diff options
Diffstat (limited to 'block')
| -rw-r--r-- | block/Makefile | 4 | ||||
| -rw-r--r-- | block/as-iosched.c | 40 | ||||
| -rw-r--r-- | block/blk-barrier.c | 319 | ||||
| -rw-r--r-- | block/blk-core.c | 2034 | ||||
| -rw-r--r-- | block/blk-exec.c | 105 | ||||
| -rw-r--r-- | block/blk-ioc.c | 194 | ||||
| -rw-r--r-- | block/blk-map.c | 264 | ||||
| -rw-r--r-- | block/blk-merge.c | 485 | ||||
| -rw-r--r-- | block/blk-settings.c | 402 | ||||
| -rw-r--r-- | block/blk-sysfs.c | 309 | ||||
| -rw-r--r-- | block/blk-tag.c | 396 | ||||
| -rw-r--r-- | block/blk.h | 53 | ||||
| -rw-r--r-- | block/blktrace.c | 24 | ||||
| -rw-r--r-- | block/bsg.c | 15 | ||||
| -rw-r--r-- | block/cfq-iosched.c | 425 | ||||
| -rw-r--r-- | block/compat_ioctl.c | 5 | ||||
| -rw-r--r-- | block/elevator.c | 34 | ||||
| -rw-r--r-- | block/genhd.c | 423 | ||||
| -rw-r--r-- | block/ll_rw_blk.c | 4214 |
19 files changed, 5047 insertions, 4698 deletions
diff --git a/block/Makefile b/block/Makefile index 826108190f0..5a43c7d7959 100644 --- a/block/Makefile +++ b/block/Makefile @@ -2,7 +2,9 @@ # Makefile for the kernel block layer # -obj-$(CONFIG_BLOCK) := elevator.o ll_rw_blk.o ioctl.o genhd.o scsi_ioctl.o +obj-$(CONFIG_BLOCK) := elevator.o blk-core.o blk-tag.o blk-sysfs.o \ + blk-barrier.o blk-settings.o blk-ioc.o blk-map.o \ + blk-exec.o blk-merge.o ioctl.o genhd.o scsi_ioctl.o obj-$(CONFIG_BLK_DEV_BSG) += bsg.o obj-$(CONFIG_IOSCHED_NOOP) += noop-iosched.o diff --git a/block/as-iosched.c b/block/as-iosched.c index cb5e53b05c7..96036846a00 100644 --- a/block/as-iosched.c +++ b/block/as-iosched.c @@ -170,9 +170,11 @@ static void free_as_io_context(struct as_io_context *aic) static void as_trim(struct io_context *ioc) { + spin_lock(&ioc->lock); if (ioc->aic) free_as_io_context(ioc->aic); ioc->aic = NULL; + spin_unlock(&ioc->lock); } /* Called when the task exits */ @@ -462,7 +464,9 @@ static void as_antic_timeout(unsigned long data) spin_lock_irqsave(q->queue_lock, flags); if (ad->antic_status == ANTIC_WAIT_REQ || ad->antic_status == ANTIC_WAIT_NEXT) { - struct as_io_context *aic = ad->io_context->aic; + struct as_io_context *aic; + spin_lock(&ad->io_context->lock); + aic = ad->io_context->aic; ad->antic_status = ANTIC_FINISHED; kblockd_schedule_work(&ad->antic_work); @@ -475,6 +479,7 @@ static void as_antic_timeout(unsigned long data) /* process not "saved" by a cooperating request */ ad->exit_no_coop = (7*ad->exit_no_coop + 256)/8; } + spin_unlock(&ad->io_context->lock); } spin_unlock_irqrestore(q->queue_lock, flags); } @@ -635,9 +640,11 @@ static int as_can_break_anticipation(struct as_data *ad, struct request *rq) ioc = ad->io_context; BUG_ON(!ioc); + spin_lock(&ioc->lock); if (rq && ioc == RQ_IOC(rq)) { /* request from same process */ + spin_unlock(&ioc->lock); return 1; } @@ -646,20 +653,25 @@ static int as_can_break_anticipation(struct as_data *ad, struct request *rq) * In this situation status should really be FINISHED, * however the timer hasn't had the chance to run yet. */ + spin_unlock(&ioc->lock); return 1; } aic = ioc->aic; - if (!aic) + if (!aic) { + spin_unlock(&ioc->lock); return 0; + } if (atomic_read(&aic->nr_queued) > 0) { /* process has more requests queued */ + spin_unlock(&ioc->lock); return 1; } if (atomic_read(&aic->nr_dispatched) > 0) { /* process has more requests dispatched */ + spin_unlock(&ioc->lock); return 1; } @@ -680,6 +692,7 @@ static int as_can_break_anticipation(struct as_data *ad, struct request *rq) } as_update_iohist(ad, aic, rq); + spin_unlock(&ioc->lock); return 1; } @@ -688,20 +701,27 @@ static int as_can_break_anticipation(struct as_data *ad, struct request *rq) if (aic->ttime_samples == 0) ad->exit_prob = (7*ad->exit_prob + 256)/8; - if (ad->exit_no_coop > 128) + if (ad->exit_no_coop > 128) { + spin_unlock(&ioc->lock); return 1; + } } if (aic->ttime_samples == 0) { - if (ad->new_ttime_mean > ad->antic_expire) + if (ad->new_ttime_mean > ad->antic_expire) { + spin_unlock(&ioc->lock); return 1; - if (ad->exit_prob * ad->exit_no_coop > 128*256) + } + if (ad->exit_prob * ad->exit_no_coop > 128*256) { + spin_unlock(&ioc->lock); return 1; + } } else if (aic->ttime_mean > ad->antic_expire) { /* the process thinks too much between requests */ + spin_unlock(&ioc->lock); return 1; } - + spin_unlock(&ioc->lock); return 0; } @@ -1255,7 +1275,13 @@ static void as_merged_requests(struct request_queue *q, struct request *req, * Don't copy here but swap, because when anext is * removed below, it must contain the unused context */ - swap_io_context(&rioc, &nioc); + if (rioc != nioc) { + double_spin_lock(&rioc->lock, &nioc->lock, + rioc < nioc); + swap_io_context(&rioc, &nioc); + double_spin_unlock(&rioc->lock, &nioc->lock, + rioc < nioc); + } } } diff --git a/block/blk-barrier.c b/block/blk-barrier.c new file mode 100644 index 00000000000..5f74fec327d --- /dev/null +++ b/block/blk-barrier.c @@ -0,0 +1,319 @@ +/* + * Functions related to barrier IO handling + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/bio.h> +#include <linux/blkdev.h> + +#include "blk.h" + +/** + * blk_queue_ordered - does this queue support ordered writes + * @q: the request queue + * @ordered: one of QUEUE_ORDERED_* + * @prepare_flush_fn: rq setup helper for cache flush ordered writes + * + * Description: + * For journalled file systems, doing ordered writes on a commit + * block instead of explicitly doing wait_on_buffer (which is bad + * for performance) can be a big win. Block drivers supporting this + * feature should call this function and indicate so. + * + **/ +int blk_queue_ordered(struct request_queue *q, unsigned ordered, + prepare_flush_fn *prepare_flush_fn) +{ + if (ordered & (QUEUE_ORDERED_PREFLUSH | QUEUE_ORDERED_POSTFLUSH) && + prepare_flush_fn == NULL) { + printk(KERN_ERR "blk_queue_ordered: prepare_flush_fn required\n"); + return -EINVAL; + } + + if (ordered != QUEUE_ORDERED_NONE && + ordered != QUEUE_ORDERED_DRAIN && + ordered != QUEUE_ORDERED_DRAIN_FLUSH && + ordered != QUEUE_ORDERED_DRAIN_FUA && + ordered != QUEUE_ORDERED_TAG && + ordered != QUEUE_ORDERED_TAG_FLUSH && + ordered != QUEUE_ORDERED_TAG_FUA) { + printk(KERN_ERR "blk_queue_ordered: bad value %d\n", ordered); + return -EINVAL; + } + + q->ordered = ordered; + q->next_ordered = ordered; + q->prepare_flush_fn = prepare_flush_fn; + + return 0; +} + +EXPORT_SYMBOL(blk_queue_ordered); + +/* + * Cache flushing for ordered writes handling + */ +inline unsigned blk_ordered_cur_seq(struct request_queue *q) +{ + if (!q->ordseq) + return 0; + return 1 << ffz(q->ordseq); +} + +unsigned blk_ordered_req_seq(struct request *rq) +{ + struct request_queue *q = rq->q; + + BUG_ON(q->ordseq == 0); + + if (rq == &q->pre_flush_rq) + return QUEUE_ORDSEQ_PREFLUSH; + if (rq == &q->bar_rq) + return QUEUE_ORDSEQ_BAR; + if (rq == &q->post_flush_rq) + return QUEUE_ORDSEQ_POSTFLUSH; + + /* + * !fs requests don't need to follow barrier ordering. Always + * put them at the front. This fixes the following deadlock. + * + * http://thread.gmane.org/gmane.linux.kernel/537473 + */ + if (!blk_fs_request(rq)) + return QUEUE_ORDSEQ_DRAIN; + + if ((rq->cmd_flags & REQ_ORDERED_COLOR) == + (q->orig_bar_rq->cmd_flags & REQ_ORDERED_COLOR)) + return QUEUE_ORDSEQ_DRAIN; + else + return QUEUE_ORDSEQ_DONE; +} + +void blk_ordered_complete_seq(struct request_queue *q, unsigned seq, int error) +{ + struct request *rq; + + if (error && !q->orderr) + q->orderr = error; + + BUG_ON(q->ordseq & seq); + q->ordseq |= seq; + + if (blk_ordered_cur_seq(q) != QUEUE_ORDSEQ_DONE) + return; + + /* + * Okay, sequence complete. + */ + q->ordseq = 0; + rq = q->orig_bar_rq; + + if (__blk_end_request(rq, q->orderr, blk_rq_bytes(rq))) + BUG(); +} + +static void pre_flush_end_io(struct request *rq, int error) +{ + elv_completed_request(rq->q, rq); + blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_PREFLUSH, error); +} + +static void bar_end_io(struct request *rq, int error) +{ + elv_completed_request(rq->q, rq); + blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_BAR, error); +} + +static void post_flush_end_io(struct request *rq, int error) +{ + elv_completed_request(rq->q, rq); + blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_POSTFLUSH, error); +} + +static void queue_flush(struct request_queue *q, unsigned which) +{ + struct request *rq; + rq_end_io_fn *end_io; + + if (which == QUEUE_ORDERED_PREFLUSH) { + rq = &q->pre_flush_rq; + end_io = pre_flush_end_io; + } else { + rq = &q->post_flush_rq; + end_io = post_flush_end_io; + } + + rq->cmd_flags = REQ_HARDBARRIER; + rq_init(q, rq); + rq->elevator_private = NULL; + rq->elevator_private2 = NULL; + rq->rq_disk = q->bar_rq.rq_disk; + rq->end_io = end_io; + q->prepare_flush_fn(q, rq); + + elv_insert(q, rq, ELEVATOR_INSERT_FRONT); +} + +static inline struct request *start_ordered(struct request_queue *q, + struct request *rq) +{ + q->orderr = 0; + q->ordered = q->next_ordered; + q->ordseq |= QUEUE_ORDSEQ_STARTED; + + /* + * Prep proxy barrier request. + */ + blkdev_dequeue_request(rq); + q->orig_bar_rq = rq; + rq = &q->bar_rq; + rq->cmd_flags = 0; + rq_init(q, rq); + if (bio_data_dir(q->orig_bar_rq->bio) == WRITE) + rq->cmd_flags |= REQ_RW; + if (q->ordered & QUEUE_ORDERED_FUA) + rq->cmd_flags |= REQ_FUA; + rq->elevator_private = NULL; + rq->elevator_private2 = NULL; + init_request_from_bio(rq, q->orig_bar_rq->bio); + rq->end_io = bar_end_io; + + /* + * Queue ordered sequence. As we stack them at the head, we + * need to queue in reverse order. Note that we rely on that + * no fs request uses ELEVATOR_INSERT_FRONT and thus no fs + * request gets inbetween ordered sequence. If this request is + * an empty barrier, we don't need to do a postflush ever since + * there will be no data written between the pre and post flush. + * Hence a single flush will suffice. + */ + if ((q->ordered & QUEUE_ORDERED_POSTFLUSH) && !blk_empty_barrier(rq)) + queue_flush(q, QUEUE_ORDERED_POSTFLUSH); + else + q->ordseq |= QUEUE_ORDSEQ_POSTFLUSH; + + elv_insert(q, rq, ELEVATOR_INSERT_FRONT); + + if (q->ordered & QUEUE_ORDERED_PREFLUSH) { + queue_flush(q, QUEUE_ORDERED_PREFLUSH); + rq = &q->pre_flush_rq; + } else + q->ordseq |= QUEUE_ORDSEQ_PREFLUSH; + + if ((q->ordered & QUEUE_ORDERED_TAG) || q->in_flight == 0) + q->ordseq |= QUEUE_ORDSEQ_DRAIN; + else + rq = NULL; + + return rq; +} + +int blk_do_ordered(struct request_queue *q, struct request **rqp) +{ + struct request *rq = *rqp; + const int is_barrier = blk_fs_request(rq) && blk_barrier_rq(rq); + + if (!q->ordseq) { + if (!is_barrier) + return 1; + + if (q->next_ordered != QUEUE_ORDERED_NONE) { + *rqp = start_ordered(q, rq); + return 1; + } else { + /* + * This can happen when the queue switches to + * ORDERED_NONE while this request is on it. + */ + blkdev_dequeue_request(rq); + if (__blk_end_request(rq, -EOPNOTSUPP, + blk_rq_bytes(rq))) + BUG(); + *rqp = NULL; + return 0; + } + } + + /* + * Ordered sequence in progress + */ + + /* Special requests are not subject to ordering rules. */ + if (!blk_fs_request(rq) && + rq != &q->pre_flush_rq && rq != &q->post_flush_rq) + return 1; + + if (q->ordered & QUEUE_ORDERED_TAG) { + /* Ordered by tag. Blocking the next barrier is enough. */ + if (is_barrier && rq != &q->bar_rq) + *rqp = NULL; + } else { + /* Ordered by draining. Wait for turn. */ + WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q)); + if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q)) + *rqp = NULL; + } + + return 1; +} + +static void bio_end_empty_barrier(struct bio *bio, int err) +{ + if (err) + clear_bit(BIO_UPTODATE, &bio->bi_flags); + + complete(bio->bi_private); +} + +/** + * blkdev_issue_flush - queue a flush + * @bdev: blockdev to issue flush for + * @error_sector: error sector + * + * Description: + * Issue a flush for the block device in question. Caller can supply + * room for storing the error offset in case of a flush error, if they + * wish to. Caller must run wait_for_completion() on its own. + */ +int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector) +{ + DECLARE_COMPLETION_ONSTACK(wait); + struct request_queue *q; + struct bio *bio; + int ret; + + if (bdev->bd_disk == NULL) + return -ENXIO; + + q = bdev_get_queue(bdev); + if (!q) + return -ENXIO; + + bio = bio_alloc(GFP_KERNEL, 0); + if (!bio) + return -ENOMEM; + + bio->bi_end_io = bio_end_empty_barrier; + bio->bi_private = &wait; + bio->bi_bdev = bdev; + submit_bio(1 << BIO_RW_BARRIER, bio); + + wait_for_completion(&wait); + + /* + * The driver must store the error location in ->bi_sector, if + * it supports it. For non-stacked drivers, this should be copied + * from rq->sector. + */ + if (error_sector) + *error_sector = bio->bi_sector; + + ret = 0; + if (!bio_flagged(bio, BIO_UPTODATE)) + ret = -EIO; + + bio_put(bio); + return ret; +} + +EXPORT_SYMBOL(blkdev_issue_flush); diff --git a/block/blk-core.c b/block/blk-core.c new file mode 100644 index 00000000000..8ff99440ee4 --- /dev/null +++ b/block/blk-core.c @@ -0,0 +1,2034 @@ +/* + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 1994, Karl Keyte: Added support for disk statistics + * Elevator latency, (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE + * Queue request tables / lock, selectable elevator, Jens Axboe <axboe@suse.de> + * kernel-doc documentation started by NeilBrown <neilb@cse.unsw.edu.au> - July2000 + * bio rewrite, highmem i/o, etc, Jens Axboe <axboe@suse.de> - may 2001 + */ + +/* + * This handles all read/write requests to block devices + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/backing-dev.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/kernel_stat.h> +#include <linux/string.h> +#include <linux/init.h> +#include <linux/completion.h> +#include <linux/slab.h> +#include <linux/swap.h> +#include <linux/writeback.h> +#include <linux/task_io_accounting_ops.h> +#include <linux/interrupt.h> +#include <linux/cpu.h> +#include <linux/blktrace_api.h> +#include <linux/fault-inject.h> + +#include "blk.h" + +static int __make_request(struct request_queue *q, struct bio *bio); + +/* + * For the allocated request tables + */ +struct kmem_cache *request_cachep; + +/* + * For queue allocation + */ +struct kmem_cache *blk_requestq_cachep = NULL; + +/* + * Controlling structure to kblockd + */ +static struct workqueue_struct *kblockd_workqueue; + +static DEFINE_PER_CPU(struct list_head, blk_cpu_done); + +static void drive_stat_acct(struct request *rq, int new_io) +{ + int rw = rq_data_dir(rq); + + if (!blk_fs_request(rq) || !rq->rq_disk) + return; + + if (!new_io) { + __disk_stat_inc(rq->rq_disk, merges[rw]); + } else { + disk_round_stats(rq->rq_disk); + rq->rq_disk->in_flight++; + } +} + +void blk_queue_congestion_threshold(struct request_queue *q) +{ + int nr; + + nr = q->nr_requests - (q->nr_requests / 8) + 1; + if (nr > q->nr_requests) + nr = q->nr_requests; + q->nr_congestion_on = nr; + + nr = q->nr_requests - (q->nr_requests / 8) - (q->nr_requests / 16) - 1; + if (nr < 1) + nr = 1; + q->nr_congestion_off = nr; +} + +/** + * blk_get_backing_dev_info - get the address of a queue's backing_dev_info + * @bdev: device + * + * Locates the passed device's request queue and returns the address of its + * backing_dev_info + * + * Will return NULL if the request queue cannot be located. + */ +struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev) +{ + struct backing_dev_info *ret = NULL; + struct request_queue *q = bdev_get_queue(bdev); + + if (q) + ret = &q->backing_dev_info; + return ret; +} +EXPORT_SYMBOL(blk_get_backing_dev_info); + +void rq_init(struct request_queue *q, struct request *rq) +{ + INIT_LIST_HEAD(&rq->queuelist); + INIT_LIST_HEAD(&rq->donelist); + + rq->errors = 0; + rq->bio = rq->biotail = NULL; + INIT_HLIST_NODE(&rq->hash); + RB_CLEAR_NODE(&rq->rb_node); + rq->ioprio = 0; + rq->buffer = NULL; + rq->ref_count = 1; + rq->q = q; + rq->special = NULL; + rq->data_len = 0; + rq->data = NULL; + rq->nr_phys_segments = 0; + rq->sense = NULL; + rq->end_io = NULL; + rq->end_io_data = NULL; + rq->completion_data = NULL; + rq->next_rq = NULL; +} + +static void req_bio_endio(struct request *rq, struct bio *bio, + unsigned int nbytes, int error) +{ + struct request_queue *q = rq->q; + + if (&q->bar_rq != rq) { + if (error) + clear_bit(BIO_UPTODATE, &bio->bi_flags); + else if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) + error = -EIO; + + if (unlikely(nbytes > bio->bi_size)) { + printk("%s: want %u bytes done, only %u left\n", + __FUNCTION__, nbytes, bio->bi_size); + nbytes = bio->bi_size; + } + + bio->bi_size -= nbytes; + bio->bi_sector += (nbytes >> 9); + if (bio->bi_size == 0) + bio_endio(bio, error); + } else { + + /* + * Okay, this is the barrier request in progress, just + * record the error; + */ + if (error && !q->orderr) + q->orderr = error; + } +} + +void blk_dump_rq_flags(struct request *rq, char *msg) +{ + int bit; + + printk("%s: dev %s: type=%x, flags=%x\n", msg, + rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->cmd_type, + rq->cmd_flags); + + printk("\nsector %llu, nr/cnr %lu/%u\n", (unsigned long long)rq->sector, + rq->nr_sectors, + rq->current_nr_sectors); + printk("bio %p, biotail %p, buffer %p, data %p, len %u\n", rq->bio, rq->biotail, rq->buffer, rq->data, rq->data_len); + + if (blk_pc_request(rq)) { + printk("cdb: "); + for (bit = 0; bit < sizeof(rq->cmd); bit++) + printk("%02x ", rq->cmd[bit]); + printk("\n"); + } +} + +EXPORT_SYMBOL(blk_dump_rq_flags); + +/* + * "plug" the device if there are no outstanding requests: this will + * force the transfer to start only after we have put all the requests + * on the list. + * + * This is called with interrupts off and no requests on the queue and + * with the queue lock held. + */ +void blk_plug_device(struct request_queue *q) +{ + WARN_ON(!irqs_disabled()); + + /* + * don't plug a stopped queue, it must be paired with blk_start_queue() + * which will restart the queueing + */ + if (blk_queue_stopped(q)) + return; + + if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags)) { + mod_timer(&q->unplug_timer, jiffies + q->unplug_delay); + blk_add_trace_generic(q, NULL, 0, BLK_TA_PLUG); + } +} + +EXPORT_SYMBOL(blk_plug_device); + +/* + * remove the queue from the plugged list, if present. called with + * queue lock held and interrupts disabled. + */ +int blk_remove_plug(struct request_queue *q) +{ + WARN_ON(!irqs_disabled()); + + if (!test_and_clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags)) + return 0; + + del_timer(&q->unplug_timer); + return 1; +} + +EXPORT_SYMBOL(blk_remove_plug); + +/* + * remove the plug and let it rip.. + */ +void __generic_unplug_device(struct request_queue *q) +{ + if (unlikely(blk_queue_stopped(q))) + return; + + if (!blk_remove_plug(q)) + return; + + q->request_fn(q); +} +EXPORT_SYMBOL(__generic_unplug_device); + +/** + * generic_unplug_device - fire a request queue + * @q: The &struct request_queue in question + * + * Description: + * Linux uses plugging to build bigger requests queues before letting + * the device have at them. If a queue is plugged, the I/O scheduler + * is still adding and merging requests on the queue. Once the queue + * gets unplugged, the request_fn defined for the queue is invoked and + * transfers started. + **/ +void generic_unplug_device(struct request_queue *q) +{ + spin_lock_irq(q->queue_lock); + __generic_unplug_device(q); + spin_unlock_irq(q->queue_lock); +} +EXPORT_SYMBOL(generic_unplug_device); + +static void blk_backing_dev_unplug(struct backing_dev_info *bdi, + struct page *page) +{ + struct request_queue *q = bdi->unplug_io_data; + + blk_unplug(q); +} + +void blk_unplug_work(struct work_struct *work) +{ + struct request_queue *q = + container_of(work, struct request_queue, unplug_work); + + blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL, + q->rq.count[READ] + q->rq.count[WRITE]); + + q->unplug_fn(q); +} + +void blk_unplug_timeout(unsigned long data) +{ + struct request_queue *q = (struct request_queue *)data; + + blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_TIMER, NULL, + q->rq.count[READ] + q->rq.count[WRITE]); + + kblockd_schedule_work(&q->unplug_work); +} + +void blk_unplug(struct request_queue *q) +{ + /* + * devices don't necessarily have an ->unplug_fn defined + */ + if (q->unplug_fn) { + blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL, + q->rq.count[READ] + q->rq.count[WRITE]); + + q->unplug_fn(q); + } +} +EXPORT_SYMBOL(blk_unplug); + +/** + * blk_start_queue - restart a previously stopped queue + * @q: The &struct request_queue in question + * + * Description: + * blk_start_queue() will clear the stop flag on the queue, and call + * the request_fn for the queue if it was in a stopped state when + * entered. Also see blk_stop_queue(). Queue lock must be held. + **/ +void blk_start_queue(struct request_queue *q) +{ + WARN_ON(!irqs_disabled()); + + clear_bit(QUEUE_FLAG_STOPPED, &q->queue_flags); + + /* + * one level of recursion is ok and is much faster than kicking + * the unplug handling + */ + if (!test_and_set_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) { + q->request_fn(q); + clear_bit(QUEUE_FLAG_REENTER, &q->queue_flags); + } else { + blk_plug_device(q); + kblockd_schedule_work(&q->unplug_work); + } +} + +EXPORT_SYMBOL(blk_start_queue); + +/** + * blk_stop_queue - stop a queue + * @q: The &struct request_queue in question + * + * Description: + * The Linux block layer assumes that a block driver will consume all + * entries on the request queue when the request_fn strategy is called. + * Often this will not happen, because of hardware limitations (queue + * depth settings). If a device driver gets a 'queue full' response, + * or if it simply chooses not to queue more I/O at one point, it can + * call this function to prevent the request_fn from being called until + * the driver has signalled it's ready to go again. This happens by calling + * blk_start_queue() to restart queue operations. Queue lock must be held. + **/ +void blk_stop_queue(struct request_queue *q) +{ + blk_remove_plug(q); + set_bit(QUEUE_FLAG_STOPPED, &q->queue_flags); +} +EXPORT_SYMBOL(blk_stop_queue); + +/** + * blk_sync_queue - cancel any pending callbacks on a queue + * @q: the queue + * + * Description: + * The block layer may perform asynchronous callback activity + * on a queue, such as calling the unplug function after a timeout. + * A block device may call blk_sync_queue to ensure that any + * such activity is cancelled, thus allowing it to release resources + * that the callbacks might use. The caller must already have made sure + * that its ->make_request_fn will not re-add plugging prior to calling + * this function. + * + */ +void blk_sync_queue(struct request_queue *q) +{ + del_timer_sync(&q->unplug_timer); + kblockd_flush_work(&q->unplug_work); +} +EXPORT_SYMBOL(blk_sync_queue); + +/** + * blk_run_queue - run a single device queue + * @q: The queue to run + */ +void blk_run_queue(struct request_queue *q) +{ + unsigned long flags; + + spin_lock_irqsave(q->queue_lock, flags); + blk_remove_plug(q); + + /* + * Only recurse once to avoid overrunning the stack, let the unplug + * handling reinvoke the handler shortly if we already got there. + */ + if (!elv_queue_empty(q)) { + if (!test_and_set_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) { + q->request_fn(q); + clear_bit(QUEUE_FLAG_REENTER, &q->queue_flags); + } else { + blk_plug_device(q); + kblockd_schedule_work(&q->unplug_work); + } + } + + spin_unlock_irqrestore(q->queue_lock, flags); +} +EXPORT_SYMBOL(blk_run_queue); + +void blk_put_queue(struct request_queue *q) +{ + kobject_put(&q->kobj); +} +EXPORT_SYMBOL(blk_put_queue); + +void blk_cleanup_queue(struct request_queue * q) +{ + mutex_lock(&q->sysfs_lock); + set_bit(QUEUE_FLAG_DEAD, &q->queue_flags); + mutex_unlock(&q->sysfs_lock); + + if (q->elevator) + elevator_exit(q->elevator); + + blk_put_queue(q); +} + +EXPORT_SYMBOL(blk_cleanup_queue); + +static int blk_init_free_list(struct request_queue *q) +{ + struct request_list *rl = &q->rq; + + rl->count[READ] = rl->count[WRITE] = 0; + rl->starved[READ] = rl->starved[WRITE] = 0; + rl->elvpriv = 0; + init_waitqueue_head(&rl->wait[READ]); + init_waitqueue_head(&rl->wait[WRITE]); + + rl->rq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab, + mempool_free_slab, request_cachep, q->node); + + if (!rl->rq_pool) + return -ENOMEM; + + return 0; +} + +struct request_queue *blk_alloc_queue(gfp_t gfp_mask) +{ + return blk_alloc_queue_node(gfp_mask, -1); +} +EXPORT_SYMBOL(blk_alloc_queue); + +struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) +{ + struct request_queue *q; + int err; + + q = kmem_cache_alloc_node(blk_requestq_cachep, + gfp_mask | __GFP_ZERO, node_id); + if (!q) + return NULL; + + q->backing_dev_info.unplug_io_fn = blk_backing_dev_unplug; + q->backing_dev_info.unplug_io_data = q; + err = bdi_init(&q->backing_dev_info); + if (err) { + kmem_cache_free(blk_requestq_cachep, q); + return NULL; + } + + init_timer(&q->unplug_timer); + + kobject_init(&q->kobj, &blk_queue_ktype); + + mutex_init(&q->sysfs_lock); + + return q; +} +EXPORT_SYMBOL(blk_alloc_queue_node); + +/** + * blk_init_queue - prepare a request queue for use with a block device + * @rfn: The function to be called to process requests that have been + * placed on the queue. + * @lock: Request queue spin lock + * + * Description: + * If a block device wishes to use the standard request handling procedures, + * which sorts requests and coalesces adjacent requests, then it must + * call blk_init_queue(). The function @rfn will be called when there + * are requests on the queue that need to be processed. If the device + * supports plugging, then @rfn may not be called immediately when requests + * are available on the queue, but may be called at some time later instead. + * Plugged queues are generally unplugged when a buffer belonging to one + * of the requests on the queue is needed, or due to memory pressure. + * + * @rfn is not required, or even expected, to remove all requests off the + * queue, but only as many as it can handle at a time. If it does leave + * requests on the queue, it is responsible for arranging that the requests + * get dealt with eventually. + * + * The queue spin lock must be held while manipulating the requests on the + * request queue; this lock will be taken also from interrupt context, so irq + * disabling is needed for it. + * + * Function returns a pointer to the initialized request queue, or NULL if + * it didn't succeed. + * + * Note: + * blk_init_queue() must be paired with a blk_cleanup_queue() call + * when the block device is deactivated (such as at module unload). + **/ + +struct request_queue *blk_init_queue(request_fn_proc *rfn, spinlock_t *lock) +{ + return blk_init_queue_node(rfn, lock, -1); +} +EXPORT_SYMBOL(blk_init_queue); + +struct request_queue * +blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id) +{ + struct request_queue *q = blk_alloc_queue_node(GFP_KERNEL, node_id); + + if (!q) + return NULL; + + q->node = node_id; + if (blk_init_free_list(q)) { + kmem_cache_free(blk_requestq_cachep, q); + return NULL; + } + + /* + * if caller didn't supply a lock, they get per-queue locking with + * our embedded lock + */ + if (!lock) { + spin_lock_init(&q->__queue_lock); + lock = &q->__queue_lock; + } + + q->request_fn = rfn; + q->prep_rq_fn = NULL; + q->unplug_fn = generic_unplug_device; + q->queue_flags = (1 << QUEUE_FLAG_CLUSTER); + q->queue_lock = lock; + + blk_queue_segment_boundary(q, 0xffffffff); + + blk_queue_make_request(q, __make_request); + blk_queue_max_segment_size(q, MAX_SEGMENT_SIZE); + + blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS); + blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS); + + q->sg_reserved_size = INT_MAX; + + /* + * all done + */ + if (!elevator_init(q, NULL)) { + blk_queue_congestion_threshold(q); + return q; + } + + blk_put_queue(q); + return NULL; +} +EXPORT_SYMBOL(blk_init_queue_node); + +int blk_get_queue(struct request_queue *q) +{ + if (likely(!test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) { + kobject_get(&q->kobj); + return 0; + } + + return 1; +} + +EXPORT_SYMBOL(blk_get_queue); + +static inline void blk_free_request(struct request_queue *q, struct request *rq) +{ + if (rq->cmd_flags & REQ_ELVPRIV) + elv_put_request(q, rq); + mempool_free(rq, q->rq.rq_pool); +} + +static struct request * +blk_alloc_request(struct request_queue *q, int rw, int priv, gfp_t gfp_mask) +{ + struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask); + + if (!rq) + return NULL; + + /* + * first three bits are identical in rq->cmd_flags and bio->bi_rw, + * see bio.h and blkdev.h + */ + rq->cmd_flags = rw | REQ_ALLOCED; + + if (priv) { + if (unlikely(elv_set_request(q, rq, gfp_mask))) { + mempool_free(rq, q->rq.rq_pool); + return NULL; + } + rq->cmd_flags |= REQ_ELVPRIV; + } + + return rq; +} + +/* + * ioc_batching returns true if the ioc is a valid batching request and + * should be given priority access to a request. + */ +static inline int ioc_batching(struct request_queue *q, struct io_context *ioc) +{ + if (!ioc) + return 0; + + /* + * Make sure the process is able to allocate at least 1 request + * even if the batch times out, otherwise we could theoretically + * lose wakeups. + */ + return ioc->nr_batch_requests == q->nr_batching || + (ioc->nr_batch_requests > 0 + && time_before(jiffies, ioc->last_waited + BLK_BATCH_TIME)); +} + +/* + * ioc_set_batching sets ioc to be a new "batcher" if it is not one. This + * will cause the process to be a "batcher" on all queues in the system. This + * is the behaviour we want though - once it gets a wakeup it should be given + * a nice run. + */ +static void ioc_set_batching(struct request_queue *q, struct io_context *ioc) +{ + if (!ioc || ioc_batching(q, ioc)) + return; + + ioc->nr_batch_requests = q->nr_batching; + ioc->last_waited = jiffies; +} + +static void __freed_request(struct request_queue *q, int rw) +{ + struct request_list *rl = &q->rq; + + if (rl->count[rw] < queue_congestion_off_threshold(q)) + blk_clear_queue_congested(q, rw); + + if (rl->count[rw] + 1 <= q->nr_requests) { + if (waitqueue_active(&rl->wait[rw])) + wake_up(&rl->wait[rw]); + + blk_clear_queue_full(q, rw); + } +} + +/* + * A request has just been released. Account for it, update the full and + * congestion status, wake up any waiters. Called under q->queue_lock. + */ +static void freed_request(struct request_queue *q, int rw, int priv) +{ + struct request_list *rl = &q->rq; + + rl->count[rw]--; + if (priv) + rl->elvpriv--; + + __freed_request(q, rw); + + if (unlikely(rl->starved[rw ^ 1])) + __freed_request(q, rw ^ 1); +} + +#define blkdev_free_rq(list) list_entry((list)->next, struct request, queuelist) +/* + * Get a free request, queue_lock must be held. + * Returns NULL on failure, with queue_lock held. + * Returns !NULL on success, with queue_lock *not held*. + */ +static struct request *get_request(struct request_queue *q, int rw_flags, + struct bio *bio, gfp_t gfp_mask) +{ + struct request *rq = NULL; + struct request_list *rl = &q->rq; + struct io_context *ioc = NULL; + const int rw = rw_flags & 0x01; + int may_queue, priv; + + may_queue = elv_may_queue(q, rw_flags); + if (may_queue == ELV_MQUEUE_NO) + goto rq_starved; + + if (rl->count[rw]+1 >= queue_congestion_on_threshold(q)) { + if (rl->count[rw]+1 >= q->nr_requests) { + ioc = current_io_context(GFP_ATOMIC, q->node); + /* + * The queue will fill after this allocation, so set + * it as full, and mark this process as "batching". + * This process will be allowed to complete a batch of + * requests, others will be blocked. + */ + if (!blk_queue_full(q, rw)) { + ioc_set_batching(q, ioc); + blk_set_queue_full(q, rw); + } else { + if (may_queue != ELV_MQUEUE_MUST + && !ioc_batching(q, ioc)) { + /* + * The queue is full and the allocating + * process is not a "batcher", and not + * exempted by the IO scheduler + */ + goto out; + } + } + } + blk_set_queue_congested(q, rw); + } + + /* + * Only allow batching queuers to allocate up to 50% over the defined + * limit of requests, otherwise we could have thousands of requests + * allocated with any setting of ->nr_requests + */ + if (rl->count[rw] >= (3 * q->nr_requests / 2)) + goto out; + + rl->count[rw]++; + rl->starved[rw] = 0; + + priv = !test_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags); + if (priv) + rl->elvpriv++; + + spin_unlock_irq(q->queue_lock); + + rq = blk_alloc_request(q, rw_flags, priv, gfp_mask); + if (unlikely(!rq)) { + /* + * Allocation failed presumably due to memory. Undo anything + * we might have messed up. + * + * Allocating task should really be put onto the front of the + * wait queue, but this is pretty rare. + */ + spin_lock_irq(q->queue_lock); + freed_request(q, rw, priv); + + /* + * in the very unlikely event that allocation failed and no + * requests for this direction was pending, mark us starved + * so that freeing of a request in the other direction will + * notice us. another possible fix would be to split the + * rq mempool into READ and WRITE + */ +rq_starved: + if (unlikely(rl->count[rw] == 0)) + rl->starved[rw] = 1; + + goto out; + } + + /* + * ioc may be NULL here, and ioc_batching will be false. That's + * OK, if the queue is under the request limit then requests need + * not count toward the nr_batch_requests limit. There will always + * be some limit enforced by BLK_BATCH_TIME. + */ + if (ioc_batching(q, ioc)) + ioc->nr_batch_requests--; + + rq_init(q, rq); + + blk_add_trace_generic(q, bio, rw, BLK_TA_GETRQ); +out: + return rq; +} + +/* + * No available requests for this queue, unplug the device and wait for some + * requests to become available. + * + * Called with q->queue_lock held, and returns with it unlocked. + */ +static struct request *get_request_wait(struct request_queue *q, int rw_flags, + struct bio *bio) +{ + const int rw = rw_flags & 0x01; + struct request *rq; + + rq = get_request(q, rw_flags, bio, GFP_NOIO); + while (!rq) { + DEFINE_WAIT(wait); + struct request_list *rl = &q->rq; + + prepare_to_wait_exclusive(&rl->wait[rw], &wait, + TASK_UNINTERRUPTIBLE); + + rq = get_request(q, rw_flags, bio, GFP_NOIO); + + if (!rq) { + struct io_context *ioc; + + blk_add_trace_generic(q, bio, rw, BLK_TA_SLEEPRQ); + + __generic_unplug_device(q); + spin_unlock_irq(q->queue_lock); + io_schedule(); + + /* + * After sleeping, we become a "batching" process and + * will be able to allocate at least one request, and + * up to a big batch of them for a small period time. + * See ioc_batching, ioc_set_batching + */ + ioc = current_io_context(GFP_NOIO, q->node); + ioc_set_batching(q, ioc); + + spin_lock_irq(q->queue_lock); + } + finish_wait(&rl->wait[rw], &wait); + } + + return rq; +} + +struct request *blk_get_request(struct request_queue *q, int rw, gfp_t gfp_mask) +{ + struct request *rq; + + BUG_ON(rw != READ && rw != WRITE); + + spin_lock_irq(q->queue_lock); + if (gfp_mask & __GFP_WAIT) { + rq = get_request_wait(q, rw, NULL); + } else { + rq = get_request(q, rw, NULL, gfp_mask); + if (!rq) + spin_unlock_irq(q->queue_lock); + } + /* q->queue_lock is unlocked at this point */ + + return rq; +} +EXPORT_SYMBOL(blk_get_request); + +/** + * blk_start_queueing - initiate dispatch of requests to device + * @q: request queue to kick into gear + * + * This is basically a helper to remove the need to know whether a queue + * is plugged or not if someone just wants to initiate dispatch of requests + * for this queue. + * + * The queue lock must be held with interrupts disabled. + */ +void blk_start_queueing(struct request_queue *q) +{ + if (!blk_queue_plugged(q)) + q->request_fn(q); + else + __generic_unplug_device(q); +} +EXPORT_SYMBOL(blk_start_queueing); + +/** + * blk_requeue_request - put a request back on queue + * @q: request queue where request should be inserted + * @rq: request to be inserted + * + * Description: + * Drivers often keep queueing requests until the hardware cannot accept + * more, when that condition happens we need to put the request back + * on the queue. Must be called with queue lock held. + */ +void blk_requeue_request(struct request_queue *q, struct request *rq) +{ + blk_add_trace_rq(q, rq, BLK_TA_REQUEUE); + + if (blk_rq_tagged(rq)) + blk_queue_end_tag(q, rq); + + elv_requeue_request(q, rq); +} + +EXPORT_SYMBOL(blk_requeue_request); + +/** + * blk_insert_request - insert a special request in to a request queue + * @q: request queue where request should be inserted + * @rq: request to be inserted + * @at_head: insert request at head or tail of queue + * @data: private data + * + * Description: + * Many block devices need to execute commands asynchronously, so they don't + * block the whole kernel from preemption during request execution. This is + * accomplished normally by inserting aritficial requests tagged as + * REQ_SPECIAL in to the corresponding request queue, and letting them be + * scheduled for actual execution by the request queue. + * + * We have the option of inserting the head or the tail of the queue. + * Typically we use the tail for new ioctls and so forth. We use the head + * of the queue for things like a QUEUE_FULL message from a device, or a + * host that is unable to accept a particular command. + */ +void blk_insert_request(struct request_queue *q, struct request *rq, + int at_head, void *data) +{ + int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK; + unsigned long flags; + + /* + * tell I/O scheduler that this isn't a regular read/write (ie it + * must not attempt merges on this) and that it acts as a soft + * barrier + */ + rq->cmd_type = REQ_TYPE_SPECIAL; + rq->cmd_flags |= REQ_SOFTBARRIER; + + rq->special = data; + + spin_lock_irqsave(q->queue_lock, flags); + + /* + * If command is tagged, release the tag + */ + if (blk_rq_tagged(rq)) + blk_queue_end_tag(q, rq); + + drive_stat_acct(rq, 1); + __elv_add_request(q, rq, where, 0); + blk_start_queueing(q); + spin_unlock_irqrestore(q->queue_lock, flags); +} + +EXPORT_SYMBOL(blk_insert_request); + +/* + * add-request adds a request to the linked list. + * queue lock is held and interrupts disabled, as we muck with the + * request queue list. + */ +static inline void add_request(struct request_queue * q, struct request * req) +{ + drive_stat_acct(req, 1); + + /* + * elevator indicated where it wants this request to be + * inserted at elevator_merge time + */ + __elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0); +} + +/* + * disk_round_stats() - Round off the performance stats on a struct + * disk_stats. + * + * The average IO queue length and utilisation statistics are maintained + * by observing the current state of the queue length and the amount of + * time it has been in this state for. + * + * Normally, that accounting is done on IO completion, but that can result + * in more than a second's worth of IO being accounted for within any one + * second, leading to >100% utilisation. To deal with that, we call this + * function to do a round-off before returning the results when reading + * /proc/diskstats. This accounts immediately for all queue usage up to + * the current jiffies and restarts the counters again. + */ +void disk_round_stats(struct gendisk *disk) +{ + unsigned long now = jiffies; + + if (now == disk->stamp) + return; + + if (disk->in_flight) { + __disk_stat_add(disk, time_in_queue, + disk->in_flight * (now - disk->stamp)); + __disk_stat_add(disk, io_ticks, (now - disk->stamp)); + } + disk->stamp = now; +} + +EXPORT_SYMBOL_GPL(disk_round_stats); + +/* + * queue lock must be held + */ +void __blk_put_request(struct request_queue *q, struct request *req) +{ + if (unlikely(!q)) + return; + if (unlikely(--req->ref_count)) + return; + + elv_completed_request(q, req); + + /* + * Request may not have originated from ll_rw_blk. if not, + * it didn't come out of our reserved rq pools + */ + if (req->cmd_flags & REQ_ALLOCED) { + int rw = rq_data_dir(req); + int priv = req->cmd_flags & REQ_ELVPRIV; + + BUG_ON(!list_empty(&req->queuelist)); + BUG_ON(!hlist_unhashed(&req->hash)); + + blk_free_request(q, req); + freed_request(q, rw, priv); + } +} + +EXPORT_SYMBOL_GPL(__blk_put_request); + +void blk_put_request(struct request *req) +{ + unsigned long flags; + struct request_queue *q = req->q; + + /* + * Gee, IDE calls in w/ NULL q. Fix IDE and remove the + * following if (q) test. + */ + if (q) { + spin_lock_irqsave(q->queue_lock, flags); + __blk_put_request(q, req); + spin_unlock_irqrestore(q->queue_lock, flags); + } +} + +EXPORT_SYMBOL(blk_put_request); + +void init_request_from_bio(struct request *req, struct bio *bio) +{ + req->cmd_type = REQ_TYPE_FS; + + /* + * inherit FAILFAST from bio (for read-ahead, and explicit FAILFAST) + */ + if (bio_rw_ahead(bio) || bio_failfast(bio)) + req->cmd_flags |= REQ_FAILFAST; + + /* + * REQ_BARRIER implies no merging, but lets make it explicit + */ + if (unlikely(bio_barrier(bio))) + req->cmd_flags |= (REQ_HARDBARRIER | REQ_NOMERGE); + + if (bio_sync(bio)) + req->cmd_flags |= REQ_RW_SYNC; + if (bio_rw_meta(bio)) + req->cmd_flags |= REQ_RW_META; + + req->errors = 0; + req->hard_sector = req->sector = bio->bi_sector; + req->ioprio = bio_prio(bio); + req->start_time = jiffies; + blk_rq_bio_prep(req->q, req, bio); +} + +static int __make_request(struct request_queue *q, struct bio *bio) +{ + struct request *req; + int el_ret, nr_sectors, barrier, err; + const unsigned short prio = bio_prio(bio); + const int sync = bio_sync(bio); + int rw_flags; + + nr_sectors = bio_sectors(bio); + + /* + * low level driver can indicate that it wants pages above a + * certain limit bounced to low memory (ie for highmem, or even + * ISA dma in theory) + */ + blk_queue_bounce(q, &bio); + + barrier = bio_barrier(bio); + if (unlikely(barrier) && (q->next_ordered == QUEUE_ORDERED_NONE)) { + err = -EOPNOTSUPP; + goto end_io; + } + + spin_lock_irq(q->queue_lock); + + if (unlikely(barrier) || elv_queue_empty(q)) + goto get_rq; + + el_ret = elv_merge(q, &req, bio); + switch (el_ret) { + case ELEVATOR_BACK_MERGE: + BUG_ON(!rq_mergeable(req)); + + if (!ll_back_merge_fn(q, req, bio)) + break; + + blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE); + + req->biotail->bi_next = bio; + req->biotail = bio; + req->nr_sectors = req->hard_nr_sectors += nr_sectors; + req->ioprio = ioprio_best(req->ioprio, prio); + drive_stat_acct(req, 0); + if (!attempt_back_merge(q, req)) + elv_merged_request(q, req, el_ret); + goto out; + + case ELEVATOR_FRONT_MERGE: + BUG_ON(!rq_mergeable(req)); + + if (!ll_front_merge_fn(q, req, bio)) + break; + + blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE); + + bio->bi_next = req->bio; + req->bio = bio; + + /* + * may not be valid. if the low level driver said + * it didn't need a bounce buffer then it better + * not touch req->buffer either... + */ + req->buffer = bio_data(bio); + req->current_nr_sectors = bio_cur_sectors(bio); + req->hard_cur_sectors = req->current_nr_sectors; + req->sector = req->hard_sector = bio->bi_sector; + req->nr_sectors = req->hard_nr_sectors += nr_sectors; + req->ioprio = ioprio_best(req->ioprio, prio); + drive_stat_acct(req, 0); + if (!attempt_front_merge(q, req)) + elv_merged_request(q, req, el_ret); + goto out; + + /* ELV_NO_MERGE: elevator says don't/can't merge. */ + default: + ; + } + +get_rq: + /* + * This sync check and mask will be re-done in init_request_from_bio(), + * but we need to set it earlier to expose the sync flag to the + * rq allocator and io schedulers. + */ + rw_flags = bio_data_dir(bio); + if (sync) + rw_flags |= REQ_RW_SYNC; + + /* + * Grab a free request. This is might sleep but can not fail. + * Returns with the queue unlocked. + */ + req = get_request_wait(q, rw_flags, bio); + + /* + * After dropping the lock and possibly sleeping here, our request + * may now be mergeable after it had proven unmergeable (above). + * We don't worry about that case for efficiency. It won't happen + * often, and the elevators are able to handle it. + */ + init_request_from_bio(req, bio); + + spin_lock_irq(q->queue_lock); + if (elv_queue_empty(q)) + blk_plug_device(q); + add_request(q, req); +out: + if (sync) + __generic_unplug_device(q); + + spin_unlock_irq(q->queue_lock); + return 0; + +end_io: + bio_endio(bio, err); + return 0; +} + +/* + * If bio->bi_dev is a partition, remap the location + */ +static inline void blk_partition_remap(struct bio *bio) +{ + struct block_device *bdev = bio->bi_bdev; + + if (bio_sectors(bio) && bdev != bdev->bd_contains) { + struct hd_struct *p = bdev->bd_part; + const int rw = bio_data_dir(bio); + + p->sectors[rw] += bio_sectors(bio); + p->ios[rw]++; + + bio->bi_sector += p->start_sect; + bio->bi_bdev = bdev->bd_contains; + + blk_add_trace_remap(bdev_get_queue(bio->bi_bdev), bio, + bdev->bd_dev, bio->bi_sector, + bio->bi_sector - p->start_sect); + } +} + +static void handle_bad_sector(struct bio *bio) +{ + char b[BDEVNAME_SIZE]; + + printk(KERN_INFO "attempt to access beyond end of device\n"); + printk(KERN_INFO "%s: rw=%ld, want=%Lu, limit=%Lu\n", + bdevname(bio->bi_bdev, b), + bio->bi_rw, + (unsigned long long)bio->bi_sector + bio_sectors(bio), + (long long)(bio->bi_bdev->bd_inode->i_size >> 9)); + + set_bit(BIO_EOF, &bio->bi_flags); +} + +#ifdef CONFIG_FAIL_MAKE_REQUEST + +static DECLARE_FAULT_ATTR(fail_make_request); + +static int __init setup_fail_make_request(char *str) +{ + return setup_fault_attr(&fail_make_request, str); +} +__setup("fail_make_request=", setup_fail_make_request); + +static int should_fail_request(struct bio *bio) +{ + if ((bio->bi_bdev->bd_disk->flags & GENHD_FL_FAIL) || + (bio->bi_bdev->bd_part && bio->bi_bdev->bd_part->make_it_fail)) + return should_fail(&fail_make_request, bio->bi_size); + + return 0; +} + +static int __init fail_make_request_debugfs(void) +{ + return init_fault_attr_dentries(&fail_make_request, + "fail_make_request"); +} + +late_initcall(fail_make_request_debugfs); + +#else /* CONFIG_FAIL_MAKE_REQUEST */ + +static inline int should_fail_request(struct bio *bio) +{ + return 0; +} + +#endif /* CONFIG_FAIL_MAKE_REQUEST */ + +/* + * Check whether this bio extends beyond the end of the device. + */ +static inline int bio_check_eod(struct bio *bio, unsigned int nr_sectors) +{ + sector_t maxsector; + + if (!nr_sectors) + return 0; + + /* Test device or partition size, when known. */ + maxsector = bio->bi_bdev->bd_inode->i_size >> 9; + if (maxsector) { + sector_t sector = bio->bi_sector; + + if (maxsector < nr_sectors || maxsector - nr_sectors < sector) { + /* + * This may well happen - the kernel calls bread() + * without checking the size of the device, e.g., when + * mounting a device. + */ + handle_bad_sector(bio); + return 1; + } + } + + return 0; +} + +/** + * generic_make_request: hand a buffer to its device driver for I/O + * @bio: The bio describing the location in memory and on the device. + * + * generic_make_request() is used to make I/O requests of block + * devices. It is passed a &struct bio, which describes the I/O that needs + * to be done. + * + * generic_make_request() does not return any status. The + * success/failure status of the request, along with notification of + * completion, is delivered asynchronously through the bio->bi_end_io + * function described (one day) else where. + * + * The caller of generic_make_request must make sure that bi_io_vec + * are set to describe the memory buffer, and that bi_dev and bi_sector are + * set to describe the device address, and the + * bi_end_io and optionally bi_private are set to describe how + * completion notification should be signaled. + * + * generic_make_request and the drivers it calls may use bi_next if this + * bio happens to be merged with someone else, and may change bi_dev and + * bi_sector for remaps as it sees fit. So the values of these fields + * should NOT be depended on after the call to generic_make_request. + */ +static inline void __generic_make_request(struct bio *bio) +{ + struct request_queue *q; + sector_t old_sector; + int ret, nr_sectors = bio_sectors(bio); + dev_t old_dev; + int err = -EIO; + + might_sleep(); + + if (bio_check_eod(bio, nr_sectors)) + goto end_io; + + /* + * Resolve the mapping until finished. (drivers are + * still free to implement/resolve their own stacking + * by explicitly returning 0) + * + * NOTE: we don't repeat the blk_size check for each new device. + * Stacking drivers are expected to know what they are doing. + */ + old_sector = -1; + old_dev = 0; + do { + char b[BDEVNAME_SIZE]; + + q = bdev_get_queue(bio->bi_bdev); + if (!q) { + printk(KERN_ERR + "generic_make_request: Trying to access " + "nonexistent block-device %s (%Lu)\n", + bdevname(bio->bi_bdev, b), + (long long) bio->bi_sector); +end_io: + bio_endio(bio, err); + break; + } + + if (unlikely(nr_sectors > q->max_hw_sectors)) { + printk("bio too big device %s (%u > %u)\n", + bdevname(bio->bi_bdev, b), + bio_sectors(bio), + q->max_hw_sectors); + goto end_io; + } + + if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) + goto end_io; + + if (should_fail_request(bio)) + goto end_io; + + /* + * If this device has partitions, remap block n + * of partition p to block n+start(p) of the disk. + */ + blk_partition_remap(bio); + + if (old_sector != -1) + blk_add_trace_remap(q, bio, old_dev, bio->bi_sector, + old_sector); + + blk_add_trace_bio(q, bio, BLK_TA_QUEUE); + + old_sector = bio->bi_sector; + old_dev = bio->bi_bdev->bd_dev; + + if (bio_check_eod(bio, nr_sectors)) + goto end_io; + if (bio_empty_barrier(bio) && !q->prepare_flush_fn) { + err = -EOPNOTSUPP; + goto end_io; + } + + ret = q->make_request_fn(q, bio); + } while (ret); +} + +/* + * We only want one ->make_request_fn to be active at a time, + * else stack usage with stacked devices could be a problem. + * So use current->bio_{list,tail} to keep a list of requests + * submited by a make_request_fn function. + * current->bio_tail is also used as a flag to say if + * generic_make_request is currently active in this task or not. + * If it is NULL, then no make_request is active. If it is non-NULL, + * then a make_request is active, and new requests should be added + * at the tail + */ +void generic_make_request(struct bio *bio) +{ + if (current->bio_tail) { + /* make_request is active */ + *(current->bio_tail) = bio; + bio->bi_next = NULL; + current->bio_tail = &bio->bi_next; + return; + } + /* following loop may be a bit non-obvious, and so deserves some + * explanation. + * Before entering the loop, bio->bi_next is NULL (as all callers + * ensure that) so we have a list with a single bio. + * We pretend that we have just taken it off a longer list, so + * we assign bio_list to the next (which is NULL) and bio_tail + * to &bio_list, thus initialising the bio_list of new bios to be + * added. __generic_make_request may indeed add some more bios + * through a recursive call to generic_make_request. If it + * did, we find a non-NULL value in bio_list and re-enter the loop + * from the top. In this case we really did just take the bio + * of the top of the list (no pretending) and so fixup bio_list and + * bio_tail or bi_next, and call into __generic_make_request again. + * + * The loop was structured like this to make only one call to + * __generic_make_request (which is important as it is large and + * inlined) and to keep the structure simple. + */ + BUG_ON(bio->bi_next); + do { + current->bio_list = bio->bi_next; + if (bio->bi_next == NULL) + current->bio_tail = ¤t->bio_list; + else + bio->bi_next = NULL; + __generic_make_request(bio); + bio = current->bio_list; + } while (bio); + current->bio_tail = NULL; /* deactivate */ +} + +EXPORT_SYMBOL(generic_make_request); + +/** + * submit_bio: submit a bio to the block device layer for I/O + * @rw: whether to %READ or %WRITE, or maybe to %READA (read ahead) + * @bio: The &struct bio which describes the I/O + * + * submit_bio() is very similar in purpose to generic_make_request(), and + * uses that function to do most of the work. Both are fairly rough + * interfaces, @bio must be presetup and ready for I/O. + * + */ +void submit_bio(int rw, struct bio *bio) +{ + int count = bio_sectors(bio); + + bio->bi_rw |= rw; + + /* + * If it's a regular read/write or a barrier with data attached, + * go through the normal accounting stuff before submission. + */ + if (!bio_empty_barrier(bio)) { + + BIO_BUG_ON(!bio->bi_size); + BIO_BUG_ON(!bio->bi_io_vec); + + if (rw & WRITE) { + count_vm_events(PGPGOUT, count); + } else { + task_io_account_read(bio->bi_size); + count_vm_events(PGPGIN, count); + } + + if (unlikely(block_dump)) { + char b[BDEVNAME_SIZE]; + printk(KERN_DEBUG "%s(%d): %s block %Lu on %s\n", + current->comm, task_pid_nr(current), + (rw & WRITE) ? "WRITE" : "READ", + (unsigned long long)bio->bi_sector, + bdevname(bio->bi_bdev,b)); + } + } + + generic_make_request(bio); +} + +EXPORT_SYMBOL(submit_bio); + +/** + * __end_that_request_first - end I/O on a request + * @req: the request being processed + * @error: 0 for success, < 0 for error + * @nr_bytes: number of bytes to complete + * + * Description: + * Ends I/O on a number of bytes attached to @req, and sets it up + * for the next range of segments (if any) in the cluster. + * + * Return: + * 0 - we are done with this request, call end_that_request_last() + * 1 - still buffers pending for this request + **/ +static int __end_that_request_first(struct request *req, int error, + int nr_bytes) +{ + int total_bytes, bio_nbytes, next_idx = 0; + struct bio *bio; + + blk_add_trace_rq(req->q, req, BLK_TA_COMPLETE); + + /* + * for a REQ_BLOCK_PC request, we want to carry any eventual + * sense key with us all the way through + */ + if (!blk_pc_request(req)) + req->errors = 0; + + if (error) { + if (blk_fs_request(req) && !(req->cmd_flags & REQ_QUIET)) + printk("end_request: I/O error, dev %s, sector %llu\n", + req->rq_disk ? req->rq_disk->disk_name : "?", + (unsigned long long)req->sector); + } + + if (blk_fs_request(req) && req->rq_disk) { + const int rw = rq_data_dir(req); + + disk_stat_add(req->rq_disk, sectors[rw], nr_bytes >> 9); + } + + total_bytes = bio_nbytes = 0; + while ((bio = req->bio) != NULL) { + int nbytes; + + /* + * For an empty barrier request, the low level driver must + * store a potential error location in ->sector. We pass + * that back up in ->bi_sector. + */ + if (blk_empty_barrier(req)) + bio->bi_sector = req->sector; + + if (nr_bytes >= bio->bi_size) { + req->bio = bio->bi_next; + nbytes = bio->bi_size; + req_bio_endio(req, bio, nbytes, error); + next_idx = 0; + bio_nbytes = 0; + } else { + int idx = bio->bi_idx + next_idx; + + if (unlikely(bio->bi_idx >= bio->bi_vcnt)) { + blk_dump_rq_flags(req, "__end_that"); + printk("%s: bio idx %d >= vcnt %d\n", + __FUNCTION__, + bio->bi_idx, bio->bi_vcnt); + break; + } + + nbytes = bio_iovec_idx(bio, idx)->bv_len; + BIO_BUG_ON(nbytes > bio->bi_size); + + /* + * not a complete bvec done + */ + if (unlikely(nbytes > nr_bytes)) { + bio_nbytes += nr_bytes; + total_bytes += nr_bytes; + break; + } + + /* + * advance to the next vector + */ + next_idx++; + bio_nbytes += nbytes; + } + + total_bytes += nbytes; + nr_bytes -= nbytes; + + if ((bio = req->bio)) { + /* + * end more in this run, or just return 'not-done' + */ + if (unlikely(nr_bytes <= 0)) + break; + } + } + + /* + * completely done + */ + if (!req->bio) + return 0; + + /* + * if the request wasn't completed, update state + */ + if (bio_nbytes) { + req_bio_endio(req, bio, bio_nbytes, error); + bio->bi_idx += next_idx; + bio_iovec(bio)->bv_offset += nr_bytes; + bio_iovec(bio)->bv_len -= nr_bytes; + } + + blk_recalc_rq_sectors(req, total_bytes >> 9); + blk_recalc_rq_segments(req); + return 1; +} + +/* + * splice the completion data to a local structure and hand off to + * process_completion_queue() to complete the requests + */ +static void blk_done_softirq(struct softirq_action *h) +{ + struct list_head *cpu_list, local_list; + + local_irq_disable(); + cpu_list = &__get_cpu_var(blk_cpu_done); + list_replace_init(cpu_list, &local_list); + local_irq_enable(); + + while (!list_empty(&local_list)) { + struct request *rq = list_entry(local_list.next, struct request, donelist); + + list_del_init(&rq->donelist); + rq->q->softirq_done_fn(rq); + } +} + +static int __cpuinit blk_cpu_notify(struct notifier_block *self, unsigned long action, + void *hcpu) +{ + /* + * If a CPU goes away, splice its entries to the current CPU + * and trigger a run of the softirq + */ + if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) { + int cpu = (unsigned long) hcpu; + + local_irq_disable(); + list_splice_init(&per_cpu(blk_cpu_done, cpu), + &__get_cpu_var(blk_cpu_done)); + raise_softirq_irqoff(BLOCK_SOFTIRQ); + local_irq_enable(); + } + + return NOTIFY_OK; +} + + +static struct notifier_block blk_cpu_notifier __cpuinitdata = { + .notifier_call = blk_cpu_notify, +}; + +/** + * blk_complete_request - end I/O on a request + * @req: the request being processed + * + * Description: + * Ends all I/O on a request. It does not handle partial completions, + * unless the driver actually implements this in its completion callback + * through requeueing. The actual completion happens out-of-order, + * through a softirq handler. The user must have registered a completion + * callback through blk_queue_softirq_done(). + **/ + +void blk_complete_request(struct request *req) +{ + struct list_head *cpu_list; + unsigned long flags; + + BUG_ON(!req->q->softirq_done_fn); + + local_irq_save(flags); + + cpu_list = &__get_cpu_var(blk_cpu_done); + list_add_tail(&req->donelist, cpu_list); + raise_softirq_irqoff(BLOCK_SOFTIRQ); + + local_irq_restore(flags); +} + +EXPORT_SYMBOL(blk_complete_request); + +/* + * queue lock must be held + */ +static void end_that_request_last(struct request *req, int error) +{ + struct gendisk *disk = req->rq_disk; + + if (blk_rq_tagged(req)) + blk_queue_end_tag(req->q, req); + + if (blk_queued_rq(req)) + blkdev_dequeue_request(req); + + if (unlikely(laptop_mode) && blk_fs_request(req)) + laptop_io_completion(); + + /* + * Account IO completion. bar_rq isn't accounted as a normal + * IO on queueing nor completion. Accounting the