)))
return;
- if (!async) {
+ if (!async && !(hctx->flags & BLK_MQ_F_BLOCKING)) {
int cpu = get_cpu();
if (cpumask_test_cpu(cpu, hctx->cpumask)) {
__blk_mq_run_hw_queue(hctx);
put_cpu();
}
- kblockd_schedule_delayed_work_on(blk_mq_hctx_next_cpu(hctx),
- &hctx->run_work, 0);
+ kblockd_schedule_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work);
}
void blk_mq_run_hw_queues(struct request_queue *q, bool async)
void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx)
{
- cancel_delayed_work(&hctx->run_work);
+ cancel_work(&hctx->run_work);
cancel_delayed_work(&hctx->delay_work);
set_bit(BLK_MQ_S_STOPPED, &hctx->state);
}
{
struct blk_mq_hw_ctx *hctx;
- hctx = container_of(work, struct blk_mq_hw_ctx, run_work.work);
+ hctx = container_of(work, struct blk_mq_hw_ctx, run_work);
__blk_mq_run_hw_queue(hctx);
}
{
struct blk_mq_ctx *ctx = rq->mq_ctx;
struct request_queue *q = rq->q;
- struct blk_mq_hw_ctx *hctx;
-
- hctx = q->mq_ops->map_queue(q, ctx->cpu);
+ struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, ctx->cpu);
spin_lock(&ctx->lock);
__blk_mq_insert_request(hctx, rq, at_head);
bool from_schedule)
{
- struct blk_mq_hw_ctx *hctx;
+ struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, ctx->cpu);
trace_block_unplug(q, depth, !from_schedule);
- hctx = q->mq_ops->map_queue(q, ctx->cpu);
-
/*
* preemption doesn't flush plug list, so it's possible ctx->cpu is
* offline now
blk_queue_enter_live(q);
ctx = blk_mq_get_ctx(q);
- hctx = q->mq_ops->map_queue(q, ctx->cpu);
+ hctx = blk_mq_map_queue(q, ctx->cpu);
if (rw_is_sync(bio_op(bio), bio->bi_opf))
op_flags |= REQ_SYNC;
trace_block_getrq(q, bio, op);
- blk_mq_set_alloc_data(&alloc_data, q, BLK_MQ_REQ_NOWAIT, ctx, hctx);
+ blk_mq_set_alloc_data(&alloc_data, q, 0, ctx, hctx);
rq = __blk_mq_alloc_request(&alloc_data, op, op_flags);
- if (unlikely(!rq)) {
- __blk_mq_run_hw_queue(hctx);
- blk_mq_put_ctx(ctx);
- trace_block_sleeprq(q, bio, op);
-
- ctx = blk_mq_get_ctx(q);
- hctx = q->mq_ops->map_queue(q, ctx->cpu);
- blk_mq_set_alloc_data(&alloc_data, q, 0, ctx, hctx);
- rq = __blk_mq_alloc_request(&alloc_data, op, op_flags);
- ctx = alloc_data.ctx;
- hctx = alloc_data.hctx;
- }
hctx->queued++;
data->hctx = hctx;
{
int ret;
struct request_queue *q = rq->q;
- struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q,
- rq->mq_ctx->cpu);
+ struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, rq->mq_ctx->cpu);
struct blk_mq_queue_data bd = {
.rq = rq,
.list = NULL,
return cookie;
}
-/*
- * Default mapping to a software queue, since we use one per CPU.
- */
-struct blk_mq_hw_ctx *blk_mq_map_queue(struct request_queue *q, const int cpu)
-{
- return q->queue_hw_ctx[q->mq_map[cpu]];
-}
-EXPORT_SYMBOL(blk_mq_map_queue);
-
static void blk_mq_free_rq_map(struct blk_mq_tag_set *set,
struct blk_mq_tags *tags, unsigned int hctx_idx)
{
return NULL;
}
-static void blk_mq_free_bitmap(struct blk_mq_ctxmap *bitmap)
-{
- kfree(bitmap->map);
-}
-
-static int blk_mq_alloc_bitmap(struct blk_mq_ctxmap *bitmap, int node)
-{
- unsigned int bpw = 8, total, num_maps, i;
-
- bitmap->bits_per_word = bpw;
-
- num_maps = ALIGN(nr_cpu_ids, bpw) / bpw;
- bitmap->map = kzalloc_node(num_maps * sizeof(struct blk_align_bitmap),
- GFP_KERNEL, node);
- if (!bitmap->map)
- return -ENOMEM;
-
- total = nr_cpu_ids;
- for (i = 0; i < num_maps; i++) {
- bitmap->map[i].depth = min(total, bitmap->bits_per_word);
- total -= bitmap->map[i].depth;
- }
-
- return 0;
-}
-
/*
* 'cpu' is going away. splice any existing rq_list entries from this
* software queue to the hw queue dispatch list, and ensure that it
blk_mq_unregister_cpu_notifier(&hctx->cpu_notifier);
blk_free_flush_queue(hctx->fq);
- blk_mq_free_bitmap(&hctx->ctx_map);
+ sbitmap_free(&hctx->ctx_map);
}
static void blk_mq_exit_hw_queues(struct request_queue *q,
if (node == NUMA_NO_NODE)
node = hctx->numa_node = set->numa_node;
- INIT_DELAYED_WORK(&hctx->run_work, blk_mq_run_work_fn);
+ INIT_WORK(&hctx->run_work, blk_mq_run_work_fn);
INIT_DELAYED_WORK(&hctx->delay_work, blk_mq_delay_work_fn);
spin_lock_init(&hctx->lock);
INIT_LIST_HEAD(&hctx->dispatch);
if (!hctx->ctxs)
goto unregister_cpu_notifier;
- if (blk_mq_alloc_bitmap(&hctx->ctx_map, node))
+ if (sbitmap_init_node(&hctx->ctx_map, nr_cpu_ids, ilog2(8), GFP_KERNEL,
+ node))
goto free_ctxs;
hctx->nr_ctx = 0;
if (set->ops->exit_hctx)
set->ops->exit_hctx(hctx, hctx_idx);
free_bitmap:
- blk_mq_free_bitmap(&hctx->ctx_map);
+ sbitmap_free(&hctx->ctx_map);
free_ctxs:
kfree(hctx->ctxs);
unregister_cpu_notifier:
if (!cpu_online(i))
continue;
- hctx = q->mq_ops->map_queue(q, i);
+ hctx = blk_mq_map_queue(q, i);
/*
* Set local node, IFF we have more than one hw queue. If
continue;
ctx = per_cpu_ptr(q->queue_ctx, i);
- hctx = q->mq_ops->map_queue(q, i);
+ hctx = blk_mq_map_queue(q, i);
cpumask_set_cpu(i, hctx->cpumask);
ctx->index_hw = hctx->nr_ctx;
mutex_unlock(&q->sysfs_lock);
queue_for_each_hw_ctx(q, hctx, i) {
- struct blk_mq_ctxmap *map = &hctx->ctx_map;
-
/*
* If no software queues are mapped to this hardware queue,
* disable it and free the request entries.
