diff options
-rw-r--r-- | block/blk-core.c | 3 | ||||
-rw-r--r-- | block/blk-flush.c | 2 | ||||
-rw-r--r-- | block/blk-iocost.c | 4 | ||||
-rw-r--r-- | block/blk-mq.c | 56 | ||||
-rw-r--r-- | block/mq-deadline.c | 2 | ||||
-rw-r--r-- | drivers/block/loop.c | 40 | ||||
-rw-r--r-- | drivers/block/ublk_drv.c | 11 | ||||
-rw-r--r-- | drivers/s390/block/dasd.c | 125 | ||||
-rw-r--r-- | drivers/s390/block/dasd_3990_erp.c | 4 | ||||
-rw-r--r-- | drivers/s390/block/dasd_ioctl.c | 1 | ||||
-rw-r--r-- | include/linux/blk-mq.h | 8 | ||||
-rw-r--r-- | include/uapi/linux/blkzoned.h | 10 | ||||
-rw-r--r-- | lib/sbitmap.c | 15 |
13 files changed, 156 insertions, 125 deletions
diff --git a/block/blk-core.c b/block/blk-core.c index 99d8b9812b18..90de50082146 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1144,8 +1144,7 @@ void __blk_flush_plug(struct blk_plug *plug, bool from_schedule) { if (!list_empty(&plug->cb_list)) flush_plug_callbacks(plug, from_schedule); - if (!rq_list_empty(plug->mq_list)) - blk_mq_flush_plug_list(plug, from_schedule); + blk_mq_flush_plug_list(plug, from_schedule); /* * Unconditionally flush out cached requests, even if the unplug * event came from schedule. Since we know hold references to the diff --git a/block/blk-flush.c b/block/blk-flush.c index dba392cf22be..8220517c2d67 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -189,7 +189,7 @@ static void blk_flush_complete_seq(struct request *rq, case REQ_FSEQ_DATA: list_move_tail(&rq->flush.list, &fq->flush_data_in_flight); spin_lock(&q->requeue_lock); - list_add_tail(&rq->queuelist, &q->flush_list); + list_add(&rq->queuelist, &q->requeue_list); spin_unlock(&q->requeue_lock); blk_mq_kick_requeue_list(q); break; diff --git a/block/blk-iocost.c b/block/blk-iocost.c index 9dfcf540f400..dd64e2066f01 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -2516,6 +2516,10 @@ static void calc_vtime_cost_builtin(struct bio *bio, struct ioc_gq *iocg, u64 seek_pages = 0; u64 cost = 0; + /* Can't calculate cost for empty bio */ + if (!bio->bi_iter.bi_size) + goto out; + switch (bio_op(bio)) { case REQ_OP_READ: coef_seqio = ioc->params.lcoefs[LCOEF_RSEQIO]; diff --git a/block/blk-mq.c b/block/blk-mq.c index 5504719b970d..b04ff6f56926 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -328,8 +328,24 @@ void blk_rq_init(struct request_queue *q, struct request *rq) } EXPORT_SYMBOL(blk_rq_init); +/* Set start and alloc time when the allocated request is actually used */ +static inline void blk_mq_rq_time_init(struct request *rq, u64 alloc_time_ns) +{ + if (blk_mq_need_time_stamp(rq)) + rq->start_time_ns = ktime_get_ns(); + else + rq->start_time_ns = 0; + +#ifdef CONFIG_BLK_RQ_ALLOC_TIME + if (blk_queue_rq_alloc_time(rq->q)) + rq->alloc_time_ns = alloc_time_ns ?: rq->start_time_ns; + else + rq->alloc_time_ns = 0; +#endif +} + static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, - struct blk_mq_tags *tags, unsigned int tag, u64 alloc_time_ns) + struct blk_mq_tags *tags, unsigned int tag) { struct blk_mq_ctx *ctx = data->ctx; struct blk_mq_hw_ctx *hctx = data->hctx; @@ -356,14 +372,7 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, } rq->timeout = 0; - if (blk_mq_need_time_stamp(rq)) - rq->start_time_ns = ktime_get_ns(); - else - rq->start_time_ns = 0; rq->part = NULL; -#ifdef CONFIG_BLK_RQ_ALLOC_TIME - rq->alloc_time_ns = alloc_time_ns; -#endif rq->io_start_time_ns = 0; rq->stats_sectors = 0; rq->nr_phys_segments = 0; @@ -393,8 +402,7 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, } static inline struct request * -__blk_mq_alloc_requests_batch(struct blk_mq_alloc_data *data, - u64 alloc_time_ns) +__blk_mq_alloc_requests_batch(struct blk_mq_alloc_data *data) { unsigned int tag, tag_offset; struct blk_mq_tags *tags; @@ -413,7 +421,7 @@ __blk_mq_alloc_requests_batch(struct blk_mq_alloc_data *data, tag = tag_offset + i; prefetch(tags->static_rqs[tag]); tag_mask &= ~(1UL << i); - rq = blk_mq_rq_ctx_init(data, tags, tag, alloc_time_ns); + rq = blk_mq_rq_ctx_init(data, tags, tag); rq_list_add(data->cached_rq, rq); nr++; } @@ -474,9 +482,11 @@ retry: * Try batched alloc if we want more than 1 tag. */ if (data->nr_tags > 1) { - rq = __blk_mq_alloc_requests_batch(data, alloc_time_ns); - if (rq) + rq = __blk_mq_alloc_requests_batch(data); + if (rq) { + blk_mq_rq_time_init(rq, alloc_time_ns); return rq; + } data->nr_tags = 1; } @@ -499,8 +509,9 @@ retry: goto retry; } - return blk_mq_rq_ctx_init(data, blk_mq_tags_from_data(data), tag, - alloc_time_ns); + rq = blk_mq_rq_ctx_init(data, blk_mq_tags_from_data(data), tag); + blk_mq_rq_time_init(rq, alloc_time_ns); + return rq; } static struct request *blk_mq_rq_cache_fill(struct request_queue *q, @@ -555,6 +566,7 @@ static struct request *blk_mq_alloc_cached_request(struct request_queue *q, return NULL; plug->cached_rq = rq_list_next(rq); + blk_mq_rq_time_init(rq, 0); } rq->cmd_flags = opf; @@ -656,8 +668,8 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, tag = blk_mq_get_tag(&data); if (tag == BLK_MQ_NO_TAG) goto out_queue_exit; - rq = blk_mq_rq_ctx_init(&data, blk_mq_tags_from_data(&data), tag, - alloc_time_ns); + rq = blk_mq_rq_ctx_init(&data, blk_mq_tags_from_data(&data), tag); + blk_mq_rq_time_init(rq, alloc_time_ns); rq->__data_len = 0; rq->__sector = (sector_t) -1; rq->bio = rq->biotail = NULL; @@ -2742,7 +2754,14 @@ void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule) { struct request *rq; - if (rq_list_empty(plug->mq_list)) + /* + * We may have been called recursively midway through handling + * plug->mq_list via a schedule() in the driver's queue_rq() callback. + * To avoid mq_list changing under our feet, clear rq_count early and + * bail out specifically if rq_count is 0 rather than checking + * whether the mq_list is empty. + */ + if (plug->rq_count == 0) return; plug->rq_count = 0; @@ -2896,6 +2915,7 @@ static inline struct request *blk_mq_get_cached_request(struct request_queue *q, plug->cached_rq = rq_list_next(rq); rq_qos_throttle(q, *bio); + blk_mq_rq_time_init(rq, 0); rq->cmd_flags = (*bio)->bi_opf; INIT_LIST_HEAD(&rq->queuelist); return rq; diff --git a/block/mq-deadline.c b/block/mq-deadline.c index 6aa5daf7ae32..02a916ba62ee 100644 --- a/block/mq-deadline.c +++ b/block/mq-deadline.c @@ -176,7 +176,7 @@ static inline struct request *deadline_from_pos(struct dd_per_prio *per_prio, * zoned writes, start searching from the start of a zone. */ if (blk_rq_is_seq_zoned_write(rq)) - pos -= round_down(pos, rq->q->limits.chunk_sectors); + pos = round_down(pos, rq->q->limits.chunk_sectors); while (node) { rq = rb_entry_rq(node); diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 37511d2b2caf..637c5bda2387 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1775,14 +1775,43 @@ static const struct block_device_operations lo_fops = { /* * If max_loop is specified, create that many devices upfront. * This also becomes a hard limit. If max_loop is not specified, + * the default isn't a hard limit (as before commit 85c50197716c + * changed the default value from 0 for max_loop=0 reasons), just * create CONFIG_BLK_DEV_LOOP_MIN_COUNT loop devices at module * init time. Loop devices can be requested on-demand with the * /dev/loop-control interface, or be instantiated by accessing * a 'dead' device node. */ static int max_loop = CONFIG_BLK_DEV_LOOP_MIN_COUNT; -module_param(max_loop, int, 0444); + +#ifdef CONFIG_BLOCK_LEGACY_AUTOLOAD +static bool max_loop_specified; + +static int max_loop_param_set_int(const char *val, + const struct kernel_param *kp) +{ + int ret; + + ret = param_set_int(val, kp); + if (ret < 0) + return ret; + + max_loop_specified = true; + return 0; +} + +static const struct kernel_param_ops max_loop_param_ops = { + .set = max_loop_param_set_int, + .get = param_get_int, +}; + +module_param_cb(max_loop, &max_loop_param_ops, &max_loop, 0444); MODULE_PARM_DESC(max_loop, "Maximum number of loop devices"); +#else +module_param(max_loop, int, 0444); +MODULE_PARM_DESC(max_loop, "Initial number of loop devices"); +#endif + module_param(max_part, int, 0444); MODULE_PARM_DESC(max_part, "Maximum number of partitions per loop device"); @@ -2093,14 +2122,18 @@ static void loop_remove(struct loop_device *lo) put_disk(lo->lo_disk); } +#ifdef CONFIG_BLOCK_LEGACY_AUTOLOAD static void loop_probe(dev_t dev) { int idx = MINOR(dev) >> part_shift; - if (max_loop && idx >= max_loop) + if (max_loop_specified && max_loop && idx >= max_loop) return; loop_add(idx); } +#else +#define loop_probe NULL +#endif /* !CONFIG_BLOCK_LEGACY_AUTOLOAD */ static int loop_control_remove(int idx) { @@ -2281,6 +2314,9 @@ module_exit(loop_exit); static int __init max_loop_setup(char *str) { max_loop = simple_strtol(str, NULL, 0); +#ifdef CONFIG_BLOCK_LEGACY_AUTOLOAD + max_loop_specified = true; +#endif return 1; } diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index 1c823750c95a..21d2e71c5514 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -1847,7 +1847,8 @@ static int ublk_ctrl_start_dev(struct ublk_device *ub, struct io_uring_cmd *cmd) if (ublksrv_pid <= 0) return -EINVAL; - wait_for_completion_interruptible(&ub->completion); + if (wait_for_completion_interruptible(&ub->completion) != 0) + return -EINTR; schedule_delayed_work(&ub->monitor_work, UBLK_DAEMON_MONITOR_PERIOD); @@ -2125,8 +2126,8 @@ static int ublk_ctrl_del_dev(struct ublk_device **p_ub) * - the device number is freed already, we will not find this * device via ublk_get_device_from_id() */ - wait_event_interruptible(ublk_idr_wq, ublk_idr_freed(idx)); - + if (wait_event_interruptible(ublk_idr_wq, ublk_idr_freed(idx))) + return -EINTR; return 0; } @@ -2323,7 +2324,9 @@ static int ublk_ctrl_end_recovery(struct ublk_device *ub, pr_devel("%s: Waiting for new ubq_daemons(nr: %d) are ready, dev id %d...