containing + * request is enough. + */ + if (disk && blk_fs_request(req) && req != &req->q->bar_rq) { + unsigned long duration = jiffies - req->start_time; + const int rw = rq_data_dir(req); + + __disk_stat_inc(disk, ios[rw]); + __disk_stat_add(disk, ticks[rw], duration); + disk_round_stats(disk); + disk->in_flight--; + } + + if (req->end_io) + req->end_io(req, error); + else { + if (blk_bidi_rq(req)) + __blk_put_request(req->next_rq->q, req->next_rq); + + __blk_put_request(req->q, req); + } +} + +static inline void __end_request(struct request *rq, int uptodate, + unsigned int nr_bytes) +{ + int error = 0; + + if (uptodate <= 0) + error = uptodate ? uptodate : -EIO; + + __blk_end_request(rq, error, nr_bytes); +} + +/** + * blk_rq_bytes - Returns bytes left to complete in the entire request + **/ +unsigned int blk_rq_bytes(struct request *rq) +{ + if (blk_fs_request(rq)) + return rq->hard_nr_sectors << 9; + + return rq->data_len; +} +EXPORT_SYMBOL_GPL(blk_rq_bytes); + +/** + * blk_rq_cur_bytes - Returns bytes left to complete in the current segment + **/ +unsigned int blk_rq_cur_bytes(struct request *rq) +{ + if (blk_fs_request(rq)) + return rq->current_nr_sectors << 9; + + if (rq->bio) + return rq->bio->bi_size; + + return rq->data_len; +} +EXPORT_SYMBOL_GPL(blk_rq_cur_bytes); + +/** + * end_queued_request - end all I/O on a queued request + * @rq: the request being processed + * @uptodate: error value or 0/1 uptodate flag + * + * Description: + * Ends all I/O on a request, and removes it from the block layer queues. + * Not suitable for normal IO completion, unless the driver still has + * the request attached to the block layer. + * + **/ +void end_queued_request(struct request *rq, int uptodate) +{ + __end_request(rq, uptodate, blk_rq_bytes(rq)); +} +EXPORT_SYMBOL(end_queued_request); + +/** + * end_dequeued_request - end all I/O on a dequeued request + * @rq: the request being processed + * @uptodate: error value or 0/1 uptodate flag + * + * Description: + * Ends all I/O on a request. The request must already have been + * dequeued using blkdev_dequeue_request(), as is normally the case + * for most drivers. + * + **/ +void end_dequeued_request(struct request *rq, int uptodate) +{ + __end_request(rq, uptodate, blk_rq_bytes(rq)); +} +EXPORT_SYMBOL(end_dequeued_request); + + +/** + * end_request - end I/O on the current segment of the request + * @req: the request being processed + * @uptodate: error value or 0/1 uptodate flag + * + * Description: + * Ends I/O on the current segment of a request. If that is the only + * remaining segment, the request is also completed and freed. + * + * This is a remnant of how older block drivers handled IO completions. + * Modern drivers typically end IO on the full request in one go, unless + * they have a residual value to account for. For that case this function + * isn't really useful, unless the residual just happens to be the + * full current segment. In other words, don't use this function in new + * code. Either use end_request_completely(), or the + * end_that_request_chunk() (along with end_that_request_last()) for + * partial completions. + * + **/ +void end_request(struct request *req, int uptodate) +{ + __end_request(req, uptodate, req->hard_cur_sectors << 9); +} +EXPORT_SYMBOL(end_request); + +/** + * blk_end_io - Generic end_io function to complete a request. + * @rq: the request being processed + * @error: 0 for success, < 0 for error + * @nr_bytes: number of bytes to complete @rq + * @bidi_bytes: number of bytes to complete @rq->next_rq + * @drv_callback: function called between completion of bios in the request + * and completion of the request. + * If the callback returns non 0, this helper returns without + * completion of the request. + * + * Description: + * Ends I/O on a number of bytes attached to @rq and @rq->next_rq. + * If @rq has leftover, sets it up for the next range of segments. + * + * Return: + * 0 - we are done with this request + * 1 - this request is not freed yet, it still has pending buffers. + **/ +static int blk_end_io(struct request *rq, int error, int nr_bytes, + int bidi_bytes, int (drv_callback)(struct request *)) +{ + struct request_queue *q = rq->q; + unsigned long flags = 0UL; + + if (blk_fs_request(rq) || blk_pc_request(rq)) { + if (__end_that_request_first(rq, error, nr_bytes)) + return 1; + + /* Bidi request must be completed as a whole */ + if (blk_bidi_rq(rq) && + __end_that_request_first(rq->next_rq, error, bidi_bytes)) + return 1; + } + + /* Special feature for tricky drivers */ + if (drv_callback && drv_callback(rq)) + return 1; + + add_disk_randomness(rq->rq_disk); + + spin_lock_irqsave(q->queue_lock, flags); + end_that_request_last(rq, error); + spin_unlock_irqrestore(q->queue_lock, flags); + + return 0; +} + +/** + * blk_end_request - Helper function for drivers to complete the request. + * @rq: the request being processed + * @error: 0 for success, < 0 for error + * @nr_bytes: number of bytes to complete + * + * Description: + * Ends I/O on a number of bytes attached to @rq. + * If @rq has leftover, sets it up for the next range of segments. + * + * Return: + * 0 - we are done with this request + * 1 - still buffers pending for this request + **/ +int blk_end_request(struct request *rq, int error, int nr_bytes) +{ + return blk_end_io(rq, error, nr_bytes, 0, NULL); +} +EXPORT_SYMBOL_GPL(blk_end_request); + +/** + * __blk_end_request - Helper function for drivers to complete the request. + * @rq: the request being processed + * @error: 0 for success, < 0 for error + * @nr_bytes: number of bytes to complete + * + * Description: + * Must be called with queue lock held unlike blk_end_request(). + * + * Return: + * 0 - we are done with this request + * 1 - still buffers pending for this request + **/ +int __blk_end_request(struct request *rq, int error, int nr_bytes) +{ + if (blk_fs_request(rq) || blk_pc_request(rq)) { + if (__end_that_request_first(rq, error, nr_bytes)) + return 1; + } + + add_disk_randomness(rq->rq_disk); + + end_that_request_last(rq, error); + + return 0; +} +EXPORT_SYMBOL_GPL(__blk_end_request); + +/** + * blk_end_bidi_request - Helper function for drivers to complete bidi request. + * @rq: the bidi request being processed + * @error: 0 for success, < 0 for error + * @nr_bytes: number of bytes to complete @rq + * @bidi_bytes: number of bytes to complete @rq->next_rq + * + * Description: + * Ends I/O on a number of bytes attached to @rq and @rq->next_rq. + * + * Return: + * 0 - we are done with this request + * 1 - still buffers pending for this request + **/ +int blk_end_bidi_request(struct request *rq, int error, int nr_bytes, + int bidi_bytes) +{ + return blk_end_io(rq, error, nr_bytes, bidi_bytes, NULL); +} +EXPORT_SYMBOL_GPL(blk_end_bidi_request); + +/** + * blk_end_request_callback - Special helper function for tricky drivers + * @rq: the request being processed + * @error: 0 for success, < 0 for error + * @nr_bytes: number of bytes to complete + * @drv_callback: function called between completion of bios in the request + * and completion of the request. + * If the callback returns non 0, this helper returns without + * completion of the request. + * + * Description: + * Ends I/O on a number of bytes attached to @rq. + * If @rq has leftover, sets it up for the next range of segments. + * + * This special helper function is used only for existing tricky drivers. + * (e.g. cdrom_newpc_intr() of ide-cd) + * This interface will be removed when such drivers are rewritten. + * Don't use this interface in other places anymore. + * + * Return: + * 0 - we are done with this request + * 1 - this request is not freed yet. + * this request still has pending buffers or + * the driver doesn't want to finish this request yet. + **/ +int blk_end_request_callback(struct request *rq, int error, int nr_bytes, + int (drv_callback)(struct request *)) +{ + return blk_end_io(rq, error, nr_bytes, 0, drv_callback); +} +EXPORT_SYMBOL_GPL(blk_end_request_callback); + +void blk_rq_bio_prep(struct request_queue *q, struct request *rq, + struct bio *bio) +{ + /* first two bits are identical in rq->cmd_flags and bio->bi_rw */ + rq->cmd_flags |= (bio->bi_rw & 3); + + rq->nr_phys_segments = bio_phys_segments(q, bio); + rq->nr_hw_segments = bio_hw_segments(q, bio); + rq->current_nr_sectors = bio_cur_sectors(bio); + rq->hard_cur_sectors = rq->current_nr_sectors; + rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio); + rq->buffer = bio_data(bio); + rq->data_len = bio->bi_size; + + rq->bio = rq->biotail = bio; + + if (bio->bi_bdev) + rq->rq_disk = bio->bi_bdev->bd_disk; +} + +int kblockd_schedule_work(struct work_struct *work) +{ + return queue_work(kblockd_workqueue, work); +} + +EXPORT_SYMBOL(kblockd_schedule_work); + +void kblockd_flush_work(struct work_struct *work) +{ + cancel_work_sync(work); +} +EXPORT_SYMBOL(kblockd_flush_work); + +int __init blk_dev_init(void) +{ + int i; + + kblockd_workqueue = create_workqueue("kblockd"); + if (!kblockd_workqueue) + panic("Failed to create kblockd\n"); + + request_cachep = kmem_cache_create("blkdev_requests", + sizeof(struct request), 0, SLAB_PANIC, NULL); + + blk_requestq_cachep = kmem_cache_create("blkdev_queue", + sizeof(struct request_queue), 0, SLAB_PANIC, NULL); + + for_each_possible_cpu(i) + INIT_LIST_HEAD(&per_cpu(blk_cpu_done, i)); + + open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL); + register_hotcpu_notifier(&blk_cpu_notifier); + + return 0; +} + diff --git a/block/blk-exec.c b/block/blk-exec.c new file mode 100644 index 00000000000..ebfb44e959a --- /dev/null +++ b/block/blk-exec.c @@ -0,0 +1,105 @@ +/* + * Functions related to setting various queue properties from drivers + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/bio.h> +#include <linux/blkdev.h> + +#include "blk.h" + +/* + * for max sense size + */ +#include <scsi/scsi_cmnd.h> + +/** + * blk_end_sync_rq - executes a completion event on a request + * @rq: request to complete + * @error: end io status of the request + */ +void blk_end_sync_rq(struct request *rq, int error) +{ + struct completion *waiting = rq->end_io_data; + + rq->end_io_data = NULL; + __blk_put_request(rq->q, rq); + + /* + * complete last, if this is a stack request the process (and thus + * the rq pointer) could be invalid right after this complete() + */ + complete(waiting); +} +EXPORT_SYMBOL(blk_end_sync_rq); + +/** + * blk_execute_rq_nowait - insert a request into queue for execution + * @q: queue to insert the request in + * @bd_disk: matching gendisk + * @rq: request to insert + * @at_head: insert request at head or tail of queue + * @done: I/O completion handler + * + * Description: + * Insert a fully prepared request at the back of the io scheduler queue + * for execution. Don't wait for completion. + */ +void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk, + struct request *rq, int at_head, + rq_end_io_fn *done) +{ + int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK; + + rq->rq_disk = bd_disk; + rq->cmd_flags |= REQ_NOMERGE; + rq->end_io = done; + WARN_ON(irqs_disabled()); + spin_lock_irq(q->queue_lock); + __elv_add_request(q, rq, where, 1); + __generic_unplug_device(q); + spin_unlock_irq(q->queue_lock); +} +EXPORT_SYMBOL_GPL(blk_execute_rq_nowait); + +/** + * blk_execute_rq - insert a request into queue for execution + * @q: queue to insert the request in + * @bd_disk: matching gendisk + * @rq: request to insert + * @at_head: insert request at head or tail of queue + * + * Description: + * Insert a fully prepared request at the back of the io scheduler queue + * for execution and wait for completion. + */ +int blk_execute_rq(struct request_queue *q, struct gendisk *bd_disk, + struct request *rq, int at_head) +{ + DECLARE_COMPLETION_ONSTACK(wait); + char sense[SCSI_SENSE_BUFFERSIZE]; + int err = 0; + + /* + * we need an extra reference to the request, so we can look at + * it after io completion + */ + rq->ref_count++; + + if (!rq->sense) { + memset(sense, 0, sizeof(sense)); + rq->sense = sense; + rq->sense_len = 0; + } + + rq->end_io_data = &wait; + blk_execute_rq_nowait(q, bd_disk, rq, at_head, blk_end_sync_rq); + wait_for_completion(&wait); + + if (rq->errors) + err = -EIO; + + return err; +} + +EXPORT_SYMBOL(blk_execute_rq); diff --git a/block/blk-ioc.c b/block/blk-ioc.c new file mode 100644 index 00000000000..6d1675508eb --- /dev/null +++ b/block/blk-ioc.c @@ -0,0 +1,194 @@ +/* + * Functions related to io context handling + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/bootmem.h> /* for max_pfn/max_low_pfn */ + +#include "blk.h" + +/* + * For io context allocations + */ +static struct kmem_cache *iocontext_cachep; + +static void cfq_dtor(struct io_context *ioc) +{ + struct cfq_io_context *cic[1]; + int r; + + /* + * We don't have a specific key to lookup with, so use the gang + * lookup to just retrieve the first item stored. The cfq exit + * function will iterate the full tree, so any member will do. + */ + r = radix_tree_gang_lookup(&ioc->radix_root, (void **) cic, 0, 1); + if (r > 0) + cic[0]->dtor(ioc); +} + +/* + * IO Context helper functions. put_io_context() returns 1 if there are no + * more users of this io context, 0 otherwise. + */ +int put_io_context(struct io_context *ioc) +{ + if (ioc == NULL) + return 1; + + BUG_ON(atomic_read(&ioc->refcount) == 0); + + if (atomic_dec_and_test(&ioc->refcount)) { + rcu_read_lock(); + if (ioc->aic && ioc->aic->dtor) + ioc->aic->dtor(ioc->aic); + rcu_read_unlock(); + cfq_dtor(ioc); + + kmem_cache_free(iocontext_cachep, ioc); + return 1; + } + return 0; +} +EXPORT_SYMBOL(put_io_context); + +static void cfq_exit(struct io_context *ioc) +{ + struct cfq_io_context *cic[1]; + int r; + + rcu_read_lock(); + /* + * See comment for cfq_dtor() + */ + r = radix_tree_gang_lookup(&ioc->radix_root, (void **) cic, 0, 1); + rcu_read_unlock(); + + if (r > 0) + cic[0]->exit(ioc); +} + +/* Called by the exitting task */ +void exit_io_context(void) +{ + struct io_context *ioc; + + task_lock(current); + ioc = current->io_context; + current->io_context = NULL; + task_unlock(current); + + if (atomic_dec_and_test(&ioc->nr_tasks)) { + if (ioc->aic && ioc->aic->exit) + ioc->aic->exit(ioc->aic); + cfq_exit(ioc); + + put_io_context(ioc); + } +} + +struct io_context *alloc_io_context(gfp_t gfp_flags, int node) +{ + struct io_context *ret; + + ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node); + if (ret) { + atomic_set(&ret->refcount, 1); + atomic_set(&ret->nr_tasks, 1); + spin_lock_init(&ret->lock); + ret->ioprio_changed = 0; + ret->ioprio = 0; + ret->last_waited = jiffies; /* doesn't matter... */ + ret->nr_batch_requests = 0; /* because this is 0 */ + ret->aic = NULL; + INIT_RADIX_TREE(&ret->radix_root, GFP_ATOMIC | __GFP_HIGH); + ret->ioc_data = NULL; + } + + return ret; +} + +/* + * If the current task has no IO context then create one and initialise it. + * Otherwise, return its existing IO context. + * + * This returned IO context doesn't have a specifically elevated refcount, + * but since the current task itself holds a reference, the context can be + * used in general code, so long as it stays within `current` context. + */ +struct io_context *current_io_context(gfp_t gfp_flags, int node) +{ + struct task_struct *tsk = current; + struct io_context *ret; + + ret = tsk->io_context; + if (likely(ret)) + return ret; + + ret = alloc_io_context(gfp_flags, node); + if (ret) { + /* make sure set_task_ioprio() sees the settings above */ + smp_wmb(); + tsk->io_context = ret; + } + + return ret; +} + +/* + * If the current task has no IO context then create one and initialise it. + * If it does have a context, take a ref on it. + * + * This is always called in the context of the task which submitted the I/O. + */ +struct io_context *get_io_context(gfp_t gfp_flags, int node) +{ + struct io_context *ret = NULL; + + /* + * Check for unlikely race with exiting task. ioc ref count is + * zero when ioc is being detached. + */ + do { + ret = current_io_context(gfp_flags, node); + if (unlikely(!ret)) + break; + } while (!atomic_inc_not_zero(&ret->refcount)); + + return ret; +} +EXPORT_SYMBOL(get_io_context); + +void copy_io_context(struct io_context **pdst, struct io_context **psrc) +{ + struct io_context *src = *psrc; + struct io_context *dst = *pdst; + + if (src) { + BUG_ON(atomic_read(&src->refcount) == 0); + atomic_inc(&src->refcount); + put_io_context(dst); + *pdst = src; + } +} +EXPORT_SYMBOL(copy_io_context); + +void swap_io_context(struct io_context **ioc1, struct io_context **ioc2) +{ + struct io_context *temp; + temp = *ioc1; + *ioc1 = *ioc2; + *ioc2 = temp; +} +EXPORT_SYMBOL(swap_io_context); + +int __init blk_ioc_init(void) +{ + iocontext_cachep = kmem_cache_create("blkdev_ioc", + sizeof(struct io_context), 0, SLAB_PANIC, NULL); + return 0; +} +subsys_initcall(blk_ioc_init); diff --git a/block/blk-map.c b/block/blk-map.c new file mode 100644 index 00000000000..916cfc96ffa --- /dev/null +++ b/block/blk-map.c @@ -0,0 +1,264 @@ +/* + * Functions related to mapping data to requests + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/bio.h> +#include <linux/blkdev.h> + +#include "blk.h" + +int blk_rq_append_bio(struct request_queue *q, struct request *rq, + struct bio *bio) +{ + if (!rq->bio) + blk_rq_bio_prep(q, rq, bio); + else if (!ll_back_merge_fn(q, rq, bio)) + return -EINVAL; + else { + rq->biotail->bi_next = bio; + rq->biotail = bio; + + rq->data_len += bio->bi_size; + } + return 0; +} +EXPORT_SYMBOL(blk_rq_append_bio); + +static int __blk_rq_unmap_user(struct bio *bio) +{ + int ret = 0; + + if (bio) { + if (bio_flagged(bio, BIO_USER_MAPPED)) + bio_unmap_user(bio); + else + ret = bio_uncopy_user(bio); + } + + return ret; +} + +static int __blk_rq_map_user(struct request_queue *q, struct request *rq, + void __user *ubuf, unsigned int len) +{ + unsigned long uaddr; + struct bio *bio, *orig_bio; + int reading, ret; + + reading = rq_data_dir(rq) == READ; + + /* + * if alignment requirement is satisfied, map in user pages for + * direct dma. else, set up kernel bounce buffers + */ + uaddr = (unsigned long) ubuf; + if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q))) + bio = bio_map_user(q, NULL, uaddr, len, reading); + else + bio = bio_copy_user(q, uaddr, len, reading); + + if (IS_ERR(bio)) + return PTR_ERR(bio); + + orig_bio = bio; + blk_queue_bounce(q, &bio); + + /* + * We link the bounce buffer in and could have to traverse it + * later so we have to get a ref to prevent it from being freed + */ + bio_get(bio); + + ret = blk_rq_append_bio(q, rq, bio); + if (!ret) + return bio->bi_size; + + /* if it was boucned we must call the end io function */ + bio_endio(bio, 0); + __blk_rq_unmap_user(orig_bio); + bio_put(bio); + return ret; +} + +/** + * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage + * @q: request queue where request should be inserted + * @rq: request structure to fill + * @ubuf: the user buffer + * @len: length of user data + * + * Description: + * Data will be mapped directly for zero copy io, if possible. Otherwise + * a kernel bounce buffer is used. + * + * A matching blk_rq_unmap_user() must be issued at the end of io, while + * still in process context. + * + * Note: The mapped bio may need to be bounced through blk_queue_bounce() + * before being submitted to the device, as pages mapped may be out of + * reach. It's the callers responsibility to make sure this happens. The + * original bio must be passed back in to blk_rq_unmap_user() for proper + * unmapping. + */ +int blk_rq_map_user(struct request_queue *q, struct request *rq, + void __user *ubuf, unsigned long len) +{ + unsigned long bytes_read = 0; + struct bio *bio = NULL; + int ret; + + if (len > (q->max_hw_sectors << 9)) + return -EINVAL; + if (!len || !ubuf) + return -EINVAL; + + while (bytes_read != len) { + unsigned long map_len, end, start; + + map_len = min_t(unsigned long, len - bytes_read, BIO_MAX_SIZE); + end = ((unsigned long)ubuf + map_len + PAGE_SIZE - 1) + >> PAGE_SHIFT; + start = (unsigned long)ubuf >> PAGE_SHIFT; + + /* + * A bad offset could cause us to require BIO_MAX_PAGES + 1 + * pages. If this happens we just lower the requested + * mapping len by a page so that we can fit + */ + if (end - start > BIO_MAX_PAGES) + map_len -= PAGE_SIZE; + + ret = __blk_rq_map_user(q, rq, ubuf, map_len); + if (ret < 0) + goto unmap_rq; + if (!bio) + bio = rq->bio; + bytes_read += ret; + ubuf += ret; + } + + rq->buffer = rq->data = NULL; + return 0; +unmap_rq: + blk_rq_unmap_user(bio); + return ret; +} + +EXPORT_SYMBOL(blk_rq_map_user); + +/** + * blk_rq_map_user_iov - map user data to a request, for REQ_BLOCK_PC usage + * @q: request queue where request should be inserted + * @rq: request to map data to + * @iov: pointer to the iovec + * @iov_count: number of elements in the iovec + * @len: I/O byte count + * + * Description: + * Data will be mapped directly for zero copy io, if possible. Otherwise + * a kernel bounce buffer is used. + * + * A matching blk_rq_unmap_user() must be issued at the end of io, while + * still in process context. + * + * Note: The mapped bio may need to be bounced through blk_queue_bounce() + * before being submitted to the device, as pages mapped may be out of + * reach. It's the callers responsibility to make sure this happens. The + * original bio must be passed back in to blk_rq_unmap_user() for proper + * unmapping. + */ +int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, + struct sg_iovec *iov, int iov_count, unsigned int len) +{ + struct bio *bio; + + if (!iov || iov_count <= 0) + return -EINVAL; + + /* we don't allow misaligned data like bio_map_user() does. If the + * user is using sg, they're expected to know the alignment constraints + * and respect them accordingly */ + bio = bio_map_user_iov(q, NULL, iov, iov_count, rq_data_dir(rq)== READ); + if (IS_ERR(bio)) + return PTR_ERR(bio); + + if (bio->bi_size != len) { + bio_endio(bio, 0); + bio_unmap_user(bio); + return -EINVAL; + } + + bio_get(bio); + blk_rq_bio_prep(q, rq, bio); + rq->buffer = rq->data = NULL; + return 0; +} + +EXPORT_SYMBOL(blk_rq_map_user_iov); + +/** + * blk_rq_unmap_user - unmap a request with user data + * @bio: start of bio list + * + * Description: + * Unmap a rq previously mapped by blk_rq_map_user(). The caller must + * supply the original rq->bio from the blk_rq_map_user() return, since + * the io completion may have changed rq->bio. + */ +int blk_rq_unmap_user(struct bio *bio) +{ + struct bio *mapped_bio; + int ret = 0, ret2; + + while (bio) { + mapped_bio = bio; + if (unlikely(bio_flagged(bio, BIO_BOUNCED))) + mapped_bio = bio->bi_private; + + ret2 = __blk_rq_unmap_user(mapped_bio); + if (ret2 && !ret) + ret = ret2; + + mapped_bio = bio; + bio = bio->bi_next; + bio_put(mapped_bio); + } + + return ret; +} + +EXPORT_SYMBOL(blk_rq_unmap_user); + +/** + * blk_rq_map_kern - map kernel data to a request, for REQ_BLOCK_PC usage + * @q: request queue where request should be inserted + * @rq: request to fill + * @kbuf: the kernel buffer + * @len: length of user data + * @gfp_mask: memory allocation flags + */ +int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, + unsigned int len, gfp_t gfp_mask) +{ + struct bio *bio; + + if (len > (q->max_hw_sectors << 9)) + return -EINVAL; + if (!len || !kbuf) + return -EINVAL; + + bio = bio_map_kern(q, kbuf, len, gfp_mask); + if (IS_ERR(bio)) + return PTR_ERR(bio); + + if (rq_data_dir(rq) == WRITE) + bio->bi_rw |= (1 << BIO_RW); + + blk_rq_bio_prep(q, rq, bio); + blk_queue_bounce(q, &rq->bio); + rq->buffer = rq->data = NULL; + return 0; +} + +EXPORT_SYMBOL(blk_rq_map_kern); diff --git a/block/blk-merge.c b/block/blk-merge.c new file mode 100644 index 00000000000..5023f0b0807 --- /dev/null +++ b/block/blk-merge.c @@ -0,0 +1,485 @@ +/* + * Functions related to segment and merge handling + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/scatterlist.h> + +#include "blk.h" + +void blk_recalc_rq_sectors(struct request *rq, int nsect) +{ + if (blk_fs_request(rq)) { + rq->hard_sector += nsect; + rq->hard_nr_sectors -= nsect; + + /* + * Move the I/O submission pointers ahead if required. + */ + if ((rq->nr_sectors >= rq->hard_nr_sectors) && + (rq->sector <= rq->hard_sector)) { + rq->sector = rq->hard_sector; + rq->nr_sectors = rq->hard_nr_sectors; + rq->hard_cur_sectors = bio_cur_sectors(rq->bio); + rq->current_nr_sectors = rq->hard_cur_sectors; + rq->buffer = bio_data(rq->bio); + } + + /* + * if total number of sectors is less than the first segment + * size, something has gone terribly wrong + */ + if (rq->nr_sectors < rq->current_nr_sectors) { + printk("blk: request botched\n"); + rq->nr_sectors = rq->current_nr_sectors; + } + } +} + +void blk_recalc_rq_segments(struct request *rq) +{ + int nr_phys_segs; + int nr_hw_segs; + unsigned int phys_size; + unsigned int hw_size; + struct bio_vec *bv, *bvprv = NULL; + int seg_size; + int hw_seg_size; + int cluster; + struct req_iterator iter; + int high, highprv = 1; + struct request_queue *q = rq->q; + + if (!rq->bio) + return; + + cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER); + hw_seg_size = seg_size = 0; + phys_size = hw_size = nr_phys_segs = nr_hw_segs = 0; + rq_for_each_segment(bv, rq, iter) { + /* + * the trick here is making sure that a high page is never + * considered part of another segment, since that might + * change with the bounce page. + */ + high = page_to_pfn(bv->bv_page) > q->bounce_pfn; + if (high || highprv) + goto new_hw_segment; + if (cluster) { + if (seg_size + bv->bv_len > q->max_segment_size) + goto new_segment; + if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv)) + goto new_segment; + if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv)) + goto new_segment; + if (BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len)) + goto new_hw_segment; + + seg_size += bv->bv_len; + hw_seg_size += bv->bv_len; + bvprv = bv; + continue; + } +new_segment: + if (BIOVEC_VIRT_MERGEABLE(bvprv, bv) && + !BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len)) + hw_seg_size += bv->bv_len; + else { +new_hw_segment: + if (nr_hw_segs == 1 && + hw_seg_size > rq->bio->bi_hw_front_size) + rq->bio->bi_hw_front_size = hw_seg_size; + hw_seg_size = BIOVEC_VIRT_START_SIZE(bv) + bv->bv_len; + nr_hw_segs++; + } + + nr_phys_segs++; + bvprv = bv; + seg_size = bv->bv_len; + highprv = high; + } + + if (nr_hw_segs == 1 && + hw_seg_size > rq->bio->bi_hw_front_size) + rq->bio->bi_hw_front_size = hw_seg_size; + if (hw_seg_size > rq->biotail->bi_hw_back_size) + rq->biotail->bi_hw_back_size = hw_seg_size; + rq->nr_phys_segments = nr_phys_segs; + rq->nr_hw_segments = nr_hw_segs; +} + +void blk_recount_segments(struct request_queue *q, struct bio *bio) +{ + struct request rq; + struct bio *nxt = bio->bi_next; + rq.q = q; + rq.bio = rq.biotail = bio; + bio->bi_next = NULL; + blk_recalc_rq_segments(&rq); + bio->bi_next = nxt; + bio->bi_phys_segments = rq.nr_phys_segments; + bio->bi_hw_segments = rq.nr_hw_segments; + bio->bi_flags |= (1 << BIO_SEG_VALID); +} +EXPORT_SYMBOL(blk_recount_segments); + +static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio, + struct bio *nxt) +{ + if (!