hctx->tags = set->tags[i];
WARN_ON(!hctx->tags);
- cpumask_copy(hctx->tags->cpumask, hctx->cpumask);
/*
* Set the map size to the number of mapped software queues.
* This is more accurate and more efficient than looping
* over all possibly mapped software queues.
*/
- map->size = DIV_ROUND_UP(hctx->nr_ctx, map->bits_per_word);
+ sbitmap_resize(&hctx->ctx_map, hctx->nr_ctx);
/*
* Initialize batch roundrobin counts
kfree(hctx);
}
- kfree(q->mq_map);
q->mq_map = NULL;
kfree(q->queue_hw_ctx);
if (!q->queue_hw_ctx)
goto err_percpu;
- q->mq_map = blk_mq_make_queue_map(set);
- if (!q->mq_map)
- goto err_map;
+ q->mq_map = set->mq_map;
blk_mq_realloc_hw_ctxs(set, q);
if (!q->nr_hw_queues)
q->sg_reserved_size = INT_MAX;
- INIT_WORK(&q->requeue_work, blk_mq_requeue_work);
+ INIT_DELAYED_WORK(&q->requeue_work, blk_mq_requeue_work);
INIT_LIST_HEAD(&q->requeue_list);
spin_lock_init(&q->requeue_lock);
return q;
err_hctxs:
- kfree(q->mq_map);
-err_map:
kfree(q->queue_hw_ctx);
err_percpu:
free_percpu(q->queue_ctx);
blk_mq_sysfs_unregister(q);
- blk_mq_update_queue_map(q->mq_map, q->nr_hw_queues, online_mask);
-
/*
* 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
return 0;
}
-struct cpumask *blk_mq_tags_cpumask(struct blk_mq_tags *tags)
-{
- return tags->cpumask;
-}
-EXPORT_SYMBOL_GPL(blk_mq_tags_cpumask);
-
/*
* Alloc a tag set to be associated with one or more request queues.
* May fail with EINVAL for various error conditions. May adjust the
*/
int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
{
+ int ret;
+
BUILD_BUG_ON(BLK_MQ_MAX_DEPTH > 1 << BLK_MQ_UNIQUE_TAG_BITS);
if (!set->nr_hw_queues)
if (set->queue_depth < set->reserved_tags + BLK_MQ_TAG_MIN)
return -EINVAL;
- if (!set->ops->queue_rq || !set->ops->map_queue)
+ if (!set->ops->queue_rq)
return -EINVAL;
if (set->queue_depth > BLK_MQ_MAX_DEPTH) {
if (!set->tags)
return -ENOMEM;
- if (blk_mq_alloc_rq_maps(set))
- goto enomem;
+ ret = -ENOMEM;
+ set->mq_map = kzalloc_node(sizeof(*set->mq_map) * nr_cpu_ids,
+ GFP_KERNEL, set->numa_node);
+ if (!set->mq_map)
+ goto out_free_tags;
+
+ if (set->ops->map_queues)
+ ret = set->ops->map_queues(set);
+ else
+ ret = blk_mq_map_queues(set);
+ if (ret)
+ goto out_free_mq_map;
+
+ ret = blk_mq_alloc_rq_maps(set);
+ if (ret)
+ goto out_free_mq_map;
mutex_init(&set->tag_list_lock);
INIT_LIST_HEAD(&set->tag_list);
return 0;
-enomem:
+
+out_free_mq_map:
+ kfree(set->mq_map);
+ set->mq_map = NULL;
+out_free_tags:
kfree(set->tags);
set->tags = NULL;
- return -ENOMEM;
+ return ret;
}
EXPORT_SYMBOL(blk_mq_alloc_tag_set);
blk_mq_free_rq_map(set, set->tags[i], i);
}
+ kfree(set->mq_map);
+ set->mq_map = NULL;
+
kfree(set->tags);
set->tags = NULL;
}
projects / cascardo / linux.git / blobdiff