\n", __func__, ub->dev_info.nr_hw_queues, header->dev_id); /* wait until new ubq_daemon sending all FETCH_REQ */ - wait_for_completion_interruptible(&ub->completion); + if (wait_for_completion_interruptible(&ub->completion)) + return -EINTR; + pr_devel("%s: All new ubq_daemons(nr: %d) are ready, dev id %d\n", __func__, ub->dev_info.nr_hw_queues, header->dev_id); diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index edcbf77852c3..50a5ff70814a 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -2943,41 +2943,32 @@ static void _dasd_wake_block_flush_cb(struct dasd_ccw_req *cqr, void *data) * Requeue a request back to the block request queue * only works for block requests */ -static int _dasd_requeue_request(struct dasd_ccw_req *cqr) +static void _dasd_requeue_request(struct dasd_ccw_req *cqr) { - struct dasd_block *block = cqr->block; struct request *req; - if (!block) - return -EINVAL; /* * If the request is an ERP request there is nothing to requeue. * This will be done with the remaining original request. */ if (cqr->refers) - return 0; + return; spin_lock_irq(&cqr->dq->lock); req = (struct request *) cqr->callback_data; blk_mq_requeue_request(req, true); spin_unlock_irq(&cqr->dq->lock); - return 0; + return; } -/* - * Go through all request on the dasd_block request queue, cancel them - * on the respective dasd_device, and return them to the generic - * block layer. - */ -static int dasd_flush_block_queue(struct dasd_block *block) +static int _dasd_requests_to_flushqueue(struct dasd_block *block, + struct list_head *flush_queue) { struct dasd_ccw_req *cqr, *n; - int rc, i; - struct list_head flush_queue; unsigned long flags; + int rc, i; - INIT_LIST_HEAD(&flush_queue); - spin_lock_bh(&block->queue_lock); + spin_lock_irqsave(&block->queue_lock, flags); rc = 0; restart: list_for_each_entry_safe(cqr, n, &block->ccw_queue, blocklist) { @@ -2992,13 +2983,32 @@ restart: * is returned from the dasd_device layer. */ cqr->callback = _dasd_wake_block_flush_cb; - for (i = 0; cqr != NULL; cqr = cqr->refers, i++) - list_move_tail(&cqr->blocklist, &flush_queue); + for (i = 0; cqr; cqr = cqr->refers, i++) + list_move_tail(&cqr->blocklist, flush_queue); if (i > 1) /* moved more than one request - need to restart */ goto restart; } - spin_unlock_bh(&block->queue_lock); + spin_unlock_irqrestore(&block->queue_lock, flags); + + return rc; +} + +/* + * Go through all request on the dasd_block request queue, cancel them + * on the respective dasd_device, and return them to the generic + * block layer. + */ +static int dasd_flush_block_queue(struct dasd_block *block) +{ + struct dasd_ccw_req *cqr, *n; + struct list_head flush_queue; + unsigned long flags; + int rc; + + INIT_LIST_HEAD(&flush_queue); + rc = _dasd_requests_to_flushqueue(block, &flush_queue); + /* Now call the callback function of flushed requests */ restart_cb: list_for_each_entry_safe(cqr, n, &flush_queue, blocklist) { @@ -3881,75 +3891,36 @@ EXPORT_SYMBOL_GPL(dasd_generic_space_avail); */ int dasd_generic_requeue_all_requests(struct dasd_device *device) { + struct dasd_block *block = device->block; struct list_head requeue_queue; struct dasd_ccw_req *cqr, *n; - struct dasd_ccw_req *refers; int rc; - INIT_LIST_HEAD(&requeue_queue); - spin_lock_irq(get_ccwdev_lock(device->cdev)); - rc = 0; - list_for_each_entry_safe(cqr, n, &device->ccw_queue, devlist) { - /* Check status and move request to flush_queue */ - if (cqr->status == DASD_CQR_IN_IO) { - rc = device->discipline->term_IO(cqr); - if (rc) { - /* unable to terminate requeust */ - dev_err(&device->cdev->dev, - "Unable to terminate request %p " - "on suspend\n", cqr); - spin_unlock_irq(get_ccwdev_lock(device->cdev)); - dasd_put_device(device); - return rc; - } - } - list_move_tail(&cqr->devlist, &requeue_queue); - } - spin_unlock_irq(get_ccwdev_lock(device->cdev)); - - list_for_each_entry_safe(cqr, n, &requeue_queue, devlist) { - wait_event(dasd_flush_wq, - (cqr->status != DASD_CQR_CLEAR_PENDING)); + if (!block) + return 0; - /* - * requeue requests to blocklayer will only work - * for block device requests - */ - if (_dasd_requeue_request(cqr)) - continue; + INIT_LIST_HEAD(&requeue_queue); + rc = _dasd_requests_to_flushqueue(block, &requeue_queue); - /* remove requests from device and block queue */ - list_del_init(&cqr->devlist); - while (cqr->refers != NULL) { - refers = cqr->refers; - /* remove the request from the block queue */ - list_del(&cqr->blocklist); - /* free the finished erp request */ - dasd_free_erp_request(cqr, cqr->memdev); - cqr = refers; + /* Now call the callback function of flushed requests */ +restart_cb: + list_for_each_entry_safe(cqr, n, &requeue_queue, blocklist) { + wait_event(dasd_flush_wq, (cqr->status < DASD_CQR_QUEUED)); + /* Process finished ERP request. */ + if (cqr->refers) { + spin_lock_bh(&block->queue_lock); + __dasd_process_erp(block->base, cqr); + spin_unlock_bh(&block->queue_lock); + /* restart list_for_xx loop since dasd_process_erp + * might remove multiple elements + */ + goto restart_cb; } - - /* - * _dasd_requeue_request already checked for a valid - * blockdevice, no need to check again - * all erp requests (cqr->refers) have a cqr->block - * pointer copy from the original cqr - */ + _dasd_requeue_request(cqr); list_del_init(&cqr->blocklist); cqr->block->base->discipline->free_cp( cqr, (struct request *) cqr->callback_data); } - - /* - * if requests remain then they are internal request - * and go back to the device queue - */ - if (!list_empty(&requeue_queue)) { - /* move freeze_queue to start of the ccw_queue */ - spin_lock_irq(get_ccwdev_lock(device->cdev)); - list_splice_tail(&requeue_queue, &device->ccw_queue); - spin_unlock_irq(get_ccwdev_lock(device->cdev)); - } dasd_schedule_device_bh(device); return rc; } diff --git a/drivers/s390/block/dasd_3990_erp.c b/drivers/s390/block/dasd_3990_erp.c index 9fd36c468706..89957bb7244d 100644 --- a/drivers/s390/block/dasd_3990_erp.c +++ b/drivers/s390/block/dasd_3990_erp.c @@ -1050,7 +1050,7 @@ dasd_3990_erp_com_rej(struct dasd_ccw_req * erp, char *sense) dev_err(&device->cdev->dev, "An I/O request was rejected" " because writing is inhibited\n"); erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED); - } else if (sense[7] & SNS7_INVALID_ON_SEC) { + } else if (sense[7] == SNS7_INVALID_ON_SEC) { dev_err(&device->cdev->dev, "An I/O request was rejected on a copy pair secondary device\n"); /* suppress dump of sense data for this error */ set_bit(DASD_CQR_SUPPRESS_CR, &erp->refers->flags); @@ -2441,7 +2441,7 @@ static struct dasd_ccw_req *dasd_3990_erp_add_erp(struct dasd_ccw_req *cqr) erp->block = cqr->block; erp->magic = cqr->magic; erp->expires = cqr->expires; - erp->retries = 256; + erp->retries = device->default_retries; erp->buildclk = get_tod_clock(); erp->status = DASD_CQR_FILLED; diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c index 513a7e6eee63..d55862605b82 100644 --- a/drivers/s390/block/dasd_ioctl.c +++ b/drivers/s390/block/dasd_ioctl.c @@ -131,6 +131,7 @@ static int dasd_ioctl_resume(struct dasd_block *block) spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags); dasd_schedule_block_bh(block); + dasd_schedule_device_bh(base); return 0; } diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 2b7fb8e87793..495ca198775f 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -158,13 +158,13 @@ struct request { /* * The rb_node is only used inside the io scheduler, requests - * are pruned when moved to the dispatch queue. So let the - * completion_data share space with the rb_node. + * are pruned when moved to the dispatch queue. special_vec must + * only be used if RQF_SPECIAL_PAYLOAD is set, and those cannot be + * insert into an IO scheduler. */ union { struct rb_node rb_node; /* sort/lookup */ struct bio_vec special_vec; - void *completion_data; }; /* @@ -397,8 +397,6 @@ struct blk_mq_hw_ctx { */ struct blk_mq_tags *sched_tags; - /** @queued: Number of queued requests. */ - unsigned long queued; /** @run: Number of dispatched requests. */ unsigned long run; diff --git a/include/uapi/linux/blkzoned.h b/include/uapi/linux/blkzoned.h index b80fcc9ea525..f85743ef6e7d 100644 --- a/include/uapi/linux/blkzoned.h +++ b/include/uapi/linux/blkzoned.h @@ -51,13 +51,13 @@ enum blk_zone_type { * * The Zone Condition state machine in the ZBC/ZAC standards maps the above * deinitions as: - * - ZC1: Empty | BLK_ZONE_EMPTY + * - ZC1: Empty | BLK_ZONE_COND_EMPTY * - ZC2: Implicit Open | BLK_ZONE_COND_IMP_OPEN * - ZC3: Explicit Open | BLK_ZONE_COND_EXP_OPEN - * - ZC4: Closed | BLK_ZONE_CLOSED - * - ZC5: Full | BLK_ZONE_FULL - * - ZC6: Read Only | BLK_ZONE_READONLY - * - ZC7: Offline | BLK_ZONE_OFFLINE + * - ZC4: Closed | BLK_ZONE_COND_CLOSED + * - ZC5: Full | BLK_ZONE_COND_FULL + * - ZC6: Read Only | BLK_ZONE_COND_READONLY + * - ZC7: Offline | BLK_ZONE_COND_OFFLINE * * Conditions 0x5 to 0xC are reserved by the current ZBC/ZAC spec and should * be considered invalid. diff --git a/lib/sbitmap.c b/lib/sbitmap.c index eff4e42c425a..d0a5081dfd12 100644 --- a/lib/sbitmap.c +++ b/lib/sbitmap.c @@ -550,7 +550,7 @@ EXPORT_SYMBOL_GPL(sbitmap_queue_min_shallow_depth); static void __sbitmap_queue_wake_up(struct sbitmap_queue *sbq, int nr) { - int i, wake_index; + int i, wake_index, woken; if (!atomic_read(&sbq->ws_active)) return; @@ -567,13 +567,12 @@ static void __sbitmap_queue_wake_up(struct sbitmap_queue *sbq, int nr) */ wake_index = sbq_index_inc(wake_index); - /* - * It is sufficient to wake up at least one waiter to - * guarantee forward progress. - */ - if (waitqueue_active(&ws->wait) && - wake_up_nr(&ws->wait, nr)) - break; + if (waitqueue_active(&ws->wait)) { + woken = wake_up_nr(&ws->wait, nr); + if (woken == nr) + break; + nr -= woken; + } } if (wake_index != atomic_read(&sbq->wake_index)) |