(q->queue_flags & (1 << QUEUE_FLAG_CLUSTER))) + return 0; + + if (!BIOVEC_PHYS_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt))) + return 0; + if (bio->bi_size + nxt->bi_size > q->max_segment_size) + return 0; + + /* + * bio and nxt are contigous in memory, check if the queue allows + * these two to be merged into one + */ + if (BIO_SEG_BOUNDARY(q, bio, nxt)) + return 1; + + return 0; +} + +static int blk_hw_contig_segment(struct request_queue *q, struct bio *bio, + struct bio *nxt) +{ + if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) + blk_recount_segments(q, bio); + if (unlikely(!bio_flagged(nxt, BIO_SEG_VALID))) + blk_recount_segments(q, nxt); + if (!BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)) || + BIOVEC_VIRT_OVERSIZE(bio->bi_hw_back_size + nxt->bi_hw_front_size)) + return 0; + if (bio->bi_hw_back_size + nxt->bi_hw_front_size > q->max_segment_size) + return 0; + + return 1; +} + +/* + * map a request to scatterlist, return number of sg entries setup. Caller + * must make sure sg can hold rq->nr_phys_segments entries + */ +int blk_rq_map_sg(struct request_queue *q, struct request *rq, + struct scatterlist *sglist) +{ + struct bio_vec *bvec, *bvprv; + struct req_iterator iter; + struct scatterlist *sg; + int nsegs, cluster; + + nsegs = 0; + cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER); + + /* + * for each bio in rq + */ + bvprv = NULL; + sg = NULL; + rq_for_each_segment(bvec, rq, iter) { + int nbytes = bvec->bv_len; + + if (bvprv && cluster) { + if (sg->length + nbytes > q->max_segment_size) + goto new_segment; + + if (!BIOVEC_PHYS_MERGEABLE(bvprv, bvec)) + goto new_segment; + if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bvec)) + goto new_segment; + + sg->length += nbytes; + } else { +new_segment: + if (!sg) + sg = sglist; + else { + /* + * If the driver previously mapped a shorter + * list, we could see a termination bit + * prematurely unless it fully inits the sg + * table on each mapping. We KNOW that there + * must be more entries here or the driver + * would be buggy, so force clear the + * termination bit to avoid doing a full + * sg_init_table() in drivers for each command. + */ + sg->page_link &= ~0x02; + sg = sg_next(sg); + } + + sg_set_page(sg, bvec->bv_page, nbytes, bvec->bv_offset); + nsegs++; + } + bvprv = bvec; + } /* segments in rq */ + + if (q->dma_drain_size) { + sg->page_link &= ~0x02; + sg = sg_next(sg); + sg_set_page(sg, virt_to_page(q->dma_drain_buffer), + q->dma_drain_size, + ((unsigned long)q->dma_drain_buffer) & + (PAGE_SIZE - 1)); + nsegs++; + } + + if (sg) + sg_mark_end(sg); + + return nsegs; +} + +EXPORT_SYMBOL(blk_rq_map_sg); + +static inline int ll_new_mergeable(struct request_queue *q, + struct request *req, + struct bio *bio) +{ + int nr_phys_segs = bio_phys_segments(q, bio); + + if (req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) { + req->cmd_flags |= REQ_NOMERGE; + if (req == q->last_merge) + q->last_merge = NULL; + return 0; + } + + /* + * A hw segment is just getting larger, bump just the phys + * counter. + */ + req->nr_phys_segments += nr_phys_segs; + return 1; +} + +static inline int ll_new_hw_segment(struct request_queue *q, + struct request *req, + struct bio *bio) +{ + int nr_hw_segs = bio_hw_segments(q, bio); + int nr_phys_segs = bio_phys_segments(q, bio); + + if (req->nr_hw_segments + nr_hw_segs > q->max_hw_segments + || req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) { + req->cmd_flags |= REQ_NOMERGE; + if (req == q->last_merge) + q->last_merge = NULL; + return 0; + } + + /* + * This will form the start of a new hw segment. Bump both + * counters. + */ + req->nr_hw_segments += nr_hw_segs; + req->nr_phys_segments += nr_phys_segs; + return 1; +} + +int ll_back_merge_fn(struct request_queue *q, struct request *req, + struct bio *bio) +{ + unsigned short max_sectors; + int len; + + if (unlikely(blk_pc_request(req))) + max_sectors = q->max_hw_sectors; + else + max_sectors = q->max_sectors; + + if (req->nr_sectors + bio_sectors(bio) > max_sectors) { + req->cmd_flags |= REQ_NOMERGE; + if (req == q->last_merge) + q->last_merge = NULL; + return 0; + } + if (unlikely(!bio_flagged(req->biotail, BIO_SEG_VALID))) + blk_recount_segments(q, req->biotail); + if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) + blk_recount_segments(q, bio); + len = req->biotail->bi_hw_back_size + bio->bi_hw_front_size; + if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(req->biotail), __BVEC_START(bio)) && + !BIOVEC_VIRT_OVERSIZE(len)) { + int mergeable = ll_new_mergeable(q, req, bio); + + if (mergeable) { + if (req->nr_hw_segments == 1) + req->bio->bi_hw_front_size = len; + if (bio->bi_hw_segments == 1) + bio->bi_hw_back_size = len; + } + return mergeable; + } + + return ll_new_hw_segment(q, req, bio); +} + +int ll_front_merge_fn(struct request_queue *q, struct request *req, + struct bio *bio) +{ + unsigned short max_sectors; + int len; + + if (unlikely(blk_pc_request(req))) + max_sectors = q->max_hw_sectors; + else + max_sectors = q->max_sectors; + + + if (req->nr_sectors + bio_sectors(bio) > max_sectors) { + req->cmd_flags |= REQ_NOMERGE; + if (req == q->last_merge) + q->last_merge = NULL; + return 0; + } + len = bio->bi_hw_back_size + req->bio->bi_hw_front_size; + if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) + blk_recount_segments(q, bio); + if (unlikely(!bio_flagged(req->bio, BIO_SEG_VALID))) + blk_recount_segments(q, req->bio); + if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(req->bio)) && + !BIOVEC_VIRT_OVERSIZE(len)) { + int mergeable = ll_new_mergeable(q, req, bio); + + if (mergeable) { + if (bio->bi_hw_segments == 1) + bio->bi_hw_front_size = len; + if (req->nr_hw_segments == 1) + req->biotail->bi_hw_back_size = len; + } + return mergeable; + } + + return ll_new_hw_segment(q, req, bio); +} + +static int ll_merge_requests_fn(struct request_queue *q, struct request *req, + struct request *next) +{ + int total_phys_segments; + int total_hw_segments; + + /* + * First check if the either of the requests are re-queued + * requests. Can't merge them if they are. + */ + if (req->special || next->special) + return 0; + + /* + * Will it become too large? + */ + if ((req->nr_sectors + next->nr_sectors) > q->max_sectors) + return 0; + + total_phys_segments = req->nr_phys_segments + next->nr_phys_segments; + if (blk_phys_contig_segment(q, req->biotail, next->bio)) + total_phys_segments--; + + if (total_phys_segments > q->max_phys_segments) + return 0; + + total_hw_segments = req->nr_hw_segments + next->nr_hw_segments; + if (blk_hw_contig_segment(q, req->biotail, next->bio)) { + int len = req->biotail->bi_hw_back_size + next->bio->bi_hw_front_size; + /* + * propagate the combined length to the end of the requests + */ + if (req->nr_hw_segments == 1) + req->bio->bi_hw_front_size = len; + if (next->nr_hw_segments == 1) + next->biotail->bi_hw_back_size = len; + total_hw_segments--; + } + + if (total_hw_segments > q->max_hw_segments) + return 0; + + /* Merge is OK... */ + req->nr_phys_segments = total_phys_segments; + req->nr_hw_segments = total_hw_segments; + return 1; +} + +/* + * Has to be called with the request spinlock acquired + */ +static int attempt_merge(struct request_queue *q, struct request *req, + struct request *next) +{ + if (!rq_mergeable(req) || !rq_mergeable(next)) + return 0; + + /* + * not contiguous + */ + if (req->sector + req->nr_sectors != next->sector) + return 0; + + if (rq_data_dir(req) != rq_data_dir(next) + || req->rq_disk != next->rq_disk + || next->special) + return 0; + + /* + * If we are allowed to merge, then append bio list + * from next to rq and release next. merge_requests_fn + * will have updated segment counts, update sector + * counts here. + */ + if (!ll_merge_requests_fn(q, req, next)) + return 0; + + /* + * At this point we have either done a back merge + * or front merge. We need the smaller start_time of + * the merged requests to be the current request + * for accounting purposes. + */ + if (time_after(req->start_time, next->start_time)) + req->start_time = next->start_time; + + req->biotail->bi_next = next->bio; + req->biotail = next->biotail; + + req->nr_sectors = req->hard_nr_sectors += next->hard_nr_sectors; + + elv_merge_requests(q, req, next); + + if (req->rq_disk) { + disk_round_stats(req->rq_disk); + req->rq_disk->in_flight--; + } + + req->ioprio = ioprio_best(req->ioprio, next->ioprio); + + __blk_put_request(q, next); + return 1; +} + +int attempt_back_merge(struct request_queue *q, struct request *rq) +{ + struct request *next = elv_latter_request(q, rq); + + if (next) + return attempt_merge(q, rq, next); + + return 0; +} + +int attempt_front_merge(struct request_queue *q, struct request *rq) +{ + struct request *prev = elv_former_request(q, rq); + + if (prev) + return attempt_merge(q, prev, rq); + + return 0; +} diff --git a/block/blk-settings.c b/block/blk-settings.c new file mode 100644 index 00000000000..4df09a1b8f4 --- /dev/null +++ b/block/blk-settings.c @@ -0,0 +1,402 @@ +/* + * Functions related to setting various queue properties from drivers + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/bootmem.h> /* for max_pfn/max_low_pfn */ + +#include "blk.h" + +unsigned long blk_max_low_pfn, blk_max_pfn; +EXPORT_SYMBOL(blk_max_low_pfn); +EXPORT_SYMBOL(blk_max_pfn); + +/** + * blk_queue_prep_rq - set a prepare_request function for queue + * @q: queue + * @pfn: prepare_request function + * + * It's possible for a queue to register a prepare_request callback which + * is invoked before the request is handed to the request_fn. The goal of + * the function is to prepare a request for I/O, it can be used to build a + * cdb from the request data for instance. + * + */ +void blk_queue_prep_rq(struct request_queue *q, prep_rq_fn *pfn) +{ + q->prep_rq_fn = pfn; +} + +EXPORT_SYMBOL(blk_queue_prep_rq); + +/** + * blk_queue_merge_bvec - set a merge_bvec function for queue + * @q: queue + * @mbfn: merge_bvec_fn + * + * Usually queues have static limitations on the max sectors or segments that + * we can put in a request. Stacking drivers may have some settings that + * are dynamic, and thus we have to query the queue whether it is ok to + * add a new bio_vec to a bio at a given offset or not. If the block device + * has such limitations, it needs to register a merge_bvec_fn to control + * the size of bio's sent to it. Note that a block device *must* allow a + * single page to be added to an empty bio. The block device driver may want + * to use the bio_split() function to deal with these bio's. By default + * no merge_bvec_fn is defined for a queue, and only the fixed limits are + * honored. + */ +void blk_queue_merge_bvec(struct request_queue *q, merge_bvec_fn *mbfn) +{ + q->merge_bvec_fn = mbfn; +} + +EXPORT_SYMBOL(blk_queue_merge_bvec); + +void blk_queue_softirq_done(struct request_queue *q, softirq_done_fn *fn) +{ + q->softirq_done_fn = fn; +} + +EXPORT_SYMBOL(blk_queue_softirq_done); + +/** + * blk_queue_make_request - define an alternate make_request function for a device + * @q: the request queue for the device to be affected + * @mfn: the alternate make_request function + * + * Description: + * The normal way for &struct bios to be passed to a device + * driver is for them to be collected into requests on a request + * queue, and then to allow the device driver to select requests + * off that queue when it is ready. This works well for many block + * devices. However some block devices (typically virtual devices + * such as md or lvm) do not benefit from the processing on the + * request queue, and are served best by having the requests passed + * directly to them. This can be achieved by providing a function + * to blk_queue_make_request(). + * + * Caveat: + * The driver that does this *must* be able to deal appropriately + * with buffers in "highmemory". This can be accomplished by either calling + * __bio_kmap_atomic() to get a temporary kernel mapping, or by calling + * blk_queue_bounce() to create a buffer in normal memory. + **/ +void blk_queue_make_request(struct request_queue * q, make_request_fn * mfn) +{ + /* + * set defaults + */ + q->nr_requests = BLKDEV_MAX_RQ; + blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS); + blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS); + q->make_request_fn = mfn; + q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; + q->backing_dev_info.state = 0; + q->backing_dev_info.capabilities = BDI_CAP_MAP_COPY; + blk_queue_max_sectors(q, SAFE_MAX_SECTORS); + blk_queue_hardsect_size(q, 512); + blk_queue_dma_alignment(q, 511); + blk_queue_congestion_threshold(q); + q->nr_batching = BLK_BATCH_REQ; + + q->unplug_thresh = 4; /* hmm */ + q->unplug_delay = (3 * HZ) / 1000; /* 3 milliseconds */ + if (q->unplug_delay == 0) + q->unplug_delay = 1; + + INIT_WORK(&q->unplug_work, blk_unplug_work); + + q->unplug_timer.function = blk_unplug_timeout; + q->unplug_timer.data = (unsigned long)q; + + /* + * by default assume old behaviour and bounce for any highmem page + */ + blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH); +} + +EXPORT_SYMBOL(blk_queue_make_request); + +/** + * blk_queue_bounce_limit - set bounce buffer limit for queue + * @q: the request queue for the device + * @dma_addr: bus address limit + * + * Description: + * Different hardware can have different requirements as to what pages + * it can do I/O directly to. A low level driver can call + * blk_queue_bounce_limit to have lower memory pages allocated as bounce + * buffers for doing I/O to pages residing above @page. + **/ +void blk_queue_bounce_limit(struct request_queue *q, u64 dma_addr) +{ + unsigned long bounce_pfn = dma_addr >> PAGE_SHIFT; + int dma = 0; + + q->bounce_gfp = GFP_NOIO; +#if BITS_PER_LONG == 64 + /* Assume anything <= 4GB can be handled by IOMMU. + Actually some IOMMUs can handle everything, but I don't + know of a way to test this here. */ + if (bounce_pfn < (min_t(u64,0xffffffff,BLK_BOUNCE_HIGH) >> PAGE_SHIFT)) + dma = 1; + q->bounce_pfn = max_low_pfn; +#else + if (bounce_pfn < blk_max_low_pfn) + dma = 1; + q->bounce_pfn = bounce_pfn; +#endif + if (dma) { + init_emergency_isa_pool(); + q->bounce_gfp = GFP_NOIO | GFP_DMA; + q->bounce_pfn = bounce_pfn; + } +} + +EXPORT_SYMBOL(blk_queue_bounce_limit); + +/** + * blk_queue_max_sectors - set max sectors for a request for this queue + * @q: the request queue for the device + * @max_sectors: max sectors in the usual 512b unit + * + * Description: + * Enables a low level driver to set an upper limit on the size of + * received requests. + **/ +void blk_queue_max_sectors(struct request_queue *q, unsigned int max_sectors) +{ + if ((max_sectors << 9) < PAGE_CACHE_SIZE) { + max_sectors = 1 << (PAGE_CACHE_SHIFT - 9); + printk("%s: set to minimum %d\n", __FUNCTION__, max_sectors); + } + + if (BLK_DEF_MAX_SECTORS > max_sectors) + q->max_hw_sectors = q->max_sectors = max_sectors; + else { + q->max_sectors = BLK_DEF_MAX_SECTORS; + q->max_hw_sectors = max_sectors; + } +} + +EXPORT_SYMBOL(blk_queue_max_sectors); + +/** + * blk_queue_max_phys_segments - set max phys segments for a request for this queue + * @q: the request queue for the device + * @max_segments: max number of segments + * + * Description: + * Enables a low level driver to set an upper limit on the number of + * physical data segments in a request. This would be the largest sized + * scatter list the driver could handle. + **/ +void blk_queue_max_phys_segments(struct request_queue *q, + unsigned short max_segments) +{ + if (!max_segments) { + max_segments = 1; + printk("%s: set to minimum %d\n", __FUNCTION__, max_segments); + } + + q->max_phys_segments = max_segments; +} + +EXPORT_SYMBOL(blk_queue_max_phys_segments); + +/** + * blk_queue_max_hw_segments - set max hw segments for a request for this queue + * @q: the request queue for the device + * @max_segments: max number of segments + * + * Description: + * Enables a low level driver to set an upper limit on the number of + * hw data segments in a request. This would be the largest number of + * address/length pairs the host adapter can actually give as once + * to the device. + **/ +void blk_queue_max_hw_segments(struct request_queue *q, + unsigned short max_segments) +{ + if (!max_segments) { + max_segments = 1; + printk("%s: set to minimum %d\n", __FUNCTION__, max_segments); + } + + q->max_hw_segments = max_segments; +} + +EXPORT_SYMBOL(blk_queue_max_hw_segments); + +/** + * blk_queue_max_segment_size - set max segment size for blk_rq_map_sg + * @q: the request queue for the device + * @max_size: max size of segment in bytes + * + * Description: + * Enables a low level driver to set an upper limit on the size of a + * coalesced segment + **/ +void blk_queue_max_segment_size(struct request_queue *q, unsigned int max_size) +{ + if (max_size < PAGE_CACHE_SIZE) { + max_size = PAGE_CACHE_SIZE; + printk("%s: set to minimum %d\n", __FUNCTION__, max_size); + } + + q->max_segment_size = max_size; +} + +EXPORT_SYMBOL(blk_queue_max_segment_size); + +/** + * blk_queue_hardsect_size - set hardware sector size for the queue + * @q: the request queue for the device + * @size: the hardware sector size, in bytes + * + * Description: + * This should typically be set to the lowest possible sector size + * that the hardware can operate on (possible without reverting to + * even internal read-modify-write operations). Usually the default + * of 512 covers most hardware. + **/ +void blk_queue_hardsect_size(struct request_queue *q, unsigned short size) +{ + q->hardsect_size = size; +} + +EXPORT_SYMBOL(blk_queue_hardsect_size); + +/* + * Returns the minimum that is _not_ zero, unless both are zero. + */ +#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r)) + +/** + * blk_queue_stack_limits - inherit underlying queue limits for stacked drivers + * @t: the stacking driver (top) + * @b: the underlying device (bottom) + **/ +void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b) +{ + /* zero is "infinity" */ + t->max_sectors = min_not_zero(t->max_sectors,b->max_sectors); + t->max_hw_sectors = min_not_zero(t->max_hw_sectors,b->max_hw_sectors); + + t->max_phys_segments = min(t->max_phys_segments,b->max_phys_segments); + t->max_hw_segments = min(t->max_hw_segments,b->max_hw_segments); + t->max_segment_size = min(t->max_segment_size,b->max_segment_size); + t->hardsect_size = max(t->hardsect_size,b->hardsect_size); + if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags)) + clear_bit(QUEUE_FLAG_CLUSTER, &t->queue_flags); +} + +EXPORT_SYMBOL(blk_queue_stack_limits); + +/** + * blk_queue_dma_drain - Set up a drain buffer for excess dma. + * + * @q: the request queue for the device + * @buf: physically contiguous buffer + * @size: size of the buffer in bytes + * + * Some devices have excess DMA problems and can't simply discard (or + * zero fill) the unwanted piece of the transfer. They have to have a + * real area of memory to transfer it into. The use case for this is + * ATAPI devices in DMA mode. If the packet command causes a transfer + * bigger than the transfer size some HBAs will lock up if there + * aren't DMA elements to contain the excess transfer. What this API + * does is adjust the queue so that the buf is always appended + * silently to the scatterlist. + * + * Note: This routine adjusts max_hw_segments to make room for + * appending the drain buffer. If you call + * blk_queue_max_hw_segments() or blk_queue_max_phys_segments() after + * calling this routine, you must set the limit to one fewer than your + * device can support otherwise there won't be room for the drain + * buffer. + */ +int blk_queue_dma_drain(struct request_queue *q, void *buf, + unsigned int size) +{ + if (q->max_hw_segments < 2 || q->max_phys_segments < 2) + return -EINVAL; + /* make room for appending the drain */ + --q->max_hw_segments; + --q->max_phys_segments; + q->dma_drain_buffer = buf; + q->dma_drain_size = size; + + return 0; +} + +EXPORT_SYMBOL_GPL(blk_queue_dma_drain); + +/** + * blk_queue_segment_boundary - set boundary rules for segment merging + * @q: the request queue for the device + * @mask: the memory boundary mask + **/ +void blk_queue_segment_boundary(struct request_queue *q, unsigned long mask) +{ + if (mask < PAGE_CACHE_SIZE - 1) { + mask = PAGE_CACHE_SIZE - 1; + printk("%s: set to minimum %lx\n", __FUNCTION__, mask); + } + + q->seg_boundary_mask = mask; +} + +EXPORT_SYMBOL(blk_queue_segment_boundary); + +/** + * blk_queue_dma_alignment - set dma length and memory alignment + * @q: the request queue for the device + * @mask: alignment mask + * + * description: + * set required memory and length aligment for direct dma transactions. + * this is used when buiding direct io requests for the queue. + * + **/ +void blk_queue_dma_alignment(struct request_queue *q, int mask) +{ + q->dma_alignment = mask; +} + +EXPORT_SYMBOL(blk_queue_dma_alignment); + +/** + * blk_queue_update_dma_alignment - update dma length and memory alignment + * @q: the request queue for the device + * @mask: alignment mask + * + * description: + * update required memory and length aligment for direct dma transactions. + * If the requested alignment is larger than the current alignment, then + * the current queue alignment is updated to the new value, otherwise it + * is left alone. The design of this is to allow multiple objects + * (driver, device, transport etc) to set their respective + * alignments without having them interfere. + * + **/ +void blk_queue_update_dma_alignment(struct request_queue *q, int mask) +{ + BUG_ON(mask > PAGE_SIZE); + + if (mask > q->dma_alignment) + q->dma_alignment = mask; +} + +EXPORT_SYMBOL(blk_queue_update_dma_alignment); + +int __init blk_settings_init(void) +{ + blk_max_low_pfn = max_low_pfn - 1; + blk_max_pfn = max_pfn - 1; + return 0; +} +subsys_initcall(blk_settings_init); diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c new file mode 100644 index 00000000000..bc28776ba76 --- /dev/null +++ b/block/blk-sysfs.c @@ -0,0 +1,309 @@ +/* + * Functions related to sysfs handling + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/blktrace_api.h> + +#include "blk.h" + +struct queue_sysfs_entry { + struct attribute attr; + ssize_t (*show)(struct request_queue *, char *); + ssize_t (*store)(struct request_queue *, const char *, size_t); +}; + +static ssize_t +queue_var_show(unsigned int var, char *page) +{ + return sprintf(page, "%d\n", var); +} + +static ssize_t +queue_var_store(unsigned long *var, const char *page, size_t count) +{ + char *p = (char *) page; + + *var = simple_strtoul(p, &p, 10); + return count; +} + +static ssize_t queue_requests_show(struct request_queue *q, char *page) +{ + return queue_var_show(q->nr_requests, (page)); +} + +static ssize_t +queue_requests_store(struct request_queue *q, const char *page, size_t count) +{ + struct request_list *rl = &q->rq; + unsigned long nr; + int ret = queue_var_store(&nr, page, count); + if (nr < BLKDEV_MIN_RQ) + nr = BLKDEV_MIN_RQ; + + spin_lock_irq(q->queue_lock); + q->nr_requests = nr; + blk_queue_congestion_threshold(q); + + if (rl->count[READ] >= queue_congestion_on_threshold(q)) + blk_set_queue_congested(q, READ); + else if (rl->count[READ] < queue_congestion_off_threshold(q)) + blk_clear_queue_congested(q, READ); + + if (rl->count[WRITE] >= queue_congestion_on_threshold(q)) + blk_set_queue_congested(q, WRITE); + else if (rl->count[WRITE] < queue_congestion_off_threshold(q)) + blk_clear_queue_congested(q, WRITE); + + if (rl->count[READ] >= q->nr_requests) { + blk_set_queue_full(q, READ); + } else if (rl->count[READ]+1 <= q->nr_requests) { + blk_clear_queue_full(q, READ); + wake_up(&rl->wait[READ]); + } + + if (rl->count[WRITE] >= q->nr_requests) { + blk_set_queue_full(q, WRITE); + } else if (rl->count[WRITE]+1 <= q->nr_requests) { + blk_clear_queue_full(q, WRITE); + wake_up(&rl->wait[WRITE]); + } + spin_unlock_irq(q->queue_lock); + return ret; +} + +static ssize_t queue_ra_show(struct request_queue *q, char *page) +{ + int ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10); + + return queue_var_show(ra_kb, (page)); +} + +static ssize_t +queue_ra_store(struct request_queue *q, const char *page, size_t count) +{ + unsigned long ra_kb; + ssize_t ret = queue_var_store(&ra_kb, page, count); + + spin_lock_irq(q->queue_lock); + q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10); + spin_unlock_irq(q->queue_lock); + + return ret; +} + +static ssize_t queue_max_sectors_show(struct request_queue *q, char *page) +{ + int max_sectors_kb = q->max_sectors >> 1; + + return queue_var_show(max_sectors_kb, (page)); +} + +static ssize_t queue_hw_sector_size_show(struct request_queue *q, char *page) +{ + return queue_var_show(q->hardsect_size, page); +} + +static ssize_t +queue_max_sectors_store(struct request_queue *q, const char *page, size_t count) +{ + unsigned long max_sectors_kb, + max_hw_sectors_kb = q->max_hw_sectors >> 1, + page_kb = 1 << (PAGE_CACHE_SHIFT - 10); + ssize_t ret = queue_var_store(&max_sectors_kb, page, count); + + if (max_sectors_kb > max_hw_sectors_kb || max_sectors_kb < page_kb) + return -EINVAL; + /* + * Take the queue lock to update the readahead and max_sectors + * values synchronously: + */ + spin_lock_irq(q->queue_lock); + q->max_sectors = max_sectors_kb << 1; + spin_unlock_irq(q->queue_lock); + + return ret; +} + +static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page) +{ + int max_hw_sectors_kb = q->max_hw_sectors >> 1; + + return queue_var_show(max_hw_sectors_kb, (page)); +} + + +static struct queue_sysfs_entry queue_requests_entry = { + .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR }, + .show = queue_requests_show, + .store = queue_requests_store, +}; + +static struct queue_sysfs_entry queue_ra_entry = { + .attr = {.name = "read_ahead_kb", .mode = S_IRUGO | S_IWUSR }, + .show = queue_ra_show, + .store = queue_ra_store, +}; + +static struct queue_sysfs_entry queue_max_sectors_entry = { + .attr = {.name = "max_sectors_kb", .mode = S_IRUGO | S_IWUSR }, + .show = queue_max_sectors_show, + .store = queue_max_sectors_store, +}; + +static struct queue_sysfs_entry queue_max_hw_sectors_entry = { + .attr = {.name = "max_hw_sectors_kb", .mode = S_IRUGO }, + .show = queue_max_hw_sectors_show, +}; + +static struct queue_sysfs_entry queue_iosched_entry = { + .attr = {.name = "scheduler", .mode = S_IRUGO | S_IWUSR }, + .show = elv_iosched_show, + .store = elv_iosched_store, +}; + +static struct queue_sysfs_entry queue_hw_sector_size_entry = { + .attr = {.name = "hw_sector_size", .mode = S_IRUGO }, + .show = queue_hw_sector_size_show, +}; + +static struct attribute *default_attrs[] = { + &queue_requests_entry.attr, + &queue_ra_entry.attr, + &queue_max_hw_sectors_entry.attr, + &queue_max_sectors_entry.attr, + &queue_iosched_entry.attr, + &queue_hw_sector_size_entry.attr, + NULL, +}; + +#define to_queue(atr) container_of((atr), struct queue_sysfs_entry, attr) + +static ssize_t +queue_attr_show(struct kobject *kobj, struct attribute *attr, char *page) +{ + struct queue_sysfs_entry *entry = to_queue(attr); + struct request_queue *q = + container_of(kobj, struct request_queue, kobj); + ssize_t res; + + if (!entry->show) + return -EIO; + mutex_lock(&q->sysfs_lock); + if (test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)) { + mutex_unlock(&q->sysfs_lock); + return -ENOENT; + } + res = entry->show(q, page); + mutex_unlock(&q->sysfs_lock); + return res; +} + +static ssize_t +queue_attr_store(struct kobject *kobj, struct attribute *attr, + const char *page, size_t length) +{ + struct queue_sysfs_entry *entry = to_queue(attr); + struct request_queue *q = container_of(kobj, struct request_queue, kobj); + + ssize_t res; + + if (!entry->store) + return -EIO; + mutex_lock(&q->sysfs_lock); + if (test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)) { + mutex_unlock(&q->sysfs_lock); + return -ENOENT; + } + res = entry->store(q, page, length); + mutex_unlock(&q->sysfs_lock); + return res; +} + +/** + * blk_cleanup_queue: - release a &struct request_queue when it is no longer needed + * @kobj: the kobj belonging of the request queue to be released + * + * Description: + * blk_cleanup_queue is the pair to blk_init_queue() or + * blk_queue_make_request(). It should be called when a request queue is + * being released; typically when a block device is being de-registered. + * Currently, its primary task it to free all the &struct request + * structures that were allocated to the queue and the queue itself. + * + * Caveat: + * Hopefully the low level driver will have finished any + * outstanding requests first... + **/ +static void blk_release_queue(struct kobject *kobj) +{ + struct request_queue *q = + container_of(kobj, struct request_queue, kobj); + struct request_list *rl = &q->rq; + + blk_sync_queue(q); + + if (rl->rq_pool) + mempool_destroy(rl->rq_pool); + + if (q->queue_tags) + __blk_queue_free_tags(q); + + blk_trace_shutdown(q); + + bdi_destroy(&q->backing_dev_info); + kmem_cache_free(blk_requestq_cachep, q); +} + +static struct sysfs_ops queue_sysfs_ops = { + .show = queue_attr_show, + .store = queue_attr_store, +}; + +struct kobj_type blk_queue_ktype = { + .sysfs_ops = &queue_sysfs_ops, + .default_attrs = default_attrs, + .release = blk_release_queue, +}; + +int blk_register_queue(struct gendisk *disk) +{ + int ret; + + struct request_queue *q = disk->queue; + + if (!q || !q->request_fn) + return -ENXIO; + + ret = kobject_add(&q->kobj, kobject_get(&disk->dev.kobj), + "%s", "queue"); + if (ret < 0) + return ret; + + kobject_uevent(&q->kobj, KOBJ_ADD); + + ret = elv_register_queue(q); + if (ret) { + kobject_uevent(&q->kobj, KOBJ_REMOVE); + kobject_del(&q->kobj); + return ret; + } + + return 0; +} + +void blk_unregister_queue(struct gendisk *disk) +{ + struct request_queue *q = disk->queue; + + if (q && q->request_fn) { + elv_unregister_queue(q); + + kobject_uevent(&q->kobj, KOBJ_REMOVE); + kobject_del(&q->kobj); + kobject_put(&disk->dev.kobj); + } +} diff --git a/block/blk-tag.c b/block/blk-tag.c new file mode 100644 index 00000000000..d1fd300e8ae --- /dev/null +++ b/block/blk-tag.c @@ -0,0 +1,396 @@ +/* + * Functions related to tagged command queuing + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/bio.h> +#include <linux/blkdev.h> + +/** + * blk_queue_find_tag - find a request by its tag and queue + * @q: The request queue for the device + * @tag: The tag of the request + * + * Notes: + * Should be used when a device returns a tag and you want to match + * it with a request. + * + * no locks need be held. + **/ +struct request *blk_queue_find_tag(struct request_queue *q, int tag) +{ + return blk_map_queue_find_tag(q->queue_tags, tag); +} + +EXPORT_SYMBOL(blk_queue_find_tag); + +/** + * __blk_free_tags - release a given set of tag maintenance info + * @bqt: the tag map to free + * + * Tries to free the specified @bqt@. Returns true if it was + * actually freed and false if there are still references using it + */ +static int __blk_free_tags(struct blk_queue_tag *bqt) +{ + int retval; + + retval = atomic_dec_and_test(&bqt->refcnt); + if (retval) { + BUG_ON(bqt->busy); + + kfree(bqt->tag_index); + bqt->tag_index = NULL; + + kfree(bqt->tag_map); + bqt->tag_map = NULL; + + kfree(bqt); + } + + return retval; +} + +/** + * __blk_queue_free_tags - release tag maintenance info + * @q: the request queue for the device + * + * Notes: + * blk_cleanup_queue() will take care of calling this function, if tagging + * has been used. So there's no need to call this directly. + **/ +void __blk_queue_free_tags(struct request_queue *q) +{ + struct blk_queue_tag *bqt = q->queue_tags; + + if (!bqt) + return; + + __blk_free_tags(bqt); + + q->queue_tags = NULL; + q->queue_flags &= ~(1 << QUEUE_FLAG_QUEUED); +} + +/** + * blk_free_tags - release a given set of tag maintenance info + * @bqt: the tag map to free + * + * For externally managed @bqt@ frees the map. Callers of this + * function must guarantee to have released all the queues that + * might have been using this tag map. + */ +void blk_free_tags(struct blk_queue_tag *bqt) +{ + if (unlikely(!__blk_free_tags(bqt))) + BUG(); +} +EXPORT_SYMBOL(blk_free_tags); + +/** + * blk_queue_free_tags - release tag maintenance info + * @q: the request queue for the device + * + * Notes: + * This is used to disabled tagged queuing to a device, yet leave + * queue in function. + **/ +void blk_queue_free_tags(struct request_queue *q) +{ + clear_bit(QUEUE_FLAG_QUEUED, &q->queue_flags); +} + +EXPORT_SYMBOL(blk_queue_free_tags); + +static int +init_tag_map(struct request_queue *q, struct blk_queue_tag *tags, int depth) +{ + struct request **tag_index; + unsigned long *tag_map; + int nr_ulongs; + + if (q && depth > q->nr_requests * 2) { + depth = q->nr_requests * 2; + printk(KERN_ERR "%s: adjusted depth to %d\n", + __FUNCTION__, depth); + } + + tag_index = kzalloc(depth * sizeof(struct request *), GFP_ATOMIC); + if (!tag_index) + goto fail; + + nr_ulongs = ALIGN(depth, BITS_PER_LONG) / BITS_PER_LONG; + tag_map = kzalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC); + if (!tag_map) + goto fail; + + tags->real_max_depth = depth; + tags->max_depth = depth; + tags->tag_index = tag_index; + tags->tag_map = tag_map; + + return 0; +fail: + kfree(tag_index); + return -ENOMEM; +} + +static struct blk_queue_tag *__blk_queue_init_tags(struct request_queue *q, + int depth) +{ + struct blk_queue_tag *tags; + + tags = kmalloc(sizeof(struct blk_queue_tag), GFP_ATOMIC); + if (!tags) + goto fail; + + if (init_tag_map(q, tags, depth)) + goto fail; + + tags->busy = 0; + atomic_set(&tags->refcnt, 1); + return tags; +fail: + kfree(tags); + return NULL; +} + +/** + * blk_init_tags - initialize the tag info for an external tag map + * @depth: the maximum queue depth supported + * @tags: the tag to use + **/ +struct blk_queue_tag *blk_init_tags(int depth) +{ + return __blk_queue_init_tags(NULL, depth); +} +EXPORT_SYMBOL(blk_init_tags); + +/** + * blk_queue_init_tags - initialize the queue tag info + * @q: the request queue for the device + * @depth: the maximum queue depth supported + * @tags: the tag to use + **/ +int blk_queue_init_tags(struct request_queue *q, int depth, + struct blk_queue_tag *tags) +{ + int rc; + + BUG_ON(tags && q->queue_tags && tags != q->queue_tags); + + if (!tags && !q->queue_tags) { + tags = __blk_queue_init_tags(q, depth); + + if (!tags) + goto fail; + } else if (q->queue_tags) { + if ((rc = blk_queue_resize_tags(q, depth))) + return rc; + set_bit(QUEUE_FLAG_QUEUED, &q->queue_flags); + return 0; + } else + atomic_inc(&tags->refcnt); + + /* + * assign it, all done + */ + q->queue_tags = tags; + q->queue_flags |= (1 << QUEUE_FLAG_QUEUED); + INIT_LIST_HEAD(&q->tag_busy_list); + return 0; +fail: + kfree(tags); + return -ENOMEM; +} + +EXPORT_SYMBOL(blk_queue_init_tags); + +/** + * blk_queue_resize_tags - change the queueing depth + * @q: the request queue for the device + * @new_depth: the new max command queueing depth + * + * Notes: + * Must be called with the queue lock held. + **/ +int blk_queue_resize_tags(struct request_queue *q, int new_depth) +{ + struct blk_queue_tag *bqt = q->queue_tags; + struct request **tag_index; + unsigned long *tag_map; + int max_depth, nr_ulongs; + + if (!bqt) + return -ENXIO; + + /* + * if we already have large enough real_max_depth. just + * adjust max_depth. *NOTE* as requests with tag value + * between new_depth and real_max_depth can be in-flight, tag + * map can not be shrunk blindly here. + */ + if (new_depth <= bqt->real_max_depth) { + bqt->max_depth = new_depth; + return 0; + } + + /* + * Currently cannot replace a shared tag map with a new + * one, so error out if this is the case + */ + if (atomic_read(&bqt->refcnt) != 1) + return -EBUSY; + + /* + * save the old state info, so we can copy it back + */ + tag_index = bqt->tag_index; + tag_map = bqt->tag_map; + max_depth = bqt->real_max_depth; + + if (init_tag_map(q, bqt, new_depth)) + return -ENOMEM; + + memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *)); + nr_ulongs = ALIGN(max_depth, BITS_PER_LONG) / BITS_PER_LONG; + memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long)); + + kfree(tag_index); + kfree(tag_map); + return 0; +} + +EXPORT_SYMBOL(blk_queue_resize_tags); + +/** + * blk_queue_end_tag - end tag operations for a request + * @q: the request queue for the device + * @rq: the request that has completed + * + * Description: + * Typically called when end_that_request_first() returns 0, meaning + * all transfers have been done for a request. It's important to call + * this function before end_that_request_last(), as that will put the + * request back on the free list thus corrupting the internal tag list. + * + * Notes: + * queue lock must be held. + **/ +void blk_queue_end_tag(struct request_queue *q, struct request *rq) +{ + struct blk_queue_tag *bqt = q->queue_tags; + int tag = rq->tag; + + BUG_ON(tag == -1); + + if (unlikely(tag >= bqt->real_max_depth)) + /* + * This can happen after tag depth has been reduced. + * FIXME: how about a warning or info message here? + */ + return; + + list_del_init(&rq->queuelist); + rq->cmd_flags &= ~REQ_QUEUED; + rq->tag = -1; + + if (unlikely(bqt->tag_index[tag] == NULL)) + printk(KERN_ERR "%s: tag %d is missing\n", + __FUNCTION__, tag); + + bqt->tag_index[tag] = NULL; + + if (unlikely(!test_bit(tag, bqt->tag_map))) { + printk(KERN_ERR "%s: attempt to clear non-busy tag (%d)\n", + __FUNCTION__, tag); + return; + } + /* + * The tag_map bit acts as a lock for tag_index[bit], so we need + * unlock memory barrier semantics. + */ + clear_bit_unlock(tag, bqt->tag_map); + bqt->busy--; +} + +EXPORT_SYMBOL(blk_queue_end_tag); + +/** + * blk_queue_start_tag - find a free tag and assign it + * @q: the request queue for the device + * @rq: the block request that needs tagging + * + * Description: + * This can either be used as a stand-alone helper, or possibly be + * assigned as the queue &prep_rq_fn (in which case &struct request + * automagically gets a tag assigned). Note that this function + * assumes that any type of request can be queued! if this is not + * true for your device, you must check the request type before + * calling this function. The request will also be removed from + * the request queue, so it's the drivers responsibility to readd + * it if it should need to be restarted for some reason. + * + * Notes: + * queue lock must be held. + **/ +int blk_queue_start_tag(struct request_queue *q, struct request *rq) +{ + struct blk_queue_tag *bqt = q->queue_tags; + int tag; + + if (unlikely((rq->cmd_flags & REQ_QUEUED))) { + printk(KERN_ERR + "%s: request %p for device [%s] already tagged %d", + __FUNCTION__, rq, + rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->tag); + BUG(); + } + + /* + * Protect against shared tag maps, as we may not have exclusive + * access to the tag map. + */ + do { + tag = find_first_zero_bit(bqt->tag_map, bqt->max_depth); + if (tag >= bqt->max_depth) + return 1; + + } while (test_and_set_bit_lock(tag, bqt->tag_map)); + /* + * We need lock ordering semantics given by test_and_set_bit_lock. + * See blk_queue_end_tag for details. + */ + + rq->cmd_flags |= REQ_QUEUED; + rq->tag = tag; + bqt->tag_index[tag] = rq; + blkdev_dequeue_request(rq); + list_add(&rq->queuelist, &q->tag_busy_list); + bqt->busy++; + return 0; +} + +EXPORT_SYMBOL(blk_queue_start_tag); + +/** + * blk_queue_invalidate_tags - invalidate all pending tags + * @q: the request queue for the device + * + * Description: + * Hardware conditions may dictate a need to stop all pending requests. + * In this case, we will safely clear the block side of the tag queue and + * readd all requests to the request queue in the right order. + * + * Notes: + * queue lock must be held. + **/ +void blk_queue_invalidate_tags(struct request_queue *q) +{ + struct list_head *tmp, *n; + + list_for_each_safe(tmp, n, &q->tag_busy_list) + blk_requeue_request(q, list_entry_rq(tmp)); +} + +EXPORT_SYMBOL(blk_queue_invalidate_tags); diff --git a/block/blk.h b/block/blk.h new file mode 100644 index 00000000000..ec898dd0c65 --- /dev/null +++ b/block/blk.h @@ -0,0 +1,53 @@ +#ifndef BLK_INTERNAL_H +#define BLK_INTERNAL_H + +/* Amount of time in which a process may batch requests */ +#define BLK_BATCH_TIME (HZ/50UL) + +/* Number of requests a "batching" process may submit */ +#define BLK_BATCH_REQ 32 + +extern struct kmem_cache *blk_requestq_cachep; +extern struct kobj_type blk_queue_ktype; + +void rq_init(struct request_queue *q, struct request *rq); +void init_request_from_bio(struct request *req, struct bio *bio); +void blk_rq_bio_prep(struct request_queue *q, struct request *rq, + struct bio *bio); +void __blk_queue_free_tags(struct request_queue *q); + +void blk_unplug_work(struct work_struct *work); +void blk_unplug_timeout(unsigned long data); + +struct io_context *current_io_context(gfp_t gfp_flags, int node); + +int ll_back_merge_fn(struct request_queue *q, struct request *req, + struct bio *bio); +int ll_front_merge_fn(struct request_queue *q, struct request *req, + struct bio *bio); +int attempt_back_merge(struct request_queue *q, struct request *rq); +int attempt_front_merge(struct request_queue *q, struct request *rq); +void blk_recalc_rq_segments(struct request *rq); +void blk_recalc_rq_sectors(struct request *rq, int nsect); + +void blk_queue_congestion_threshold(struct request_queue *q); + +/* + * Return the threshold (number of used requests) at which the queue is + * considered to be congested. It include a little hysteresis to keep the + * context switch rate down. + */ +static inline int queue_congestion_on_threshold(struct request_queue *q) +{ + return q->nr_congestion_on; +} + +/* + * The threshold at which a queue is considered to be uncongested + */ +static inline int queue_congestion_off_threshold(struct request_queue *q) +{ + return q->nr_congestion_off; +} + +#endif diff --git a/block/blktrace.c b/block/blktrace.c index 9b4da4ae3c7..568588cd16b 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -235,7 +235,7 @@ static void blk_trace_cleanup(struct blk_trace *bt) kfree(bt); } -static int blk_trace_remove(struct request_queue *q) +int blk_trace_remove(struct request_queue *q) { struct blk_trace *bt; @@ -249,6 +249,7 @@ static int blk_trace_remove(struct request_queue *q) return 0; } +EXPORT_SYMBOL_GPL(blk_trace_remove); static int blk_dropped_open(struct inode *inode, struct file *filp) { @@ -316,18 +317,17 @@ static struct rchan_callbacks blk_relay_callbacks = { /* * Setup everything required to start tracing */ -int do_blk_trace_setup(struct request_queue *q, struct block_device *bdev, +int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, struct blk_user_trace_setup *buts) { struct blk_trace *old_bt, *bt = NULL; struct dentry *dir = NULL; - char b[BDEVNAME_SIZE]; int ret, i; if (!buts->buf_size || !buts->buf_nr) return -EINVAL; - strcpy(buts->name, bdevname(bdev, b)); + strcpy(buts->name, name); /* * some device names have larger paths - convert the slashes @@ -352,7 +352,7 @@ int do_blk_trace_setup(struct request_queue *q, struct block_device *bdev, goto err; bt->dir = dir; - bt->dev = bdev->bd_dev; + bt->dev = dev; atomic_set(&bt->dropped, 0); ret = -EIO; @@ -399,8 +399,8 @@ err: return ret; } -static int blk_trace_setup(struct request_queue *q, struct block_device *bdev, - char __user *arg) +int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + char __user *arg) { struct blk_user_trace_setup buts; int ret; @@ -409,7 +409,7 @@ static int blk_trace_setup(struct request_queue *q, struct block_device *bdev, if (ret) return -EFAULT; - ret = do_blk_trace_setup(q, bdev, &buts); + ret = do_blk_trace_setup(q, name, dev, &buts); if (ret) return ret; @@ -418,8 +418,9 @@ static int blk_trace_setup(struct request_queue *q, struct block_device *bdev, return 0; } +EXPORT_SYMBOL_GPL(blk_trace_setup); -static int blk_trace_startstop(struct request_queue *q, int start) +int blk_trace_startstop(struct request_queue *q, int start) { struct blk_trace *bt; int ret; @@ -452,6 +453,7 @@ static int blk_trace_startstop(struct request_queue *q, int start) return ret; } +EXPORT_SYMBOL_GPL(blk_trace_startstop); /** * blk_trace_ioctl: - handle the ioctls associated with tracing @@ -464,6 +466,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) { struct request_queue *q; int ret, start = 0; + char b[BDEVNAME_SIZE]; q = bdev_get_queue(bdev); if (!q) @@ -473,7 +476,8 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) switch (cmd) { case BLKTRACESETUP: - ret = blk_trace_setup(q, bdev, arg); + strcpy(b, bdevname(bdev, b)); + ret = blk_trace_setup(q, b, bdev->bd_dev, arg); break; case BLKTRACESTART: start = 1; diff --git a/block/bsg.c b/block/bsg.c index 8e181ab3afb..8917c5174dc 100644 --- a/block/bsg.c +++ b/block/bsg.c @@ -279,6 +279,7 @@ bsg_map_hdr(struct bsg_device *bd, struct sg_io_v4 *hdr) goto out; } rq->next_rq = next_rq; + next_rq->cmd_type = rq->cmd_type; dxferp = (void*)(unsigned long)hdr->din_xferp; ret = blk_rq_map_user(q, next_rq, dxferp, hdr->din_xfer_len); @@ -445,6 +446,15 @@ static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr, else hdr->dout_resid = rq->data_len; + /* + * If the request generated a negative error number, return it + * (providing we aren't already returning an error); if it's + * just a protocol response (i.e. non negative), that gets + * processed above. + */ + if (!ret && rq->errors < 0) + ret = rq->errors; + blk_rq_unmap_user(bio); blk_put_request(rq); @@ -837,6 +847,7 @@ static long bsg_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct bsg_device *bd = file->private_data; int __user *uarg = (int __user *) arg; + int ret; switch (cmd) { /* @@ -889,12 +900,12 @@ static long bsg_ioctl(struct file *file, unsigned int cmd, unsigned long arg) if (rq->next_rq) bidi_bio = rq->next_rq->bio; blk_execute_rq(bd->queue, NULL, rq, 0); - blk_complete_sgv4_hdr_rq(rq, &hdr, bio, bidi_bio); + ret = blk_complete_sgv4_hdr_rq(rq, &hdr, bio, bidi_bio); if (copy_to_user(uarg, &hdr, sizeof(hdr))) return -EFAULT; - return 0; + return ret; } /* * block device ioctls diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 13553e015d7..f28d1fb3060 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -26,9 +26,9 @@ static const int cfq_slice_async_rq = 2; static int cfq_slice_idle = HZ / 125; /* - * grace period before allowing idle class to get disk access + * offset from end of service tree */ -#define CFQ_IDLE_GRACE (HZ / 10) +#define CFQ_IDLE_DELAY (HZ / 5) /* * below this threshold, we consider thinktime immediate @@ -98,8 +98,6 @@ struct cfq_data { struct cfq_queue *async_cfqq[2][IOPRIO_BE_NR]; struct cfq_queue *async_idle_cfqq; - struct timer_list idle_class_timer; - sector_t last_position; unsigned long last_end_request; @@ -199,8 +197,8 @@ CFQ_CFQQ_FNS(sync); static void cfq_dispatch_insert(struct request_queue *, struct request *); static struct cfq_queue *cfq_get_queue(struct cfq_data *, int, - struct task_struct *, gfp_t); -static struct cfq_io_context *cfq_cic_rb_lookup(struct cfq_data *, + struct io_context *, gfp_t); +static struct cfq_io_context *cfq_cic_lookup(struct cfq_data *, struct io_context *); static inline struct cfq_queue *cic_to_cfqq(struct cfq_io_context *cic, @@ -384,12 +382,15 @@ cfq_choose_req(struct cfq_data *cfqd, struct request *rq1, struct request *rq2) /* * The below is leftmost cache rbtree addon */ -static struct rb_node *cfq_rb_first(struct cfq_rb_root *root) +static struct cfq_queue *cfq_rb_first(struct cfq_rb_root *root) { if (!root->left) root->left = rb_first(&root->rb); - return root->left; + if (root->left) + return rb_entry(root->left, struct cfq_queue, rb_node); + + return NULL; } static void cfq_rb_erase(struct rb_node *n, struct cfq_rb_root *root) @@ -446,12 +447,20 @@ static unsigned long cfq_slice_offset(struct cfq_data *cfqd, static void cfq_service_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq, int add_front) { - struct rb_node **p = &cfqd->service_tree.rb.rb_node; - struct rb_node *parent = NULL; + struct rb_node **p, *parent; + struct cfq_queue *__cfqq; unsigned long rb_key; int left; - if (!add_front) { + if (cfq_class_idle(cfqq)) { + rb_key = CFQ_IDLE_DELAY; + parent = rb_last(&cfqd->service_tree.rb); + if (parent && parent != &cfqq->rb_node) { + __cfqq = rb_entry(parent, struct cfq_queue, rb_node); + rb_key += __cfqq->rb_key; + } else + rb_key += jiffies; + } else if (!add_front) { rb_key = cfq_slice_offset(cfqd, cfqq) + jiffies; rb_key += cfqq->slice_resid; cfqq->slice_resid = 0; @@ -469,8 +478,9 @@ static void cfq_service_tree_add(struct cfq_data *cfqd, } left = 1; + parent = NULL; + p = &cfqd->service_tree.rb.rb_node; while (*p) { - struct cfq_queue *__cfqq; struct rb_node **n; parent = *p; @@ -524,8 +534,7 @@ static void cfq_resort_rr_list(struct cfq_data *cfqd, struct cfq_queue *cfqq) * add to busy list of queues for service, trying to be fair in ordering * the pending list according to last request service */ -static inline void -cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) +static void cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) { BUG_ON(cfq_cfqq_on_rr(cfqq)); cfq_mark_cfqq_on_rr(cfqq); @@ -538,8 +547,7 @@ cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) * Called when the cfqq no longer has requests pending, remove it from * the service tree. */ -static inline void -cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) +static void cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) { BUG_ON(!cfq_cfqq_on_rr(cfqq)); cfq_clear_cfqq_on_rr(cfqq); @@ -554,7 +562,7 @@ cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) /* * rb tree support functions */ -static inline void cfq_del_rq_rb(struct request *rq) +static void cfq_del_rq_rb(struct request *rq) { struct cfq_queue *cfqq = RQ_CFQQ(rq); struct cfq_data *cfqd = cfqq->cfqd; @@ -594,8 +602,7 @@ static void cfq_add_rq_rb(struct request *rq) BUG_ON(!cfqq->next_rq); } -static inline void -cfq_reposition_rq_rb(struct cfq_queue *cfqq, struct request *rq) +static void cfq_reposition_rq_rb(struct cfq_queue *cfqq, struct request *rq) { elv_rb_del(&cfqq->sort_list, rq); cfqq->queued[rq_is_sync(rq)]--; @@ -609,7 +616,7 @@ cfq_find_rq_fmerge(struct cfq_data *cfqd, struct bio *bio) struct cfq_io_context *cic; struct cfq_queue *cfqq; - cic = cfq_cic_rb_lookup(cfqd, tsk->io_context); + cic = cfq_cic_lookup(cfqd, tsk->io_context); if (!cic) return NULL; @@ -721,7 +728,7 @@ static int cfq_allow_merge(struct request_queue *q, struct request *rq, * Lookup the cfqq that this bio will be queued with. Allow * merge only if rq is queued there. */ - cic = cfq_cic_rb_lookup(cfqd, current->io_context); + cic = cfq_cic_lookup(cfqd, current->io_context); if (!cic) return 0; @@ -732,15 +739,10 @@ static int cfq_allow_merge(struct request_queue *q, struct request *rq, return 0; } -static inline void -__cfq_set_active_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq) +static void __cfq_set_active_queue(struct cfq_data *cfqd, + struct cfq_queue *cfqq) { if (cfqq) { - /* - * stop potential idle class queues waiting service - */ - del_timer(&cfqd->idle_class_timer); - cfqq->slice_end = 0; cfq_clear_cfqq_must_alloc_slice(cfqq); cfq_clear_cfqq_fifo_expire(cfqq); @@ -789,47 +791,16 @@ static inline void cfq_slice_expired(struct cfq_data *cfqd, int timed_out) __cfq_slice_expired(cfqd, cfqq, timed_out); } -static int start_idle_class_timer(struct cfq_data *cfqd) -{ - unsigned long end = cfqd->last_end_request + CFQ_IDLE_GRACE; - unsigned long now = jiffies; - - if (time_before(now, end) && - time_after_eq(now, cfqd->last_end_request)) { - mod_timer(&cfqd->idle_class_timer, end); - return 1; - } - - return 0; -} - /* * Get next queue for service. Unless we have a queue preemption, * we'll simply select the first cfqq in the service tree. */ static struct cfq_queue *cfq_get_next_queue(struct cfq_data *cfqd) { - struct cfq_queue *cfqq; - struct rb_node *n; - if (RB_EMPTY_ROOT(&cfqd->service_tree.rb)) return NULL; - n = cfq_rb_first(&cfqd->service_tree); - cfqq = rb_entry(n, struct cfq_queue, rb_node); - - if (cfq_class_idle(cfqq)) { - /* - * if we have idle queues and no rt or be queues had - * pending requests, either allow immediate service if - * the grace period has passed or arm the idle grace - * timer - */ - if (start_idle_class_timer(cfqd)) - cfqq = NULL; - } - - return cfqq; + return cfq_rb_first(&cfqd->service_tree); } /* @@ -895,7 +866,7 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) * task has exited, don't wait */ cic = cfqd->active_cic; - if (!cic || !cic->ioc->task) + if (!cic || !atomic_read(&cic->ioc->nr_tasks)) return; /* @@ -939,7 +910,7 @@ static void cfq_dispatch_insert(struct request_queue *q, struct request *rq) /* * return expired entry, or NULL to just start from scratch in rbtree */ -static inline struct request *cfq_check_fifo(struct cfq_queue *cfqq) +static struct request *cfq_check_fifo(struct cfq_queue *cfqq) { struct cfq_data *cfqd = cfqq->cfqd; struct request *rq; @@ -1068,7 +1039,7 @@ __cfq_dispatch_requests(struct cfq_data *cfqd, struct cfq_queue *cfqq, return dispatched; } -static inline int __cfq_forced_dispatch_cfqq(struct cfq_queue *cfqq) +static int __cfq_forced_dispatch_cfqq(struct cfq_queue *cfqq) { int dispatched = 0; @@ -1087,14 +1058,11 @@ static inline int __cfq_forced_dispatch_cfqq(struct cfq_queue *cfqq) */ static int cfq_forced_dispatch(struct cfq_data *cfqd) { + struct cfq_queue *cfqq; int dispatched = 0; - struct rb_node *n; - - while ((n = cfq_rb_first(&cfqd->service_tree)) != NULL) { - struct cfq_queue *cfqq = rb_entry(n, struct cfq_queue, rb_node); + while ((cfqq = cfq_rb_first(&cfqd->service_tree)) != NULL) dispatched += __cfq_forced_dispatch_cfqq(cfqq); - } cfq_slice_expired(cfqd, 0); @@ -1170,20 +1138,69 @@ static void cfq_put_queue(struct cfq_queue *cfqq) kmem_cache_free(cfq_pool, cfqq); } -static void cfq_free_io_context(struct io_context *ioc) +/* + * Call func for each cic attached to this ioc. Returns number of cic's seen. + */ +#define CIC_GANG_NR 16 +static unsigned int +call_for_each_cic(struct io_context *ioc, + void (*func)(struct io_context *, struct cfq_io_context *)) { - struct cfq_io_context *__cic; - struct rb_node *n; - int freed = 0; + struct cfq_io_context *cics[CIC_GANG_NR]; + unsigned long index = 0; + unsigned int called = 0; + int nr; - ioc->ioc_data = NULL; + rcu_read_lock(); - while ((n = rb_first(&ioc->cic_root)) != NULL) { - __cic = rb_entry(n, struct cfq_io_context, rb_node); - rb_erase(&__cic->rb_node, &ioc->cic_root); - kmem_cache_free(cfq_ioc_pool, __cic); - freed++; - } + do { + int i; + + /* + * Perhaps there's a better way - this just gang lookups from + * 0 to the end, restarting after each CIC_GANG_NR from the + * last key + 1. + */ + nr = radix_tree_gang_lookup(&ioc->radix_root, (void **) cics, + index, CIC_GANG_NR); + if (!nr) + break; + + called += nr; + index = 1 + (unsigned long) cics[nr - 1]->key; + + for (i = 0; i < nr; i++) + func(ioc, cics[i]); + } while (nr == CIC_GANG_NR); + + rcu_read_unlock(); + + return called; +} + +static void cic_free_func(struct io_context *ioc, struct cfq_io_context *cic) +{ + unsigned long flags; + + BUG_ON(!cic->dead_key); + + spin_lock_irqsave(&ioc->lock, flags); + radix_tree_delete(&ioc->radix_root, cic->dead_key); + spin_unlock_irqrestore(&ioc->lock, flags); + + kmem_cache_free(cfq_ioc_pool, cic); +} + +static void cfq_free_io_context(struct io_context *ioc) +{ + int freed; + + /* + * ioc->refcount is zero here, so no more cic's are allowed to be + * linked into this ioc. So it should be ok to iterate over the known + * list, we will see all cic's since no new ones are added. + */ + freed = call_for_each_cic(ioc, cic_free_func); elv_ioc_count_mod(ioc_count, -freed); @@ -1205,7 +1222,12 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd, struct cfq_io_context *cic) { list_del_init(&cic->queue_list); + + /* + * Make sure key == NULL is seen for dead queues + */ smp_wmb(); + cic->dead_key = (unsigned long) cic->key; cic->key = NULL; if (cic->cfqq[ASYNC]) { @@ -1219,16 +1241,18 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd, } } -static void cfq_exit_single_io_context(struct cfq_io_context *cic) +static void cfq_exit_single_io_context(struct io_context *ioc, + struct cfq_io_context *cic) { struct cfq_data *cfqd = cic->key; if (cfqd) { struct request_queue *q = cfqd->queue; + unsigned long flags; - spin_lock_irq(q->queue_lock); + spin_lock_irqsave(q->queue_lock, flags); __cfq_exit_single_io_context(cfqd, cic); - spin_unlock_irq(q->queue_lock); + spin_unlock_irqrestore(q->queue_lock, flags); } } @@ -1238,21 +1262,8 @@ static void cfq_exit_single_io_context(struct cfq_io_context *cic) */ static void cfq_exit_io_context(struct io_context *ioc) { - struct cfq_io_context *__cic; - struct rb_node *n; - - ioc->ioc_data = NULL; - - /* - * put the reference this task is holding to the various queues - */ - n = rb_first(&ioc->cic_root); - while (n != NULL) { - __cic = rb_entry(n, struct cfq_io_context, rb_node); - - cfq_exit_single_io_context(__cic); - n = rb_next(n); - } + rcu_assign_pointer(ioc->ioc_data, NULL); + call_for_each_cic(ioc, cfq_exit_single_io_context); } static struct cfq_io_context * @@ -1273,7 +1284,7 @@ cfq_alloc_io_context(struct cfq_data *cfqd, gfp_t gfp_mask) return cic; } -static void cfq_init_prio_data(struct cfq_queue *cfqq) +static void cfq_init_prio_data(struct cfq_queue *cfqq, struct io_context *ioc) { struct task_struct *tsk = current; int ioprio_class; @@ -1281,7 +1292,7 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq) if (!cfq_cfqq_prio_changed(cfqq)) return; - ioprio_class = IOPRIO_PRIO_CLASS(tsk->ioprio); + ioprio_class = IOPRIO_PRIO_CLASS(ioc->ioprio); switch (ioprio_class) { default: printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class); @@ -1293,11 +1304,11 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq) cfqq->ioprio_class = IOPRIO_CLASS_BE; break; case IOPRIO_CLASS_RT: - cfqq->ioprio = task_ioprio(tsk); + cfqq->ioprio = task_ioprio(ioc); cfqq->ioprio_class = IOPRIO_CLASS_RT; break; case IOPRIO_CLASS_BE: - cfqq->ioprio = task_ioprio(tsk); + cfqq->ioprio = task_ioprio(ioc); cfqq->ioprio_class = IOPRIO_CLASS_BE; break; case IOPRIO_CLASS_IDLE: @@ -1316,7 +1327,7 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq) cfq_clear_cfqq_prio_changed(cfqq); } -static inline void changed_ioprio(struct cfq_io_context *cic) +static void changed_ioprio(struct io_context *ioc, struct cfq_io_context *cic) { struct cfq_data *cfqd = cic->key; struct cfq_queue *cfqq; @@ -1330,8 +1341,7 @@ static inline void changed_ioprio(struct cfq_io_context *cic) cfqq = cic->cfqq[ASYNC]; if (cfqq) { struct cfq_queue *new_cfqq; - new_cfqq = cfq_get_queue(cfqd, ASYNC, cic->ioc->task, - GFP_ATOMIC); + new_cfqq = cfq_get_queue(cfqd, ASYNC, cic->ioc, GFP_ATOMIC); if (new_cfqq) { cic->cfqq[ASYNC] = new_cfqq; cfq_put_queue(cfqq); @@ -1347,29 +1357,19 @@ static inline void changed_ioprio(struct cfq_io_context *cic) static void cfq_ioc_set_ioprio(struct io_context *ioc) { - struct cfq_io_context *cic; - struct rb_node *n; - + call_for_each_cic(ioc, changed_ioprio); ioc->ioprio_changed = 0; - - n = rb_first(&ioc->cic_root); - while (n != NULL) { - cic = rb_entry(n, struct cfq_io_context, rb_node); - - changed_ioprio(cic); - n = rb_next(n); - } } static struct cfq_queue * cfq_find_alloc_queue(struct cfq_data *cfqd, int is_sync, - struct task_struct *tsk, gfp_t gfp_mask) + struct io_context *ioc, gfp_t gfp_mask) { struct cfq_queue *cfqq, *new_cfqq = NULL; struct cfq_io_context *cic; retry: - cic = cfq_cic_rb_lookup(cfqd, tsk->io_context); + cic = cfq_cic_lookup(cfqd, ioc); /* cic always exists here */ cfqq = cic_to_cfqq(cic, is_sync); @@ -1404,15 +1404,16 @@ retry: atomic_set(&cfqq->ref, 0); cfqq->cfqd = cfqd; - if (is_sync) { - cfq_mark_cfqq_idle_window(cfqq); - cfq_mark_cfqq_sync(cfqq); - } - cfq_mark_cfqq_prio_changed(cfqq); cfq_mark_cfqq_queue_new(cfqq); - cfq_init_prio_data(cfqq); + cfq_init_prio_data(cfqq, ioc); + + if (is_sync) { + if (!cfq_class_idle(cfqq)) + cfq_mark_cfqq_idle_window(cfqq); + cfq_mark_cfqq_sync(cfqq); + } } if (new_cfqq) @@ -1439,11 +1440,11 @@ cfq_async_queue_prio(struct cfq_data *cfqd, int ioprio_class, int ioprio) } static struct cfq_queue * -cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct task_struct *tsk, +cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct io_context *ioc, gfp_t gfp_mask) { - const int ioprio = task_ioprio(tsk); - const int ioprio_class = task_ioprio_class(tsk); + const int ioprio = task_ioprio(ioc); + const int ioprio_class = task_ioprio_class(ioc); struct cfq_queue **async_cfqq = NULL; struct cfq_queue *cfqq = NULL; @@ -1453,7 +1454,7 @@ cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct task_struct *tsk, } if (!cfqq) { - cfqq = cfq_find_alloc_queue(cfqd, is_sync, tsk, gfp_mask); + cfqq = cfq_find_alloc_queue(cfqd, is_sync, ioc, gfp_mask); if (!cfqq) return NULL; } @@ -1470,28 +1471,42 @@ cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct task_struct *tsk, return cfqq; } +static void cfq_cic_free(struct cfq_io_context *cic) +{ + kmem_cache_free(cfq_ioc_pool, cic); + elv_ioc_count_dec(ioc_count); + + if (ioc_gone && !elv_ioc_count_read(ioc_count)) + complete(ioc_gone); +} + /* * We drop cfq io contexts lazily, so we may find a dead one. */ static void -cfq_drop_dead_cic(struct io_context *ioc, struct cfq_io_context *cic) +cfq_drop_dead_cic(struct cfq_data *cfqd, struct io_context *ioc, + struct cfq_io_context *cic) { + unsigned long flags; + WARN_ON(!list_empty(&cic->queue_list)); + spin_lock_irqsave(&ioc->lock, flags); + if (ioc->ioc_data == cic) - ioc->ioc_data = NULL; + rcu_assign_pointer(ioc->ioc_data, NULL); - rb_erase(&cic->rb_node, &ioc->cic_root); - kmem_cache_free(cfq_ioc_pool, cic); - elv_ioc_count_dec(ioc_count); + radix_tree_delete(&ioc->radix_root, (unsigned long) cfqd); + spin_unlock_irqrestore(&ioc->lock, flags); + + cfq_cic_free(cic); } static struct cfq_io_context * -cfq_cic_rb_lookup(struct cfq_data *cfqd, struct io_context *ioc) +cfq_cic_lookup(struct cfq_data *cfqd, struct io_context *ioc) { - struct rb_node *n; struct cfq_io_context *cic; - void *k, *key = cfqd; + void *k; if (unlikely(!ioc)) return NULL; @@ -1499,74 +1514,64 @@ cfq_cic_rb_lookup(struct cfq_data *cfqd, struct io_context *ioc) /* * we maintain a last-hit cache, to avoid browsing over the tree */ - cic = ioc->ioc_data; + cic = rcu_dereference(ioc->ioc_data); if (cic && cic->key == cfqd) return cic; -restart: - n = ioc->cic_root.rb_node; - while (n) { - cic = rb_entry(n, struct cfq_io_context, rb_node); + do { + rcu_read_lock(); + cic = radix_tree_lookup(&ioc->radix_root, (unsigned long) cfqd); + rcu_read_unlock(); + if (!cic) + break; /* ->key must be copied to avoid race with cfq_exit_queue() */ k = cic->key; if (unlikely(!k)) { - cfq_drop_dead_cic(ioc, cic); - goto restart; + cfq_drop_dead_cic(cfqd, ioc, cic); + continue; } - if (key < k) - n = n->rb_left; - else if (key > k) - n = n->rb_right; - else { - ioc->ioc_data = cic; - return cic; - } - } + rcu_assign_pointer(ioc->ioc_data, cic); + break; + } while (1); - return NULL; + return cic; } -static inline void -cfq_cic_link(struct cfq_data *cfqd, struct io_context *ioc, - struct cfq_io_context *cic) +/* + * Add cic into ioc, using cfqd as the search key. This enables us to lookup + * the process specific cfq io context when entered from the block layer. + * Also adds the cic to a per-cfqd list, used when this queue is removed. + */ +static int cfq_cic_link(struct cfq_data *cfqd, struct io_context *ioc, + struct cfq_io_context *cic, gfp_t gfp_mask) { - struct rb_node **p; - struct rb_node *parent; - struct cfq_io_context *__cic; unsigned long flags; - void *k; + int ret; - cic->ioc = ioc; - cic->key = cfqd; + ret = radix_tree_preload(gfp_mask); + if (!ret) { + cic->ioc = ioc; + cic->key = cfqd; -restart: - parent = NULL; - p = &ioc->cic_root.rb_node; - while (*p) { - parent = *p; - __cic = rb_entry(parent, struct cfq_io_context, rb_node); - /* ->key must be copied to avoid race with cfq_exit_queue() */ - k = __cic->key; - if (unlikely(!k)) { - cfq_drop_dead_cic(ioc, __cic); - goto restart; - } + spin_lock_irqsave(&ioc->lock, flags); + ret = radix_tree_insert(&ioc->radix_root, + (unsigned long) cfqd, cic); + spin_unlock_irqrestore(&ioc->lock, flags); - if (cic->key < k) - p = &(*p)->rb_left; - else if (cic->key > k) - p = &(*p)->rb_right; - else - BUG(); + radix_tree_preload_end(); + + if (!ret) { + spin_lock_irqsave(cfqd->queue->queue_lock, flags); + list_add(&cic->queue_list, &cfqd->cic_list); + spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); + } } - rb_link_node(&cic->rb_node, parent, p); - rb_insert_color(&cic->rb_node, &ioc->cic_root); + if (ret) + printk(KERN_ERR "cfq: cic link failed!\n"); - spin_lock_irqsave(cfqd->queue->queue_lock, flags); - list_add(&cic->queue_list, &cfqd->cic_list); - spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); + return ret; } /* @@ -1586,7 +1591,7 @@ cfq_get_io_context(struct cfq_data *cfqd, gfp_t gfp_mask) if (!ioc) return NULL; - cic = cfq_cic_rb_lookup(cfqd, ioc); + cic = cfq_cic_lookup(cfqd, ioc); if (cic) goto out; @@ -1594,13 +1599,17 @@ cfq_get_io_context(struct cfq_data *cfqd, gfp_t gfp_mask) if (cic == NULL) goto err; - cfq_cic_link(cfqd, ioc, cic); + if (cfq_cic_link(cfqd, ioc, cic, gfp_mask)) + goto err_free; + out: smp_read_barrier_depends(); if (unlikely(ioc->ioprio_changed)) cfq_ioc_set_ioprio(ioc); return cic; +err_free: + cfq_cic_free(cic); err: put_io_context(ioc); return NULL; @@ -1655,12 +1664,15 @@ cfq_update_idle_window(struct cfq_data *cfqd, struct cfq_queue *cfqq, { int enable_idle; - if (!cfq_cfqq_sync(cfqq)) + /* + * Don't idle for async or idle io prio class + */ + if (!cfq_cfqq_sync(cfqq) || cfq_class_idle(cfqq)) return; enable_idle = cfq_cfqq_idle_window(cfqq); - if (!cic->ioc->task || !cfqd->cfq_slice_idle || + if (!atomic_read(&cic->ioc->nr_tasks) || !cfqd->cfq_slice_idle || (cfqd->hw_tag && CIC_SEEKY(cic))) enable_idle = 0; else if (sample_valid(cic->ttime_samples)) { @@ -1793,7 +1805,7 @@ static void cfq_insert_request(struct request_queue *q, struct request *rq) struct cfq_data *cfqd = q->elevator->elevator_data; struct cfq_queue *cfqq = RQ_CFQQ(rq); - cfq_init_prio_data(cfqq); + cfq_init_prio_data(cfqq, RQ_CIC(rq)->ioc); cfq_add_rq_rb(rq); @@ -1834,7 +1846,7 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq) cfq_set_prio_slice(cfqd, cfqq); cfq_clear_cfqq_slice_new(cfqq); } - if (cfq_slice_used(cfqq)) + if (cfq_slice_used(cfqq) || cfq_class_idle(cfqq)) cfq_slice_expired(cfqd, 1); else if (sync && RB_EMPTY_ROOT(&cfqq->sort_list)) cfq_arm_slice_timer(cfqd); @@ -1894,13 +1906,13 @@ static int cfq_may_queue(struct request_queue *q, int rw) * so just lookup a possibly existing queue, or return 'may queue' * if that fails */ - cic = cfq_cic_rb_lookup(cfqd, tsk->io_context); + cic = cfq_cic_lookup(cfqd, tsk->io_context); if (!cic) return ELV_MQUEUE_MAY; cfqq = cic_to_cfqq(cic, rw & REQ_RW_SYNC); if (cfqq) { - cfq_init_prio_data(cfqq); + cfq_init_prio_data(cfqq, cic->ioc); cfq_prio_boost(cfqq); return __cfq_may_queue(cfqq); @@ -1938,7 +1950,6 @@ static int cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask) { struct cfq_data *cfqd = q->elevator->elevator_data; - struct task_struct *tsk = current; struct cfq_io_context *cic; const int rw = rq_data_dir(rq); const int is_sync = rq_is_sync(rq); @@ -1956,7 +1967,7 @@ cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask) cfqq = cic_to_cfqq(cic, is_sync); if (!cfqq) { - cfqq = cfq_get_queue(cfqd, is_sync, tsk, gfp_mask); + cfqq = cfq_get_queue(cfqd, is_sync, cic->ioc, gfp_mask); if (!cfqq) goto queue_fail; @@ -2039,29 +2050,9 @@ out_cont: spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); } -/* - * Timer running if an idle class queue is waiting for service - */ -static void cfq_idle_class_timer(unsigned long data) -{ - struct cfq_data *cfqd = (struct cfq_data *) data; - unsigned long flags; - - spin_lock_irqsave(cfqd->queue->queue_lock, flags); - - /* - * race with a non-idle queue, reset timer - */ - if (!start_idle_class_timer(cfqd)) - cfq_schedule_dispatch(cfqd); - - spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); -} - static void cfq_shutdown_timer_wq(struct cfq_data *cfqd) { del_timer_sync(&cfqd->idle_slice_timer); - del_timer_sync(&cfqd->idle_class_timer); kblockd_flush_work(&cfqd->unplug_work); } @@ -2126,10 +2117,6 @@ static void *cfq_init_queue(struct request_queue *q) cfqd->idle_slice_timer.function = cfq_idle_slice_timer; cfqd->idle_slice_timer.data = (unsigned long) cfqd; - init_timer(&cfqd->idle_class_timer); - cfqd->idle_class_timer.function = cfq_idle_class_timer; - cfqd->idle_class_timer.data = (unsigned long) cfqd; - INIT_WORK(&cfqd->unplug_work, cfq_kick_queue); cfqd->last_end_request = jiffies; @@ -2160,7 +2147,7 @@ static int __init cfq_slab_setup(void) if (!cfq_pool) goto fail; - cfq_ioc_pool = KMEM_CACHE(cfq_io_context, 0); + cfq_ioc_pool = KMEM_CACHE(cfq_io_context, SLAB_DESTROY_BY_RCU); if (!cfq_ioc_pool) goto fail; diff --git a/block/compat_ioctl.c b/block/compat_ioctl.c index cae0a852619..b73373216b0 100644 --- a/block/compat_ioctl.c +++ b/block/compat_ioctl.c @@ -545,6 +545,7 @@ static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg) struct blk_user_trace_setup buts; struct compat_blk_user_trace_setup cbuts; struct request_queue *q; + char b[BDEVNAME_SIZE]; int ret; q = bdev_get_queue(bdev); @@ -554,6 +555,8 @@ static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg) if (copy_from_user(&cbuts, arg, sizeof(cbuts))) return -EFAULT; + strcpy(b, bdevname(bdev, b)); + buts = (struct blk_user_trace_setup) { .act_mask = cbuts.act_mask, .buf_size = cbuts.buf_size, @@ -565,7 +568,7 @@ static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg) memcpy(&buts.name, &cbuts.name, 32); mutex_lock(&bdev->bd_mutex); - ret = do_blk_trace_setup(q, bdev, &buts); + ret = do_blk_trace_setup(q, b, bdev->bd_dev, &buts); mutex_unlock(&bdev->bd_mutex); if (ret) return ret; diff --git a/block/elevator.c b/block/elevator.c index e452deb8039..8cd5775acd7 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -185,9 +185,7 @@ static elevator_t *elevator_alloc(struct request_queue *q, eq->ops = &e->ops; eq->elevator_type = e; - kobject_init(&eq->kobj); - kobject_set_name(&eq->kobj, "%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, @@ -743,7 +741,21 @@ struct request *elv_next_request(struct request_queue *q) 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++; + rq->nr_hw_segments++; + } + + if (!q->prep_rq_fn) break; ret = q->prep_rq_fn(q, rq); @@ -756,6 +768,16 @@ struct request *elv_next_request(struct request_queue *q) * 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->nr_hw_segments; + } + rq = NULL; break; } else if (ret == BLKPREP_KILL) { @@ -931,9 +953,7 @@ int elv_register_queue(struct request_queue *q) 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) { diff --git a/block/genhd.c b/block/genhd.c index f2ac914160d..de2ebb2fab4 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -17,8 +17,10 @@ #include <linux/buffer_head.h> #include <linux/mutex.h> -struct kset block_subsys; -static DEFINE_MUTEX(block_subsys_lock); +static DEFINE_MUTEX(block_class_lock); +#ifndef CONFIG_SYSFS_DEPRECATED +struct kobject *block_depr; +#endif /* * Can be deleted altogether. Later. @@ -37,19 +39,17 @@ static inline int major_to_index(int major) } #ifdef CONFIG_PROC_FS - void blkdev_show(struct seq_file *f, off_t offset) { struct blk_major_name *dp; if (offset < BLKDEV_MAJOR_HASH_SIZE) { - mutex_lock(&block_subsys_lock); + mutex_lock(&block_class_lock); for (dp = major_names[offset]; dp; dp = dp->next) seq_printf(f, "%3d %s\n", dp->major, dp->name); - mutex_unlock(&block_subsys_lock); + mutex_unlock(&block_class_lock); } } - #endif /* CONFIG_PROC_FS */ int register_blkdev(unsigned int major, const char *name) @@ -57,7 +57,7 @@ int register_blkdev(unsigned int major, const char *name) struct blk_major_name **n, *p; int index, ret = 0; - mutex_lock(&block_subsys_lock); + mutex_lock(&block_class_lock); /* temporary */ if (major == 0) { @@ -102,7 +102,7 @@ int register_blkdev(unsigned int major, const char *name) kfree(p); } out: - mutex_unlock(&block_subsys_lock); + mutex_unlock(&block_class_lock); return ret; } @@ -114,7 +114,7 @@ void unregister_blkdev(unsigned int major, const char *name) struct blk_major_name *p = NULL; int index = major_to_index(major); - mutex_lock(&block_subsys_lock); + mutex_lock(&block_class_lock); for (n = &major_names[index]; *n; n = &(*n)->next) if ((*n)->major == major) break; @@ -124,7 +124,7 @@ void unregister_blkdev(unsigned int major, const char *name) p = *n; *n = p->next; } - mutex_unlock(&block_subsys_lock); + mutex_unlock(&block_class_lock); kfree(p); } @@ -137,29 +137,30 @@ static struct kobj_map *bdev_map; * range must be nonzero * The hash chain is sorted on range, so that subranges can override. */ -void blk_register_region(dev_t dev, unsigned long range, struct module *module, +void blk_register_region(dev_t devt, unsigned long range, struct module *module, struct kobject *(*probe)(dev_t, int *, void *), int (*lock)(dev_t, void *), void *data) { - kobj_map(bdev_map, dev, range, module, probe, lock, data); + kobj_map(bdev_map, devt, range, module, probe, lock, data); } EXPORT_SYMBOL(blk_register_region); -void blk_unregister_region(dev_t dev, unsigned long range) +void blk_unregister_region(dev_t devt, unsigned long range) { - kobj_unmap(bdev_map, dev, range); + kobj_unmap(bdev_map, devt, range); } EXPORT_SYMBOL(blk_unregister_region); -static struct kobject *exact_match(dev_t dev, int *part, void *data) +static struct kobject *exact_match(dev_t devt, int *part, void *data) { struct gendisk *p = data; - return &p->kobj; + + return &p->dev.kobj; } -static int exact_lock(dev_t dev, void *data) +static int exact_lock(dev_t devt, void *data) { struct gendisk *p = data; @@ -194,8 +195,6 @@ void unlink_gendisk(struct gendisk *disk) disk->minors); } -#define to_disk(obj) container_of(obj,struct gendisk,kobj) - /** * get_gendisk - get partitioning information for a given device * @dev: device to get partitioning information for @@ -203,10 +202,12 @@ void unlink_gendisk(struct gendisk *disk) * This function gets the structure containing partitioning * information for the given device @dev. */ -struct gendisk *get_gendisk(dev_t dev, int *part) +struct gendisk *get_gendisk(dev_t devt, int *part) { - struct kobject *kobj = kobj_lookup(bdev_map, dev, part); - return kobj ? to_disk(kobj) : NULL; + struct kobject *kobj = kobj_lookup(bdev_map, devt, part); + struct device *dev = kobj_to_dev(kobj); + + return kobj ? dev_to_disk(dev) : NULL; } /* @@ -216,13 +217,17 @@ struct gendisk *get_gendisk(dev_t dev, int *part) */ void __init printk_all_partitions(void) { - int n; + struct device *dev; struct gendisk *sgp; + char buf[BDEVNAME_SIZE]; + int n; - mutex_lock(&block_subsys_lock); + mutex_lock(&block_class_lock); /* For each block device... */ - list_for_each_entry(sgp, &block_subsys.list, kobj.entry) { - char buf[BDEVNAME_SIZE]; + list_for_each_entry(dev, &block_class.devices, node) { + if (dev->type != &disk_type) + continue; + sgp = dev_to_disk(dev); /* * Don't show empty devices or things that have been surpressed */ @@ -255,38 +260,46 @@ void __init printk_all_partitions(void) sgp->major, n + 1 + sgp->first_minor, (unsigned long long)sgp->part[n]->nr_sects >> 1, disk_name(sgp, n + 1, buf)); - } /* partition subloop */ - } /* Block device loop */ + } + } - mutex_unlock(&block_subsys_lock); - return; + mutex_unlock(&block_class_lock); } #ifdef CONFIG_PROC_FS /* iterator */ static void *part_start(struct seq_file *part, loff_t *pos) { - struct list_head *p; - loff_t l = *pos; + loff_t k = *pos; + struct device *dev; - mutex_lock(&block_subsys_lock); - list_for_each(p, &block_subsys.list) - if (!l--) - return list_entry(p, struct gendisk, kobj.entry); + mutex_lock(&block_class_lock); + list_for_each_entry(dev, &block_class.devices, node) { + if (dev->type != &disk_type) + continue; + if (!k--) + return dev_to_disk(dev); + } return NULL; } static void *part_next(struct seq_file *part, void *v, loff_t *pos) { - struct list_head *p = ((struct gendisk *)v)->kobj.entry.next; + struct gendisk *gp = v; + struct device *dev; ++*pos; - return p==&block_subsys.list ? NULL : - list_entry(p, struct gendisk, kobj.entry); + list_for_each_entry(dev, &gp->dev.node, node) { + if (&dev->node == &block_class.devices) + return NULL; + if (dev->type == &disk_type) + return dev_to_disk(dev); + } + return NULL; } static void part_stop(struct seq_file *part, void *v) { - mutex_unlock(&block_subsys_lock); + mutex_unlock(&block_class_lock); } static int show_partition(struct seq_file *part, void *v) @@ -295,7 +308,7 @@ static int show_partition(struct seq_file *part, void *v) int n; char buf[BDEVNAME_SIZE]; - if (&sgp->kobj.entry == block_subsys.list.next) + if (&sgp->dev.node == block_class.devices.next) seq_puts(part, "major minor #blocks name\n\n"); /* Don't show non-partitionable removeable devices or empty devices */ @@ -324,111 +337,82 @@ static int show_partition(struct seq_file *part, void *v) return 0; } -struct seq_operations partitions_op = { - .start =part_start, - .next = part_next, - .stop = part_stop, - .show = show_partition +const struct seq_operations partitions_op = { + .start = part_start, + .next = part_next, + .stop = part_stop, + .show = show_partition }; #endif extern int blk_dev_init(void); -static struct kobject *base_probe(dev_t dev, int *part, void *data) +static struct kobject *base_probe(dev_t devt, int *part, void *data) { - if (request_module("block-major-%d-%d", MAJOR(dev), MINOR(dev)) > 0) + if (request_module("block-major-%d-%d", MAJOR(devt), MINOR(devt)) > 0) /* Make old-style 2.4 aliases work */ - request_module("block-major-%d", MAJOR(dev)); + request_module("block-major-%d", MAJOR(devt)); return NULL; } static int __init genhd_device_init(void) { - int err; - - bdev_map = kobj_map_init(base_probe, &block_subsys_lock); + class_register(&block_class); + bdev_map = kobj_map_init(base_probe, &block_class_lock); blk_dev_init(); - err = subsystem_register(&block_subsys); - if (err < 0) - printk(KERN_WARNING "%s: subsystem_register error: %d\n", - __FUNCTION__, err); - return err; + +#ifndef CONFIG_SYSFS_DEPRECATED + /* create top-level block dir */ + block_depr = kobject_create_and_add("block", NULL); +#endif + return 0; } subsys_initcall(genhd_device_init); - - -/* - * kobject & sysfs bindings for block devices - */ -static ssize_t disk_attr_show(struct kobject *kobj, struct attribute *attr, - char *page) +static ssize_t disk_range_show(struct device *dev, + struct device_attribute *attr, char *buf) { - struct gendisk *disk = to_disk(kobj); - struct disk_attribute *disk_attr = - container_of(attr,struct disk_attribute,attr); - ssize_t ret = -EIO; + struct gendisk *disk = dev_to_disk(dev); - if (disk_attr->show) - ret = disk_attr->show(disk,page); - return ret; + return sprintf(buf, "%d\n", disk->minors); } -static ssize_t disk_attr_store(struct kobject * kobj, struct attribute * attr, - const char *page, size_t count) +static ssize_t disk_removable_show(struct device *dev, + struct device_attribute *attr, char *buf) { - struct gendisk *disk = to_disk(kobj); - struct disk_attribute *disk_attr = - container_of(attr,struct disk_attribute,attr); - ssize_t ret = 0; + struct gendisk *disk = dev_to_disk(dev); - if (disk_attr->store) - ret = disk_attr->store(disk, page, count); - return ret; + return sprintf(buf, "%d\n", + (disk->flags & GENHD_FL_REMOVABLE ? 1 : 0)); } -static struct sysfs_ops disk_sysfs_ops = { - .show = &disk_attr_show, - .store = &disk_attr_store, -}; - -static ssize_t disk_uevent_store(struct gendisk * disk, - const char *buf, size_t count) -{ - kobject_uevent(&disk->kobj, KOBJ_ADD); - return count; -} -static ssize_t disk_dev_read(struct gendisk * disk, char *page) -{ - dev_t base = MKDEV(disk->major, disk->first_minor); - return print_dev_t(page, base); -} -static ssize_t disk_range_read(struct gendisk * disk, char *page) +static ssize_t disk_size_show(struct device *dev, + struct device_attribute *attr, char *buf) { - return sprintf(page, "%d\n", disk->minors); -} -static ssize_t disk_removable_read(struct gendisk * disk, char *page) -{ - return sprintf(page, "%d\n", - (disk->flags & GENHD_FL_REMOVABLE ? 1 : 0)); + struct gendisk *disk = dev_to_disk(dev); + return sprintf(buf, "%llu\n", (unsigned long long)get_capacity(disk)); } -static ssize_t disk_size_read(struct gendisk * disk, char *page) -{ - return sprintf(page, "%llu\n", (unsigned long long)get_capacity(disk)); -} -static ssize_t disk_capability_read(struct gendisk *disk, char *page) + +static ssize_t disk_capability_show(struct device *dev, + struct device_attribute *attr, char *buf) { - return sprintf(page, "%x\n", disk->flags); + struct gendisk *disk = dev_to_disk(dev); + + return sprintf(buf, "%x\n", disk->flags); } -static ssize_t disk_stats_read(struct gendisk * disk, char *page) + +static ssize_t disk_stat_show(struct device *dev, + struct device_attribute *attr, char *buf) { + struct gendisk *disk = dev_to_disk(dev); + preempt_disable(); disk_round_stats(disk); preempt_enable(); - return sprintf(page, + return sprintf(buf, "%8lu %8lu %8llu %8u " "%8lu %8lu %8llu %8u " "%8u %8u %8u" @@ -445,40 +429,21 @@ static ssize_t disk_stats_read(struct gendisk * disk, char *page) jiffies_to_msecs(disk_stat_read(disk, io_ticks)), jiffies_to_msecs(disk_stat_read(disk, time_in_queue))); } -static struct disk_attribute disk_attr_uevent = { - .attr = {.name = "uevent", .mode = S_IWUSR }, - .store = disk_uevent_store -}; -static struct disk_attribute disk_attr_dev = { - .attr = {.name = "dev", .mode = S_IRUGO }, - .show = disk_dev_read -}; -static struct disk_attribute disk_attr_range = { - .attr = {.name = "range", .mode = S_IRUGO }, - .show = disk_range_read -}; -static struct disk_attribute disk_attr_removable = { - .attr = {.name = "removable", .mode = S_IRUGO }, - .show = disk_removable_read -}; -static struct disk_attribute disk_attr_size = { - .attr = {.name = "size", .mode = S_IRUGO }, - .show = disk_size_read -}; -static struct disk_attribute disk_attr_capability = { - .attr = {.name = "capability", .mode = S_IRUGO }, - .show = disk_capability_read -}; -static struct disk_attribute disk_attr_stat = { - .attr = {.name = "stat", .mode = S_IRUGO }, - .show = disk_stats_read -}; #ifdef CONFIG_FAIL_MAKE_REQUEST +static ssize_t disk_fail_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct gendisk *disk = dev_to_disk(dev); + + return sprintf(buf, "%d\n", disk->flags & GENHD_FL_FAIL ? 1 : 0); +} -static ssize_t disk_fail_store(struct gendisk * disk, +static ssize_t disk_fail_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) { + struct gendisk *disk = dev_to_disk(dev); int i; if (count > 0 && sscanf(buf, "%d", &i) > 0) { @@ -490,136 +455,100 @@ static ssize_t disk_fail_store(struct gendisk * disk, return count; } -static ssize_t disk_fail_read(struct gendisk * disk, char *page) -{ - return sprintf(page, "%d\n", disk->flags & GENHD_FL_FAIL ? 1 : 0); -} -static struct disk_attribute disk_attr_fail = { - .attr = {.name = "make-it-fail", .mode = S_IRUGO | S_IWUSR }, - .store = disk_fail_store, - .show = disk_fail_read -}; #endif -static struct attribute * default_attrs[] = { - &disk_attr_uevent.attr, - &disk_attr_dev.attr, - &disk_attr_range.attr, - &disk_attr_removable.attr, - &disk_attr_size.attr, - &disk_attr_stat.attr, - &disk_attr_capability.attr, +static DEVICE_ATTR(range, S_IRUGO, disk_range_show, NULL); +static DEVICE_ATTR(removable, S_IRUGO, disk_removable_show, NULL); +static DEVICE_ATTR(size, S_IRUGO, disk_size_show, NULL); +static DEVICE_ATTR(capability, S_IRUGO, disk_capability_show, NULL); +static DEVICE_ATTR(stat, S_IRUGO, disk_stat_show, NULL); +#ifdef CONFIG_FAIL_MAKE_REQUEST +static struct device_attribute dev_attr_fail = + __ATTR(make-it-fail, S_IRUGO|S_IWUSR, disk_fail_show, disk_fail_store); +#endif + +static struct attribute *disk_attrs[] = { + &dev_attr_range.attr, + &dev_attr_removable.attr, + &dev_attr_size.attr, + &dev_attr_capability.attr, + &dev_attr_stat.attr, #ifdef CONFIG_FAIL_MAKE_REQUEST - &disk_attr_fail.attr, + &dev_attr_fail.attr, #endif - NULL, + NULL +}; + +static struct attribute_group disk_attr_group = { + .attrs = disk_attrs, }; -static void disk_release(struct kobject * kobj) +static struct attribute_group *disk_attr_groups[] = { + &disk_attr_group, + NULL +}; + +static void disk_release(struct device *dev) { - struct gendisk *disk = to_disk(kobj); + struct gendisk *disk = dev_to_disk(dev); + kfree(disk->random); kfree(disk->part); free_disk_stats(disk); kfree(disk); } - -static struct kobj_type ktype_block = { - .release = disk_release, - .sysfs_ops = &disk_sysfs_ops, - .default_attrs = default_attrs, +struct class block_class = { + .name = "block", }; -extern struct kobj_type ktype_part; - -static int block_uevent_filter(struct kset *kset, struct kobject *kobj) -{ - struct kobj_type *ktype = get_ktype(kobj); - - return ((ktype == &ktype_block) || (ktype == &ktype_part)); -} - -static int block_uevent(struct kset *kset, struct kobject *kobj, - struct kobj_uevent_env *env) -{ - struct kobj_type *ktype = get_ktype(kobj); - struct device *physdev; - struct gendisk *disk; - struct hd_struct *part; - - if (ktype == &ktype_block) { - disk = container_of(kobj, struct gendisk, kobj); - add_uevent_var(env, "MINOR=%u", disk->first_minor); - } else if (ktype == &ktype_part) { - disk = container_of(kobj->parent, struct gendisk, kobj); - part = container_of(kobj, struct hd_struct, kobj); - add_uevent_var(env, "MINOR=%u", - disk->first_minor + part->partno); - } else - return 0; - - add_uevent_var(env, "MAJOR=%u", disk->major); - - /* add physical device, backing this device */ - physdev = disk->driverfs_dev; - if (physdev) { - char *path = kobject_get_path(&physdev->kobj, GFP_KERNEL); - - add_uevent_var(env, "PHYSDEVPATH=%s", path); - kfree(path); - - if (physdev->bus) - add_uevent_var(env, "PHYSDEVBUS=%s", physdev->bus->name); - - if (physdev->driver) - add_uevent_var(env, physdev->driver->name); - } - - return 0; -} - -static struct kset_uevent_ops block_uevent_ops = { - .filter = block_uevent_filter, - .uevent = block_uevent, +struct device_type disk_type = { + .name = "disk", + .groups = disk_attr_groups, + .release = disk_release, }; -decl_subsys(block, &ktype_block, &block_uevent_ops); - /* * aggregate disk stat collector. Uses the same stats that the sysfs * entries do, above, but makes them available through one seq_file. - * Watching a few disks may be efficient through sysfs, but watching - * all of them will be more efficient through this interface. * * The output looks suspiciously like /proc/partitions with a bunch of * extra fields. */ -/* iterator */ static void *diskstats_start(struct seq_file *part, loff_t *pos) { loff_t k = *pos; - struct list_head *p; + struct device *dev; - mutex_lock(&block_subsys_lock); - list_for_each(p, &block_subsys.list) + mutex_lock(&block_class_lock); + list_for_each_entry(dev, &block_class.devices, node) { + if (dev->type != &disk_type) + continue; if (!k--) - return list_entry(p, struct gendisk, kobj.entry); + return dev_to_disk(dev); + } return NULL; } static void *diskstats_next(struct seq_file *part, void *v, loff_t *pos) { - struct list_head *p = ((struct gendisk *)v)->kobj.entry.next; + struct gendisk *gp = v; + struct device *dev; + ++*pos; - return p==&block_subsys.list ? NULL : - list_entry(p, struct gendisk, kobj.entry); + list_for_each_entry(dev, &gp->dev.node, node) { + if (&dev->node == &block_class.devices) + return NULL; + if (dev->type == &disk_type) + return dev_to_disk(dev); + } + return NULL; } static void diskstats_stop(struct seq_file *part, void *v) { - mutex_unlock(&block_subsys_lock); + mutex_unlock(&block_class_lock); } static int diskstats_show(struct seq_file *s, void *v) @@ -629,7 +558,7 @@ static int diskstats_show(struct seq_file *s, void *v) int n = 0; /* - if (&sgp->kobj.entry == block_subsys.kset.list.next) + if (&gp->dev.kobj.entry == block_class.devices.next) seq_puts(s, "major minor name" " rio rmerge rsect ruse wio wmerge " "wsect wuse running use aveq" @@ -666,7 +595,7 @@ static int diskstats_show(struct seq_file *s, void *v) return 0; } -struct seq_operations diskstats_op = { +const struct seq_operations diskstats_op = { .start = diskstats_start, .next = diskstats_next, .stop = diskstats_stop, @@ -683,7 +612,7 @@ static void media_change_notify_thread(struct work_struct *work) * set enviroment vars to indicate which event this is for * so that user space will know to go check the media status. */ - kobject_uevent_env(&gd->kobj, KOBJ_CHANGE, envp); + kobject_uevent_env(&gd->dev.kobj, KOBJ_CHANGE, envp); put_device(gd->driverfs_dev); } @@ -694,6 +623,25 @@ void genhd_media_change_notify(struct gendisk *disk) } EXPORT_SYMBOL_GPL(genhd_media_change_notify); +dev_t blk_lookup_devt(const char *name) +{ + struct device *dev; + dev_t devt = MKDEV(0, 0); + + mutex_lock(&block_class_lock); + list_for_each_entry(dev, &block_class.devices, node) { + if (strcmp(dev->bus_id, name) == 0) { + devt = dev->devt; + break; + } + } + mutex_unlock(&block_class_lock); + + return devt; +} + +EXPORT_SYMBOL(blk_lookup_devt); + struct gendisk *alloc_disk(int minors) { return alloc_disk_node(minors, -1); @@ -721,9 +669,10 @@ struct gendisk *alloc_disk_node(int minors, int node_id) } } disk->minors = minors; - kobj_set_kset_s(disk,block_subsys); - kobject_init(&disk->kobj); rand_initialize_disk(disk); + disk->dev.class = &block_class; + disk->dev.type = &disk_type; + device_initialize(&disk->dev); INIT_WORK(&disk->async_notify, media_change_notify_thread); } @@ -743,7 +692,7 @@ struct kobject *get_disk(struct gendisk *disk) owner = disk->fops->owner; if (owner && !try_module_get(owner)) return NULL; - kobj = kobject_get(&disk->kobj); + kobj = kobject_get(&disk->dev.kobj); if (kobj == NULL) { module_put(owner); return NULL; @@ -757,7 +706,7 @@ EXPORT_SYMBOL(get_disk); void put_disk(struct gendisk *disk) { if (disk) - kobject_put(&disk->kobj); + kobject_put(&disk->dev.kobj); } EXPORT_SYMBOL(put_disk); diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c deleted file mode 100644 index 8b919940b2a..00000000000 --- a/block/ll_rw_blk.c +++ /dev/null @@ -1,4214 +0,0 @@ -/* - * Copyright (C) 1991, 1992 Linus Torvalds - * Copyright (C) 1994, Karl Keyte: Added support for disk statistics - * Elevator latency, (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE - * Queue request tables / lock, selectable elevator, Jens Axboe <axboe@suse.de> - * kernel-doc documentation started by NeilBrown <neilb@cse.unsw.edu.au> - July2000 - * bio rewrite, highmem i/o, etc, Jens Axboe <axboe@suse.de> - may 2001 - */ - -/* - * This handles all read/write requests to block devices - */ -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/backing-dev.h> -#include <linux/bio.h> -#include <linux/blkdev.h> -#include <linux/highmem.h> -#include <linux/mm.h> -#include <linux/kernel_stat.h> -#include <linux/string.h> -#include <linux/init.h> -#include <linux/bootmem.h> /* for max_pfn/max_low_pfn */ -#include <linux/completion.h> -#include <linux/slab.h> -#include <linux/swap.h> -#include <linux/writeback.h> -#include <linux/task_io_accounting_ops.h> -#include <linux/interrupt.h> -#include <linux/cpu.h> -#include <linux/blktrace_api.h> -#include <linux/fault-inject.h> -#include <linux/scatterlist.h> - -/* - * for max sense size - */ -#include <scsi/scsi_cmnd.h> - -static void blk_unplug_work(struct work_struct *work); -static void blk_unplug_timeout(unsigned long data); -static void drive_stat_acct(struct request *rq, int new_io); -static void init_request_from_bio(struct request *req, struct bio *bio); -static int __make_request(struct request_queue *q, struct bio *bio); -static struct io_context *current_io_context(gfp_t gfp_flags, int node); -static void blk_recalc_rq_segments(struct request *rq); -static void blk_rq_bio_prep(struct request_queue *q, struct request *rq, - struct bio *bio); - -/* - * For the allocated request tables - */ -static struct kmem_cache *request_cachep; - -/* - * For queue allocation - */ -static struct kmem_cache *requestq_cachep; - -/* - * For io context allocations - */ -static struct kmem_cache *iocontext_cachep; - -/* - * Controlling structure to kblockd - */ -static struct workqueue_struct *kblockd_workqueue; - -unsigned long blk_max_low_pfn, blk_max_pfn; - -EXPORT_SYMBOL(blk_max_low_pfn); -EXPORT_SYMBOL(blk_max_pfn); - -static DEFINE_PER_CPU(struct list_head, blk_cpu_done); - -/* Amount of time in which a process may batch requests */ -#define BLK_BATCH_TIME (HZ/50UL) - -/* Number of requests a "batching" process may submit */ -#define BLK_BATCH_REQ 32 - -/* - * Return the threshold (number of used requests) at which the queue is - * considered to be congested. It include a little hysteresis to keep the - * context switch rate down. - */ -static inline int queue_congestion_on_threshold(struct request_queue *q) -{ - return q->nr_congestion_on; -} - -/* - * The threshold at which a queue is considered to be uncongested - */ -static inline int queue_congestion_off_threshold(struct request_queue *q) -{ - return q->nr_congestion_off; -} - -static void blk_queue_congestion_threshold(struct request_queue *q) -{ - int nr; - - nr = q->nr_requests - (q->nr_requests / 8) + 1; - if (nr > q->nr_requests) - nr = q->nr_requests; - q->nr_congestion_on = nr; - - nr = q->nr_requests - (q->nr_requests / 8) - (q->nr_requests / 16) - 1; - if (nr < 1) - nr = 1; - q->nr_congestion_off = nr; -} - -/** - * blk_get_backing_dev_info - get the address of a queue's backing_dev_info - * @bdev: device - * - * Locates the passed device's request queue and returns the address of its - * backing_dev_info - * - * Will return NULL if the request queue cannot be located. - */ -struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev) -{ - struct backing_dev_info *ret = NULL; - struct request_queue *q = bdev_get_queue(bdev); - - if (q) - ret = &q->backing_dev_info; - return ret; -} -EXPORT_SYMBOL(blk_get_backing_dev_info); - -/** - * blk_queue_prep_rq - set a prepare_request function for queue - * @q: queue - * @pfn: prepare_request function - * - * It's possible for a queue to register a prepare_request callback which - * is invoked before the request is handed to the request_fn. The goal of - * the function is to prepare a request for I/O, it can be used to build a - * cdb from the request data for instance. - * - */ -void blk_queue_prep_rq(struct request_queue *q, prep_rq_fn *pfn) -{ - q->prep_rq_fn = pfn; -} - -EXPORT_SYMBOL(blk_queue_prep_rq); - -/** - * blk_queue_merge_bvec - set a merge_bvec function for queue - * @q: queue - * @mbfn: merge_bvec_fn - * - * Usually queues have static limitations on the max sectors or segments that - * we can put in a request. Stacking drivers may have some settings that - * are dynamic, and thus we have to query the queue whether it is ok to - * add a new bio_vec to a bio at a given offset or not. If the block device - * has such limitations, it needs to register a merge_bvec_fn to control - * the size of bio's sent to it. Note that a block device *must* allow a - * single page to be added to an empty bio. The block device driver may want - * to use the bio_split() function to deal with these bio's. By default - * no merge_bvec_fn is defined for a queue, and only the fixed limits are - * honored. - */ -void blk_queue_merge_bvec(struct request_queue *q, merge_bvec_fn *mbfn) -{ - q->merge_bvec_fn = mbfn; -} - -EXPORT_SYMBOL(blk_queue_merge_bvec); - -void blk_queue_softirq_done(struct request_queue *q, softirq_done_fn *fn) -{ - q->softirq_done_fn = fn; -} - -EXPORT_SYMBOL(blk_queue_softirq_done); - -/** - * blk_queue_make_request - define an alternate make_request function for a device - * @q: the request queue for the device to be affected - * @mfn: the alternate make_request function - * - * Description: - * The normal way for &struct bios to be passed to a device - * driver is for them to be collected into requests on a request - * queue, and then to allow the device driver to select requests - * off that queue when it is ready. This works well for many block - * devices. However some block devices (typically virtual devices - * such as md or lvm) do not benefit from the processing on the - * request queue, and are served best by having the requests passed - * directly to them. This can be achieved by providing a function - * to blk_queue_make_request(). - * - * Caveat: - * The driver that does this *must* be able to deal appropriately - * with buffers in "highmemory". This can be accomplished by either calling - * __bio_kmap_atomic() to get a temporary kernel mapping, or by calling - * blk_queue_bounce() to create a buffer in normal memory. - **/ -void blk_queue_make_request(struct request_queue * q, make_request_fn * mfn) -{ - /* - * set defaults - */ - q->nr_requests = BLKDEV_MAX_RQ; - blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS); - blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS); - q->make_request_fn = mfn; - q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; - q->backing_dev_info.state = 0; - q->backing_dev_info.capabilities = BDI_CAP_MAP_COPY; - blk_queue_max_sectors(q, SAFE_MAX_SECTORS); - blk_queue_hardsect_size(q, 512); - blk_queue_dma_alignment(q, 511); - blk_queue_congestion_threshold(q); - q->nr_batching = BLK_BATCH_REQ; - - q->unplug_thresh = 4; /* hmm */ - q->unplug_delay = (3 * HZ) / 1000; /* 3 milliseconds */ - if (q->unplug_delay == 0) - q->unplug_delay = 1; - - INIT_WORK(&q->unplug_work, blk_unplug_work); - - q->unplug_timer.function = blk_unplug_timeout; - q->unplug_timer.data = (unsigned long)q; - - /* - * by default assume old behaviour and bounce for any highmem page - */ - blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH); -} - -EXPORT_SYMBOL(blk_queue_make_request); - -static void rq_init(struct request_queue *q, struct request *rq) -{ - INIT_LIST_HEAD(&rq->queuelist); - INIT_LIST_HEAD(&rq->donelist); - - rq->errors = 0; - rq->bio = rq->biotail = NULL; - INIT_HLIST_NODE(&rq->hash); - RB_CLEAR_NODE(&rq->rb_node); - rq->ioprio = 0; - rq->buffer = NULL; - rq->ref_count = 1; - rq->q = q; - rq->special = NULL; - rq->data_len = 0; - rq->data = NULL; - rq->nr_phys_segments = 0; - rq->sense = NULL; - rq->end_io = NULL; - rq->end_io_data = NULL; - rq->completion_data = NULL; - rq->next_rq = NULL; -} - -/** - * blk_queue_ordered - does this queue support ordered writes - * @q: the request queue - * @ordered: one of QUEUE_ORDERED_* - * @prepare_flush_fn: rq setup helper for cache flush ordered writes - * - * Description: - * For journalled file systems, doing ordered writes on a commit - * block instead of explicitly doing wait_on_buffer (which is bad - * for performance) can be a big win. Block drivers supporting this - * feature should call this function and indicate so. - * - **/ -int blk_queue_ordered(struct request_queue *q, unsigned ordered, - prepare_flush_fn *prepare_flush_fn) -{ - if (ordered & (QUEUE_ORDERED_PREFLUSH | QUEUE_ORDERED_POSTFLUSH) && - prepare_flush_fn == NULL) { - printk(KERN_ERR "blk_queue_ordered: prepare_flush_fn required\n"); - return -EINVAL; - } - - if (ordered != QUEUE_ORDERED_NONE && - ordered != QUEUE_ORDERED_DRAIN && - ordered != QUEUE_ORDERED_DRAIN_FLUSH && - ordered != QUEUE_ORDERED_DRAIN_FUA && - ordered != QUEUE_ORDERED_TAG && - ordered != QUEUE_ORDERED_TAG_FLUSH && - ordered != QUEUE_ORDERED_TAG_FUA) { - printk(KERN_ERR "blk_queue_ordered: bad value %d\n", ordered); - return -EINVAL; - } - - q->ordered = ordered; - q->next_ordered = ordered; - q->prepare_flush_fn = prepare_flush_fn; - - return 0; -} - -EXPORT_SYMBOL(blk_queue_ordered); - -/* - * Cache flushing for ordered writes handling - */ -inline unsigned blk_ordered_cur_seq(struct request_queue *q) -{ - if (!q->ordseq) - return 0; - return 1 << ffz(q->ordseq); -} - -unsigned blk_ordered_req_seq(struct request *rq) -{ - struct request_queue *q = rq->q; - - BUG_ON(q->ordseq == 0); - - if (rq == &q->pre_flush_rq) - return QUEUE_ORDSEQ_PREFLUSH; - if (rq == &q->bar_rq) - return QUEUE_ORDSEQ_BAR; - if (rq == &q->post_flush_rq) - return QUEUE_ORDSEQ_POSTFLUSH; - - /* - * !fs requests don't need to follow barrier ordering. Always - * put them at the front. This fixes the following deadlock. - * - * http://thread.gmane.org/gmane.linux.kernel/537473 - */ - if (!blk_fs_request(rq)) - return QUEUE_ORDSEQ_DRAIN; - - if ((rq->cmd_flags & REQ_ORDERED_COLOR) == - (q->orig_bar_rq->cmd_flags & REQ_ORDERED_COLOR)) - return QUEUE_ORDSEQ_DRAIN; - else - return QUEUE_ORDSEQ_DONE; -} - -void blk_ordered_complete_seq(struct request_queue *q, unsigned seq, int error) -{ - struct request *rq; - int uptodate; - - if (error && !q->orderr) - q->orderr = error; - - BUG_ON(q->ordseq & seq); - q->ordseq |= seq; - - if (blk_ordered_cur_seq(q) != QUEUE_ORDSEQ_DONE) - return; - - /* - * Okay, sequence complete. - */ - uptodate = 1; - if (q->orderr) - uptodate = q->orderr; - - q->ordseq = 0; - rq = q->orig_bar_rq; - - end_that_request_first(rq, uptodate, rq->hard_nr_sectors); - end_that_request_last(rq, uptodate); -} - -static void pre_flush_end_io(struct request *rq, int error) -{ - elv_completed_request(rq->q, rq); - blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_PREFLUSH, error); -} - -static void bar_end_io(struct request *rq, int error) -{ - elv_completed_request(rq->q, rq); - blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_BAR, error); -} - -static void post_flush_end_io(struct request *rq, int error) -{ - elv_completed_request(rq->q, rq); - blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_POSTFLUSH, error); -} - -static void queue_flush(struct request_queue *q, unsigned which) -{ - struct request *rq; - rq_end_io_fn *end_io; - - if (which == QUEUE_ORDERED_PREFLUSH) { - rq = &q->pre_flush_rq; - end_io = pre_flush_end_io; - } else { - rq = &q->post_flush_rq; - end_io = post_flush_end_io; - } - - rq->cmd_flags = REQ_HARDBARRIER; - rq_init(q, rq); - rq->elevator_private = NULL; - rq->elevator_private2 = NULL; - rq->rq_disk = q->bar_rq.rq_disk; - rq->end_io = end_io; - q->prepare_flush_fn(q, rq); - - elv_insert(q, rq, ELEVATOR_INSERT_FRONT); -} - -static inline struct request *start_ordered(struct request_queue *q, - struct request *rq) -{ - q->orderr = 0; - q->ordered = q->next_ordered; - q->ordseq |= QUEUE_ORDSEQ_STARTED; - - /* - * Prep proxy barrier request. - */ - blkdev_dequeue_request(rq); - q->orig_bar_rq = rq; - rq = &q->bar_rq; - rq->cmd_flags = 0; - rq_init(q, rq); - if (bio_data_dir(q->orig_bar_rq->bio) == WRITE) - rq->cmd_flags |= REQ_RW; - if (q->ordered & QUEUE_ORDERED_FUA) - rq->cmd_flags |= REQ_FUA; - rq->elevator_private = NULL; - rq->elevator_private2 = NULL; - init_request_from_bio(rq, q->orig_bar_rq->bio); - rq->end_io = bar_end_io; - - /* - * Queue ordered sequence. As we stack them at the head, we - * need to queue in reverse order. Note that we rely on that - * no fs request uses ELEVATOR_INSERT_FRONT and thus no fs - * request gets inbetween ordered sequence. If this request is - * an empty barrier, we don't need to do a postflush ever since - * there will be no data written between the pre and post flush. - * Hence a single flush will suffice. - */ - if ((q->ordered & QUEUE_ORDERED_POSTFLUSH) && !blk_empty_barrier(rq)) - queue_flush(q, QUEUE_ORDERED_POSTFLUSH); - else - q->ordseq |= QUEUE_ORDSEQ_POSTFLUSH; - - elv_insert(q, rq, ELEVATOR_INSERT_FRONT); - - if (q->ordered & QUEUE_ORDERED_PREFLUSH) { - queue_flush(q, QUEUE_ORDERED_PREFLUSH); - rq = &q->pre_flush_rq; - } else - q->ordseq |= QUEUE_ORDSEQ_PREFLUSH; - - if ((q->ordered & QUEUE_ORDERED_TAG) || q->in_flight == 0) - q->ordseq |= QUEUE_ORDSEQ_DRAIN; - else - rq = NULL; - - return rq; -} - -int blk_do_ordered(struct request_queue *q, struct request **rqp) -{ - struct request *rq = *rqp; - const int is_barrier = blk_fs_request(rq) && blk_barrier_rq(rq); - - if (!q->ordseq) { - if (!is_barrier) - return 1; - - if (q->next_ordered != QUEUE_ORDERED_NONE) { - *rqp = start_ordered(q, rq); - return 1; - } else { - /* - * This can happen when the queue switches to - * ORDERED_NONE while this request is on it. - */ - blkdev_dequeue_request(rq); - end_that_request_first(rq, -EOPNOTSUPP, - rq->hard_nr_sectors); - end_that_request_last(rq, -EOPNOTSUPP); - *rqp = NULL; - return 0; - } - } - - /* - * Ordered sequence in progress - */ - - /* Special requests are not subject to ordering rules. */ - if (!blk_fs_request(rq) && - rq != &q->pre_flush_rq && rq != &q->post_flush_rq) - return 1; - - if (q->ordered & QUEUE_ORDERED_TAG) { - /* Ordered by tag. Blocking the next barrier is enough. */ - if (is_barrier && rq != &q->bar_rq) - *rqp = NULL; - } else { - /* Ordered by draining. Wait for turn. */ - WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q)); - if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q)) - *rqp = NULL; - } - - return 1; -} - -static void req_bio_endio(struct request *rq, struct bio *bio, - unsigned int nbytes, int error) -{ - struct request_queue *q = rq->q; - - if (&q->bar_rq != rq) { - if (error) - clear_bit(BIO_UPTODATE, &bio->bi_flags); - else if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) - error = -EIO; - - if (unlikely(nbytes > bio->bi_size)) { - printk("%s: want %u bytes done, only %u left\n", - __FUNCTION__, nbytes, bio->bi_size); - nbytes = bio->bi_size; - } - - bio->bi_size -= nbytes; - bio->bi_sector += (nbytes >> 9); - if (bio->bi_size == 0) - bio_endio(bio, error); - } else { - - /* - * Okay, this is the barrier request in progress, just - * record the error; - */ - if (error && !q->orderr) - q->orderr = error; - } -} - -/** - * blk_queue_bounce_limit - set bounce buffer limit for queue - * @q: the request queue for the device - * @dma_addr: bus address limit - * - * Description: - * Different hardware can have different requirements as to what pages - * it can do I/O directly to. A low level driver can call - * blk_queue_bounce_limit to have lower memory pages allocated as bounce - * buffers for doing I/O to pages residing above @page. - **/ -void blk_queue_bounce_limit(struct request_queue *q, u64 dma_addr) -{ - unsigned long bounce_pfn = dma_addr >> PAGE_SHIFT; - int dma = 0; - - q->bounce_gfp = GFP_NOIO; -#if BITS_PER_LONG == 64 - /* Assume anything <= 4GB can be handled by IOMMU. - Actually some IOMMUs can handle everything, but I don't - know of a way to test this here. */ - if (bounce_pfn < (min_t(u64,0xffffffff,BLK_BOUNCE_HIGH) >> PAGE_SHIFT)) - dma = 1; - q->bounce_pfn = max_low_pfn; -#else - if (bounce_pfn < blk_max_low_pfn) - dma = 1; - q->bounce_pfn = bounce_pfn; -#endif - if (dma) { - init_emergency_isa_pool(); - q->bounce_gfp = GFP_NOIO | GFP_DMA; - q->bounce_pfn = bounce_pfn; - } -} - -EXPORT_SYMBOL(blk_queue_bounce_limit); - -/** - * blk_queue_max_sectors - set max sectors for a request for this queue - * @q: the request queue for the device - * @max_sectors: max sectors in the usual 512b unit - * - * Description: - * Enables a low level driver to set an upper limit on the size of - * received requests. - **/ -void blk_queue_max_sectors(struct request_queue *q, unsigned int max_sectors) -{ - if ((max_sectors << 9) < PAGE_CACHE_SIZE) { - max_sectors = 1 << (PAGE_CACHE_SHIFT - 9); - printk("%s: set to minimum %d\n", __FUNCTION__, max_sectors); - } - - if (BLK_DEF_MAX_SECTORS > max_sectors) - q->max_hw_sectors = q->max_sectors = max_sectors; - else { - q->max_sectors = BLK_DEF_MAX_SECTORS; - q->max_hw_sectors = max_sectors; - } -} - -EXPORT_SYMBOL(blk_queue_max_sectors); - -/** - * blk_queue_max_phys_segments - set max phys segments for a request for this queue - * @q: the request queue for the device - * @max_segments: max number of segments - * - * Description: - * Enables a low level driver to set an upper limit on the number of - * physical data segments in a request. This would be the largest sized - * scatter list the driver could handle. - **/ -void blk_queue_max_phys_segments(struct request_queue *q, - unsigned short max_segments) -{ - if (!max_segments) { - max_segments = 1; - printk("%s: set to minimum %d\n", __FUNCTION__, max_segments); - } - - q->max_phys_segments = max_segments; -} - -EXPORT_SYMBOL(blk_queue_max_phys_segments); - -/** - * blk_queue_max_hw_segments - set max hw segments for a request for this queue - * @q: the request queue for the device - * @max_segments: max number of segments - * - * Description: - * Enables a low level driver to set an upper limit on the number of - * hw data segments in a request. This would be the largest number of - * address/length pairs the host adapter can actually give as once - * to the device. - **/ -void blk_queue_max_hw_segments(struct request_queue *q, - unsigned short max_segments) -{ - if (!max_segments) { - max_segments = 1; - printk("%s: set to minimum %d\n", __FUNCTION__, max_segments); - } - - q->max_hw_segments = max_segments; -} - -EXPORT_SYMBOL(blk_queue_max_hw_segments); - -/** - * blk_queue_max_segment_size - set max segment size for blk_rq_map_sg - * @q: the request queue for the device - * @max_size: max size of segment in bytes - * - * Description: - * Enables a low level driver to set an upper limit on the size of a - * coalesced segment - **/ -void blk_queue_max_segment_size(struct request_queue *q, unsigned int max_size) -{ - if (max_size < PAGE_CACHE_SIZE) { - max_size = PAGE_CACHE_SIZE; - printk("%s: set to minimum %d\n", __FUNCTION__, max_size); - } - - q->max_segment_size = max_size; -} - -EXPORT_SYMBOL(blk_queue_max_segment_size); - -/** - * blk_queue_hardsect_size - set hardware sector size for the queue - * @q: the request queue for the device - * @size: the hardware sector size, in bytes - * - * Description: - * This should typically be set to the lowest possible sector size - * that the hardware can operate on (possible without reverting to - * even internal read-modify-write operations). Usually the default - * of 512 covers most hardware. - **/ -void blk_queue_hardsect_size(struct request_queue *q, unsigned short size) -{ - q->hardsect_size = size; -} - -EXPORT_SYMBOL(blk_queue_hardsect_size); - -/* - * Returns the minimum that is _not_ zero, unless both are zero. - */ -#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r)) - -/** - * blk_queue_stack_limits - inherit underlying queue limits for stacked drivers - * @t: the stacking driver (top) - * @b: the underlying device (bottom) - **/ -void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b) -{ - /* zero is "infinity" */ - t->max_sectors = min_not_zero(t->max_sectors,b->max_sectors); - t->max_hw_sectors = min_not_zero(t->max_hw_sectors,b->max_hw_sectors); - - t->max_phys_segments = min(t->max_phys_segments,b->max_phys_segments); - t->max_hw_segments = min(t->max_hw_segments,b->max_hw_segments); - t->max_segment_size = min(t->max_segment_size,b->max_segment_size); - t->hardsect_size = max(t->hardsect_size,b->hardsect_size); - if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags)) - clear_bit(QUEUE_FLAG_CLUSTER, &t->queue_flags); -} - -EXPORT_SYMBOL(blk_queue_stack_limits); - -/** - * blk_queue_segment_boundary - set boundary rules for segment merging - * @q: the request queue for the device - * @mask: the memory boundary mask - **/ -void blk_queue_segment_boundary(struct request_queue *q, unsigned long mask) -{ - if (mask < PAGE_CACHE_SIZE - 1) { - mask = PAGE_CACHE_SIZE - 1; - printk("%s: set to minimum %lx\n", __FUNCTION__, mask); - } - - q->seg_boundary_mask = mask; -} - -EXPORT_SYMBOL(blk_queue_segment_boundary); - -/** - * blk_queue_dma_alignment - set dma length and memory alignment - * @q: the request queue for the device - * @mask: alignment mask - * - * description: - * set required memory and length aligment for direct dma transactions. - * this is used when buiding direct io requests for the queue. - * - **/ -void blk_queue_dma_alignment(struct request_queue *q, int mask) -{ - q->dma_alignment = mask; -} - -EXPORT_SYMBOL(blk_queue_dma_alignment); - -/** - * blk_queue_find_tag - find a request by its tag and queue - * @q: The request queue for the device - * @tag: The tag of the request - * - * Notes: - * Should be used when a device returns a tag and you want to match - * it with a request. - * - * no locks need be held. - **/ -struct request *blk_queue_find_tag(struct request_queue *q, int tag) -{ - return blk_map_queue_find_tag(q->queue_tags, tag); -} - -EXPORT_SYMBOL(blk_queue_find_tag); - -/** - * __blk_free_tags - release a given set of tag maintenance info - * @bqt: the tag map to free - * - * Tries to free the specified @bqt@. Returns true if it was - * actually freed and false if there are still references using it - */ -static int __blk_free_tags(struct blk_queue_tag *bqt) -{ - int retval; - - retval = atomic_dec_and_test(&bqt->refcnt); - if (retval) { - BUG_ON(bqt->busy); - - kfree(bqt->tag_index); - bqt->tag_index = NULL; - - kfree(bqt->tag_map); - bqt->tag_map = NULL; - - kfree(bqt); - - } - - return retval; -} - -/** - * __blk_queue_free_tags - release tag maintenance info - * @q: the request queue for the device - * - * Notes: - * blk_cleanup_queue() will take care of calling this function, if tagging - * has been used. So there's no need to call this directly. - **/ -static void __blk_queue_free_tags(struct request_queue *q) -{ - struct blk_queue_tag *bqt = q->queue_tags; - - if (!bqt) - return; - - __blk_free_tags(bqt); - - q->queue_tags = NULL; - q->queue_flags &= ~(1 << QUEUE_FLAG_QUEUED); -} - - -/** - * blk_free_tags - release a given set of tag maintenance info - * @bqt: the tag map to free - * - * For externally managed @bqt@ frees the map. Callers of this - * function must guarantee to have released all the queues that - * might have been using this tag map. - */ -void blk_free_tags(struct blk_queue_tag *bqt) -{ - if (unlikely(!__blk_free_tags(bqt))) - BUG(); -} -EXPORT_SYMBOL(blk_free_tags); - -/** - * blk_queue_free_tags - release tag maintenance info - * @q: the request queue for the device - * - * Notes: - * This is used to disabled tagged queuing to a device, yet leave - * queue in function. - **/ -void blk_queue_free_tags(struct request_queue *q) -{ - clear_bit(QUEUE_FLAG_QUEUED, &q->queue_flags); -} - -EXPORT_SYMBOL(blk_queue_free_tags); - -static int -init_tag_map(struct request_queue *q, struct blk_queue_tag *tags, int depth) -{ - struct request **tag_index; - unsigned long *tag_map; - int nr_ulongs; - - if (q && depth > q->nr_requests * 2) { - depth = q->nr_requests * 2; - printk(KERN_ERR "%s: adjusted depth to %d\n", - __FUNCTION__, depth); - } - - tag_index = kzalloc(depth * sizeof(struct request *), GFP_ATOMIC); - if (!tag_index) - goto fail; - - nr_ulongs = ALIGN(depth, BITS_PER_LONG) / BITS_PER_LONG; - tag_map = kzalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC); - if (!tag_map) - goto fail; - - tags->real_max_depth = depth; - tags->max_depth = depth; - tags->tag_index = tag_index; - tags->tag_map = tag_map; - - return 0; -fail: - kfree(tag_index); - return -ENOMEM; -} - -static struct blk_queue_tag *__blk_queue_init_tags(struct request_queue *q, - int depth) -{ - struct blk_queue_tag *tags; - - tags = kmalloc(sizeof(struct blk_queue_tag), GFP_ATOMIC); - if (!tags) - goto fail; - - if (init_tag_map(q, tags, depth)) - goto fail; - - tags->busy = 0; - atomic_set(&tags->refcnt, 1); - return tags; -fail: - kfree(tags); - return NULL; -} - -/** - * blk_init_tags - initialize the tag info for an external tag map - * @depth: the maximum queue depth supported - * @tags: the tag to use - **/ -struct blk_queue_tag *blk_init_tags(int depth) -{ - return __blk_queue_init_tags(NULL, depth); -} -EXPORT_SYMBOL(blk_init_tags); - -/** - * blk_queue_init_tags - initialize the queue tag info - * @q: the request queue for the device - * @depth: the maximum queue depth supported - * @tags: the tag to use - **/ -int blk_queue_init_tags(struct request_queue *q, int depth, - struct blk_queue_tag *tags) -{ - int rc; - - BUG_ON(tags && q->queue_tags && tags != q->queue_tags); - - if (!tags && !q->queue_tags) { - tags = __blk_queue_init_tags(q, depth); - - if (!tags) - goto fail; - } else if (q->queue_tags) { - if ((rc = blk_queue_resize_tags(q, depth))) - return rc; - set_bit(QUEUE_FLAG_QUEUED, &q->queue_flags); - return 0; - } else - atomic_inc(&tags->refcnt); - - /* - * assign it, all done - */ - q->queue_tags = tags; - q->queue_flags |= (1 << QUEUE_FLAG_QUEUED); - INIT_LIST_HEAD(&q->tag_busy_list); - return 0; -fail: - kfree(tags); - return -ENOMEM; -} - -EXPORT_SYMBOL(blk_queue_init_tags); - -/** - * blk_queue_resize_tags - change the queueing depth - * @q: the request queue for the device - * @new_depth: the new max command queueing depth - * - * Notes: - * Must be called with the queue lock held. - **/ -int blk_queue_resize_tags(struct request_queue *q, int new_depth) -{ - struct blk_queue_tag *bqt = q->queue_tags; - struct request **tag_index; - unsigned long *tag_map; - int max_depth, nr_ulongs; - - if (!bqt) - return -ENXIO; - - /* - * if we already have large enough real_max_depth. just - * adjust max_depth. *NOTE* as requests with tag value - * between new_depth and real_max_depth can be in-flight, tag - * map can not be shrunk blindly here. - */ - if (new_depth <= bqt->real_max_depth) { - bqt->max_depth = new_depth; - return 0; - } - - /* - * Currently cannot replace a shared tag map with a new - * one, so error out if this is the case - */ - if (atomic_read(&bqt->refcnt) != 1) - return -EBUSY; - - /* - * save the old state info, so we can copy it back - */ - tag_index = bqt->tag_index; - tag_map = bqt->tag_map; - max_depth = bqt->real_max_depth; - - if (init_tag_map(q, bqt, new_depth)) - return -ENOMEM; - - memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *)); - nr_ulongs = ALIGN(max_depth, BITS_PER_LONG) / BITS_PER_LONG; - memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long)); - - kfree(tag_index); - kfree(tag_map); - return 0; -} - -EXPORT_SYMBOL(blk_queue_resize_tags); - -/** - * blk_queue_end_tag - end tag operations for a request - * @q: the request queue for the device - * @rq: the request that has completed - * - * Description: - * Typically called when end_that_request_first() returns 0, meaning - * all transfers have been done for a request. It's important to call - * this function before end_that_request_last(), as that will put the - * request back on the free list thus corrupting the internal tag list. - * - * Notes: - * queue lock must be held. - **/ -void blk_queue_end_tag(struct request_queue *q, struct request *rq) -{ - struct blk_queue_tag *bqt = q->queue_tags; - int tag = rq->tag; - - BUG_ON(tag == -1); - - if (unlikely(tag >= bqt->real_max_depth)) - /* - * This can happen after tag depth has been reduced. - * FIXME: how about a warning or info message here? - */ - return; - - list_del_init(&rq->queuelist); - rq->cmd_flags &= ~REQ_QUEUED; - rq->tag = -1; - - if (unlikely(bqt->tag_index[tag] == NULL)) - printk(KERN_ERR "%s: tag %d is missing\n", - __FUNCTION__, tag); - - bqt->tag_index[tag] = NULL; - - if (unlikely(!test_bit(tag, bqt->tag_map))) { - printk(KERN_ERR "%s: attempt to clear non-busy tag (%d)\n", - __FUNCTION__, tag); - return; - } - /* - * The tag_map bit acts as a lock for tag_index[bit], so we need - * unlock memory barrier semantics. - */ - clear_bit_unlock(tag, bqt->tag_map); - bqt->busy--; -} - -EXPORT_SYMBOL(blk_queue_end_tag); - -/** - * blk_queue_start_tag - find a free tag and assign it - * @q: the request queue for the device - * @rq: the block request that needs tagging - * - * Description: - * This can either be used as a stand-alone helper, or possibly be - * assigned as the queue &prep_rq_fn (in which case &struct request - * automagically gets a tag assigned). Note that this function - * assumes that any type of request can be queued! if this is not - * true for your device, you must check the request type before - * calling this function. The request will also be removed from - * the request queue, so it's the drivers responsibility to readd - * it if it should need to be restarted for some reason. - * - * Notes: - * queue lock must be held. - **/ -int blk_queue_start_tag(struct request_queue *q, struct request *rq) -{ - struct blk_queue_tag *bqt = q->queue_tags; - int tag; - - if (unlikely((rq->cmd_flags & REQ_QUEUED))) { - printk(KERN_ERR - "%s: request %p for device [%s] already tagged %d", - __FUNCTION__, rq, - rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->tag); - BUG(); - } - - /* - * Protect against shared tag maps, as we may not have exclusive - * access to the tag map. - */ - do { - tag = find_first_zero_bit(bqt->tag_map, bqt->max_depth); - if (tag >= bqt->max_depth) - return 1; - - } while (test_and_set_bit_lock(tag, bqt->tag_map)); - /* - * We need lock ordering semantics given by test_and_set_bit_lock. - * See blk_queue_end_tag for details. - */ - - rq->cmd_flags |= REQ_QUEUED; - rq->tag = tag; - bqt->tag_index[tag] = rq; - blkdev_dequeue_request(rq); - list_add(&rq->queuelist, &q->tag_busy_list); - bqt->busy++; - return 0; -} - -EXPORT_SYMBOL(blk_queue_start_tag); - -/** - * blk_queue_invalidate_tags - invalidate all pending tags - * @q: the request queue for the device - * - * Description: - * Hardware conditions may dictate a need to stop all pending requests. - * In this case, we will safely clear the block side of the tag queue and - * readd all requests to the request queue in the right order. - * - * Notes: - * queue lock must be held. - **/ -void blk_queue_invalidate_tags(struct request_queue *q) -{ - struct list_head *tmp, *n; - - list_for_each_safe(tmp, n, &q->tag_busy_list) - blk_requeue_request(q, list_entry_rq(tmp)); -} - -EXPORT_SYMBOL(blk_queue_invalidate_tags); - -void blk_dump_rq_flags(struct request *rq, char *msg) -{ - int bit; - - printk("%s: dev %s: type=%x, flags=%x\n", msg, - rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->cmd_type, - rq->cmd_flags); - - printk("\nsector %llu, nr/cnr %lu/%u\n", (unsigned long long)rq->sector, - rq->nr_sectors, - rq->current_nr_sectors); - printk("bio %p, biotail %p, buffer %p, data %p, len %u\n", rq->bio, rq->biotail, rq->buffer, rq->data, rq->data_len); - - if (blk_pc_request(rq)) { - printk("cdb: "); - for (bit = 0; bit < sizeof(rq->cmd); bit++) - printk("%02x ", rq->cmd[bit]); - printk("\n"); - } -} - -EXPORT_SYMBOL(blk_dump_rq_flags); - -void blk_recount_segments(struct request_queue *q, struct bio *bio) -{ - struct request rq; - struct bio *nxt = bio->bi_next; - rq.q = q; - rq.bio = rq.biotail = bio; - bio->bi_next = NULL; - blk_recalc_rq_segments(&rq); - bio->bi_next = nxt; - bio->bi_phys_segments = rq.nr_phys_segments; - bio->bi_hw_segments = rq.nr_hw_segments; - bio->bi_flags |= (1 << BIO_SEG_VALID); -} -EXPORT_SYMBOL(blk_recount_segments); - -static void blk_recalc_rq_segments(struct request *rq) -{ - int nr_phys_segs; - int nr_hw_segs; - unsigned int phys_size; - unsigned int hw_size; - struct bio_vec *bv, *bvprv = NULL; - int seg_size; - int hw_seg_size; - int cluster; - struct req_iterator iter; - int high, highprv = 1; - struct request_queue *q = rq->q; - - if (!rq->bio) - return; - - cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER); - hw_seg_size = seg_size = 0; - phys_size = hw_size = nr_phys_segs = nr_hw_segs = 0; - rq_for_each_segment(bv, rq, iter) { - /* - * the trick here is making sure that a high page is never - * considered part of another segment, since that might - * change with the bounce page. - */ - high = page_to_pfn(bv->bv_page) > q->bounce_pfn; - if (high || highprv) - goto new_hw_segment; - if (cluster) { - if (seg_size + bv->bv_len > q->max_segment_size) - goto new_segment; - if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv)) - goto new_segment; - if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv)) - goto new_segment; - if (BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len)) - goto new_hw_segment; - - seg_size += bv->bv_len; - hw_seg_size += bv->bv_len; - bvprv = bv; - continue; - } -new_segment: - if (BIOVEC_VIRT_MERGEABLE(bvprv, bv) && - !BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len)) - hw_seg_size += bv->bv_len; - else { -new_hw_segment: - if (nr_hw_segs == 1 && - hw_seg_size > rq->bio->bi_hw_front_size) - rq->bio->bi_hw_front_size = hw_seg_size; - hw_seg_size = BIOVEC_VIRT_START_SIZE(bv) + bv->bv_len; - nr_hw_segs++; - } - - nr_phys_segs++; - bvprv = bv; - seg_size = bv->bv_len; - highprv = high; - } - - if (nr_hw_segs == 1 && - hw_seg_size > rq->bio->bi_hw_front_size) - rq->bio->bi_hw_front_size = hw_seg_size; - if (hw_seg_size > rq->biotail->bi_hw_back_size) - rq->biotail->bi_hw_back_size = hw_seg_size; - rq->nr_phys_segments = nr_phys_segs; - rq->nr_hw_segments = nr_hw_segs; -} - -static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio, - struct bio *nxt) -{ - if (!(q->queue_flags & (1 << QUEUE_FLAG_CLUSTER))) - return 0; - - if (!BIOVEC_PHYS_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt))) - return 0; - if (bio->bi_size + nxt->bi_size > q->max_segment_size) - return 0; - - /* - * bio and nxt are contigous in memory, check if the queue allows - * these two to be merged into one - */ - if (BIO_SEG_BOUNDARY(q, bio, nxt)) - return 1; - - return 0; -} - -static int blk_hw_contig_segment(struct request_queue *q, struct bio *bio, - struct bio *nxt) -{ - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) - blk_recount_segments(q, bio); - if (unlikely(!bio_flagged(nxt, BIO_SEG_VALID))) - blk_recount_segments(q, nxt); - if (!BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)) || - BIOVEC_VIRT_OVERSIZE(bio->bi_hw_back_size + nxt->bi_hw_front_size)) - return 0; - if (bio->bi_hw_back_size + nxt->bi_hw_front_size > q->max_segment_size) - return 0; - - return 1; -} - -/* - * map a request to scatterlist, return number of sg entries setup. Caller - * must make sure sg can hold rq->nr_phys_segments entries - */ -int blk_rq_map_sg(struct request_queue *q, struct request *rq, - struct scatterlist *sglist) -{ - struct bio_vec *bvec, *bvprv; - struct req_iterator iter; - struct scatterlist *sg; - int nsegs, cluster; - - nsegs = 0; - cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER); - - /* - * for each bio in rq - */ - bvprv = NULL; - sg = NULL; - rq_for_each_segment(bvec, rq, iter) { - int nbytes = bvec->bv_len; - - if (bvprv && cluster) { - if (sg->length + nbytes > q->max_segment_size) - goto new_segment; - - if (!BIOVEC_PHYS_MERGEABLE(bvprv, bvec)) - goto new_segment; - if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bvec)) - goto new_segment; - - sg->length += nbytes; - } else { -new_segment: - if (!sg) - sg = sglist; - else { - /* - * If the driver previously mapped a shorter - * list, we could see a termination bit - * prematurely unless it fully inits the sg - * table on each mapping. We KNOW that there - * must be more entries here or the driver - * would be buggy, so force clear the - * termination bit to avoid doing a full - * sg_init_table() in drivers for each command. - */ - sg->page_link &= ~0x02; - sg = sg_next(sg); - } - - sg_set_page(sg, bvec->bv_page, nbytes, bvec->bv_offset); - nsegs++; - } - bvprv = bvec; - } /* segments in rq */ - - if (sg) - sg_mark_end(sg); - - return nsegs; -} - -EXPORT_SYMBOL(blk_rq_map_sg); - -/* - * the standard queue merge functions, can be overridden with device - * specific ones if so desired - */ - -static inline int ll_new_mergeable(struct request_queue *q, - struct request *req, - struct bio *bio) -{ - int nr_phys_segs = bio_phys_segments(q, bio); - - if (req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) { - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; - return 0; - } - - /* - * A hw segment is just getting larger, bump just the phys - * counter. - */ - req->nr_phys_segments += nr_phys_segs; - return 1; -} - -static inline int ll_new_hw_segment(struct request_queue *q, - struct request *req, - struct bio *bio) -{ - int nr_hw_segs = bio_hw_segments(q, bio); - int nr_phys_segs = bio_phys_segments(q, bio); - - if (req->nr_hw_segments + nr_hw_segs > q->max_hw_segments - || req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) { - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; - return 0; - } - - /* - * This will form the start of a new hw segment. Bump both - * counters. - */ - req->nr_hw_segments += nr_hw_segs; - req->nr_phys_segments += nr_phys_segs; - return 1; -} - -static int ll_back_merge_fn(struct request_queue *q, struct request *req, - struct bio *bio) -{ - unsigned short max_sectors; - int len; - - if (unlikely(blk_pc_request(req))) - max_sectors = q->max_hw_sectors; - else - max_sectors = q->max_sectors; - - if (req->nr_sectors + bio_sectors(bio) > max_sectors) { - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; - return 0; - } - if (unlikely(!bio_flagged(req->biotail, BIO_SEG_VALID))) - blk_recount_segments(q, req->biotail); - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) - blk_recount_segments(q, bio); - len = req->biotail->bi_hw_back_size + bio->bi_hw_front_size; - if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(req->biotail), __BVEC_START(bio)) && - !BIOVEC_VIRT_OVERSIZE(len)) { - int mergeable = ll_new_mergeable(q, req, bio); - - if (mergeable) { - if (req->nr_hw_segments == 1) - req->bio->bi_hw_front_size = len; - if (bio->bi_hw_segments == 1) - bio->bi_hw_back_size = len; - } - return mergeable; - } - - return ll_new_hw_segment(q, req, bio); -} - -static int ll_front_merge_fn(struct request_queue *q, struct request *req, - struct bio *bio) -{ - unsigned short max_sectors; - int len; - - if (unlikely(blk_pc_request(req))) - max_sectors = q->max_hw_sectors; - else - max_sectors = q->max_sectors; - - - if (req->nr_sectors + bio_sectors(bio) > max_sectors) { - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; - return 0; - } - len = bio->bi_hw_back_size + req->bio->bi_hw_front_size; - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) - blk_recount_segments(q, bio); - if (unlikely(!bio_flagged(req->bio, BIO_SEG_VALID))) - blk_recount_segments(q, req->bio); - if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(req->bio)) && - !BIOVEC_VIRT_OVERSIZE(len)) { - int mergeable = ll_new_mergeable(q, req, bio); - - if (mergeable) { - if (bio->bi_hw_segments == 1) - bio->bi_hw_front_size = len; - if (req->nr_hw_segments == 1) - req->biotail->bi_hw_back_size = len; - } - return mergeable; - } - - return ll_new_hw_segment(q, req, bio); -} - -static int ll_merge_requests_fn(struct request_queue *q, struct request *req, - struct request *next) -{ - int total_phys_segments; - int total_hw_segments; - - /* - * First check if the either of the requests are re-queued - * requests. Can't merge them if they are. - */ - if (req->special || next->special) - return 0; - - /* - * Will it become too large? - */ - if ((req->nr_sectors + next->nr_sectors) > q->max_sectors) - return 0; - - total_phys_segments = req->nr_phys_segments + next->nr_phys_segments; - if (blk_phys_contig_segment(q, req->biotail, next->bio)) - total_phys_segments--; - - if (total_phys_segments > q->max_phys_segments) - return 0; - - total_hw_segments = req->nr_hw_segments + next->nr_hw_segments; - if (blk_hw_contig_segment(q, req->biotail, next->bio)) { - int len = req->biotail->bi_hw_back_size + next->bio->bi_hw_front_size; - /* - * propagate the combined length to the end of the requests - */ - if (req->nr_hw_segments == 1) - req->bio->bi_hw_front_size = len; - if (next->nr_hw_segments == 1) - next->biotail->bi_hw_back_size = len; - total_hw_segments--; - } - - if (total_hw_segments > q->max_hw_segments) - return 0; - - /* Merge is OK... */ - req->nr_phys_segments = total_phys_segments; - req->nr_hw_segments = total_hw_segments; - return 1; -} - -/* - * "plug" the device if there are no outstanding requests: this will - * force the transfer to start only after we have put all the requests - * on the list. - * - * This is called with interrupts off and no requests on the queue and - * with the queue lock held. - */ -void blk_plug_device(struct request_queue *q) -{ - WARN_ON(!irqs_disabled()); - - /* - * don't plug a stopped queue, it must be paired with blk_start_queue() - * which will restart the queueing - */ - if (blk_queue_stopped(q)) - return; - - if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags)) { - mod_timer(&q->unplug_timer, jiffies + q->unplug_delay); - blk_add_trace_generic(q, NULL, 0, BLK_TA_PLUG); - } -} - -EXPORT_SYMBOL(blk_plug_device); - -/* - * remove the queue from the plugged list, if present. called with - * queue lock held and interrupts disabled. - */ -int blk_remove_plug(struct request_queue *q) -{ - WARN_ON(!irqs_disabled()); - - if (!test_and_clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags)) - return 0; - - del_timer(&q->unplug_timer); - return 1; -} - -EXPORT_SYMBOL(blk_remove_plug); - -/* - * remove the plug and let it rip.. - */ -void __generic_unplug_device(struct request_queue *q) -{ - if (unlikely(blk_queue_stopped(q))) - return; - - if (!blk_remove_plug(q)) - return; - - q->request_fn(q); -} -EXPORT_SYMBOL(__generic_unplug_device); - -/** - * generic_unplug_device - fire a request queue - * @q: The &struct request_queue in question - * - * Description: - * Linux uses plugging to build bigger requests queues before letting - * the device have at them. If a queue is plugged, the I/O scheduler - * is still adding and merging requests on the queue. Once the queue - * gets unplugged, the request_fn defined for the queue is invoked and - * transfers started. - **/ -void generic_unplug_device(struct request_queue *q) -{ - spin_lock_irq(q->queue_lock); - __generic_unplug_device(q); - spin_unlock_irq(q->queue_lock); -} -EXPORT_SYMBOL(generic_unplug_device); - -static void blk_backing_dev_unplug(struct backing_dev_info *bdi, - struct page *page) -{ - struct request_queue *q = bdi->unplug_io_data; - - blk_unplug(q); -} - -static void blk_unplug_work(struct work_struct *work) -{ - struct request_queue *q = - container_of(work, struct request_queue, unplug_work); - - blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL, - q->rq.count[READ] + q->rq.count[WRITE]); - - q->unplug_fn(q); -} - -static void blk_unplug_timeout(unsigned long data) -{ - struct request_queue *q = (struct request_queue *)data; - - blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_TIMER, NULL, - q->rq.count[READ] + q->rq.count[WRITE]); - - kblockd_schedule_work(&q->unplug_work); -} - -void blk_unplug(struct request_queue *q) -{ - /* - * devices don't necessarily have an ->unplug_fn defined - */ - if (q->unplug_fn) { - blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL, - q->rq.count[READ] + q->rq.count[WRITE]); - - q->unplug_fn(q); - } -} -EXPORT_SYMBOL(blk_unplug); - -/** - * blk_start_queue - restart a previously stopped queue - * @q: The &struct request_queue in question - * - * Description: - * blk_start_queue() will clear the stop flag on the queue, and call - * the request_fn for the queue if it was in a stopped state when - * entered. Also see blk_stop_queue(). Queue lock must be held. - **/ -void blk_start_queue(struct request_queue *q) -{ - WARN_ON(!irqs_disabled()); - - clear_bit(QUEUE_FLAG_STOPPED, &q->queue_flags); - - /* - * one level of recursion is ok and is much faster than kicking - * the unplug handling - */ - if (!test_and_set_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) { - q->request_fn(q); - clear_bit(QUEUE_FLAG_REENTER, &q->queue_flags); - } else { - blk_plug_device(q); - kblockd_schedule_work(&q->unplug_work); - } -} - -EXPORT_SYMBOL(blk_start_queue); - -/** - * blk_stop_queue - stop a queue - * @q: The &struct request_queue in question - * - * Description: - * The Linux block layer assumes that a block driver will consume all - * entries on the request queue when the request_fn strategy is called. - * Often this will not happen, because of hardware limitations (queue - * depth settings). If a device driver gets a 'queue full' response, - * or if it simply chooses not to queue more I/O at one point, it can - * call this function to prevent the request_fn from being called until - * the driver has signalled it's ready to go again. This happens by calling - * blk_start_queue() to restart queue operations. Queue lock must be held. - **/ -void blk_stop_queue(struct request_queue *q) -{ - blk_remove_plug(q); - set_bit(QUEUE_FLAG_STOPPED, &q->queue_flags); -} -EXPORT_SYMBOL(blk_stop_queue); - -/** - * blk_sync_queue - cancel any pending callbacks on a queue - * @q: the queue - * - * Description: - * The block layer may perform asynchronous callback activity - * on a queue, such as calling the unplug function after a timeout. - * A block device may call blk_sync_queue to ensure that any - * such activity is cancelled, thus allowing it to release resources - * that the callbacks might use. The caller must already have made sure - * that its ->make_request_fn will not re-add plugging prior to calling - * this function. - * - */ -void blk_sync_queue(struct request_queue *q) -{ - del_timer_sync(&q->unplug_timer); - kblockd_flush_work(&q->unplug_work); -} -EXPORT_SYMBOL(blk_sync_queue); - -/** - * blk_run_queue - run a single device queue - * @q: The queue to run - */ -void blk_run_queue(struct request_queue *q) -{ - unsigned long flags; - - spin_lock_irqsave(q->queue_lock, flags); - blk_remove_plug(q); - - /* - * Only recurse once to avoid overrunning the stack, let the unplug - * handling reinvoke the handler shortly if we already got there. - */ - if (!elv_queue_empty(q)) { - if (!test_and_set_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) { - q->request_fn(q); - clear_bit(QUEUE_FLAG_REENTER, &q->queue_flags); - } else { - blk_plug_device(q); - kblockd_schedule_work(&q->unplug_work); - } - } - - spin_unlock_irqrestore(q->queue_lock, flags); -} -EXPORT_SYMBOL(blk_run_queue); - -/** - * blk_cleanup_queue: - release a &struct request_queue when it is no longer needed - * @kobj: the kobj belonging of the request queue to be released - * - * Description: - * blk_cleanup_queue is the pair to blk_init_queue() or - * blk_queue_make_request(). It should be called when a request queue is - * being released; typically when a block device is being de-registered. - * Currently, its primary task it to free all the &struct request - * structures that were allocated to the queue and the queue itself. - * - * Caveat: - * Hopefully the low level driver will have finished any - * outstanding requests first... - **/ -static void blk_release_queue(struct kobject *kobj) -{ - struct request_queue *q = - container_of(kobj, struct request_queue, kobj); - struct request_list *rl = &q->rq; - - blk_sync_queue(q); - - if (rl->rq_pool) - mempool_destroy(rl->rq_pool); - - if (q->queue_tags) - __blk_queue_free_tags(q); - - blk_trace_shutdown(q); - - bdi_destroy(&q->backing_dev_info); - kmem_cache_free(requestq_cachep, q); -} - -void blk_put_queue(struct request_queue *q) -{ - kobject_put(&q->kobj); -} -EXPORT_SYMBOL(blk_put_queue); - -void blk_cleanup_queue(struct request_queue * q) -{ - mutex_lock(&q->sysfs_lock); - set_bit(QUEUE_FLAG_DEAD, &q->queue_flags); - mutex_unlock(&q->sysfs_lock); - - if (q->elevator) - elevator_exit(q->elevator); - - blk_put_queue(q); -} - -EXPORT_SYMBOL(blk_cleanup_queue); - -static int blk_init_free_list(struct request_queue *q) -{ - struct request_list *rl = &q->rq; - - rl->count[READ] = rl->count[WRITE] = 0; - rl->starved[READ] = rl->starved[WRITE] = 0; - rl->elvpriv = 0; - init_waitqueue_head(&rl->wait[READ]); - init_waitqueue_head(&rl->wait[WRITE]); - - rl->rq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab, - mempool_free_slab, request_cachep, q->node); - - if (!rl->rq_pool) - return -ENOMEM; - - return 0; -} - -struct request_queue *blk_alloc_queue(gfp_t gfp_mask) -{ - return blk_alloc_queue_node(gfp_mask, -1); -} -EXPORT_SYMBOL(blk_alloc_queue); - -static struct kobj_type queue_ktype; - -struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) -{ - struct request_queue *q; - int err; - - q = kmem_cache_alloc_node(requestq_cachep, - gfp_mask | __GFP_ZERO, node_id); - if (!q) - return NULL; - - q->backing_dev_info.unplug_io_fn = blk_backing_dev_unplug; - q->backing_dev_info.unplug_io_data = q; - err = bdi_init(&q->backing_dev_info); - if (err) { - kmem_cache_free(requestq_cachep, q); - return NULL; - } - - init_timer(&q->unplug_timer); - - kobject_set_name(&q->kobj, "%s", "queue"); - q->kobj.ktype = &queue_ktype; - kobject_init(&q->kobj); - - mutex_init(&q->sysfs_lock); - - return q; -} -EXPORT_SYMBOL(blk_alloc_queue_node); - -/** - * blk_init_queue - prepare a request queue for use with a block device - * @rfn: The function to be called to process requests that have been - * placed on the queue. - * @lock: Request queue spin lock - * - * Description: - * If a block device wishes to use the standard request handling procedures, - * which sorts requests and coalesces adjacent requests, then it must - * call blk_init_queue(). The function @rfn will be called when there - * are requests on the queue that need to be processed. If the device - * supports plugging, then @rfn may not be called immediately when requests - * are available on the queue, but may be called at some time later instead. - * Plugged queues are generally unplugged when a buffer belonging to one - * of the requests on the queue is needed, or due to memory pressure. - * - * @rfn is not required, or even expected, to remove all requests off the - * queue, but only as many as it can handle at a time. If it does leave - * requests on the queue, it is responsible for arranging that the requests - * get dealt with eventually. - * - * The queue spin lock must be held while manipulating the requests on the - * request queue; this lock will be taken also from interrupt context, so irq - * disabling is needed for it. - * - * Function returns a pointer to the initialized request queue, or NULL if - * it didn't succeed. - * - * Note: - * blk_init_queue() must be paired with a blk_cleanup_queue() call - * when the block device is deactivated (such as at module unload). - **/ - -struct request_queue *blk_init_queue(request_fn_proc *rfn, spinlock_t *lock) -{ - return blk_init_queue_node(rfn, lock, -1); -} -EXPORT_SYMBOL(blk_init_queue); - -struct request_queue * -blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id) -{ - struct request_queue *q = blk_alloc_queue_node(GFP_KERNEL, node_id); - - if (!q) - return NULL; - - q->node = node_id; - if (blk_init_free_list(q)) { - kmem_cache_free(requestq_cachep, q); - return NULL; - } - - /* - * if caller didn't supply a lock, they get per-queue locking with - * our embedded lock - */ - if (!lock) { - spin_lock_init(&q->__queue_lock); - lock = &q->__queue_lock; - } - - q->request_fn = rfn; - q->prep_rq_fn = NULL; - q->unplug_fn = generic_unplug_device; - q->queue_flags = (1 << QUEUE_FLAG_CLUSTER); - q->queue_lock = lock; - - blk_queue_segment_boundary(q, 0xffffffff); - - blk_queue_make_request(q, __make_request); - blk_queue_max_segment_size(q, MAX_SEGMENT_SIZE); - - blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS); - blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS); - - q->sg_reserved_size = INT_MAX; - - /* - * all done - */ - if (!elevator_init(q, NULL)) { - blk_queue_congestion_threshold(q); - return q; - } - - blk_put_queue(q); - return NULL; -} -EXPORT_SYMBOL(blk_init_queue_node); - -int blk_get_queue(struct request_queue *q) -{ - if (likely(!test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) { - kobject_get(&q->kobj); - return 0; - } - - return 1; -} - -EXPORT_SYMBOL(blk_get_queue); - -static inline void blk_free_request(struct request_queue *q, struct request *rq) -{ - if (rq->cmd_flags & REQ_ELVPRIV) - elv_put_request(q, rq); - mempool_free(rq, q->rq.rq_pool); -} - -static struct request * -blk_alloc_request(struct request_queue *q, int rw, int priv, gfp_t gfp_mask) -{ - struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask); - - if (!rq) - return NULL; - - /* - * first three bits are identical in rq->cmd_flags and bio->bi_rw, - * see bio.h and blkdev.h - */ - rq->cmd_flags = rw | REQ_ALLOCED; - - if (priv) { - if (unlikely(elv_set_request(q, rq, gfp_mask))) { - mempool_free(rq, q->rq.rq_pool); - return NULL; - } - rq->cmd_flags |= REQ_ELVPRIV; - } - - return rq; -} - -/* - * ioc_batching returns true if the ioc is a valid batching request and - * should be given priority access to a request. - */ -static inline int ioc_batching(struct request_queue *q, struct io_context *ioc) -{ - if (!ioc) - return 0; - - /* - * Make sure the process is able to allocate at least 1 request - * even if the batch times out, otherwise we could theoretically - * lose wakeups. - */ - return ioc->nr_batch_requests == q->nr_batching || - (ioc->nr_batch_requests > 0 - && time_before(jiffies, ioc->last_waited + BLK_BATCH_TIME)); -} - -/* - * ioc_set_batching sets ioc to be a new "batcher" if it is not one. This - * will cause the process to be a "batcher" on all queues in the system. This - * is the behaviour we want though - once it gets a wakeup it should be given - * a nice run. - */ -static void ioc_set_batching(struct request_queue *q, struct io_context *ioc) -{ - if (!ioc || ioc_batching(q, ioc)) - return; - - ioc->nr_batch_requests = q->nr_batching; - ioc->last_waited = jiffies; -} - -static void __freed_request(struct request_queue *q, int rw) -{ - struct request_list *rl = &q->rq; - - if (rl->count[rw] < queue_congestion_off_threshold(q)) - blk_clear_queue_congested(q, rw); - - if (rl->count[rw] + 1 <= q->nr_requests) { - if (waitqueue_active(&rl->wait[rw])) - wake_up(&rl->wait[rw]); - - blk_clear_queue_full(q, rw); - } -} - -/* - * A request has just been released. Account for it, update the full and - * congestion status, wake up any waiters. Called under q->queue_lock. - */ -static void freed_request(struct request_queue *q, int rw, int priv) -{ - struct request_list *rl = &q->rq; - - rl->count[rw]--; - if (priv) - rl->elvpriv--; - - __freed_request(q, rw); - - if (unlikely(rl->starved[rw ^ 1])) - __freed_request(q, rw ^ 1); -} - -#define blkdev_free_rq(list) list_entry((list)->next, struct request, queuelist) -/* - * Get a free request, queue_lock must be held. - * Returns NULL on failure, with queue_lock held. - * Returns !NULL on success, with queue_lock *not held*. - */ -static struct request *get_request(struct request_queue *q, int rw_flags, - struct bio *bio, gfp_t gfp_mask) -{ - struct request *rq = NULL; - struct request_list *rl = &q->rq; - struct io_context *ioc = NULL; - const int rw = rw_flags & 0x01; - int may_queue, priv; - - may_queue = elv_may_queue(q, rw_flags); - if (may_queue == ELV_MQUEUE_NO) - goto rq_starved; - - if (rl->count[rw]+1 >= queue_congestion_on_threshold(q)) { - if (rl->count[rw]+1 >= q->nr_requests) { - ioc = current_io_context(GFP_ATOMIC, q->node); - /* - * The queue will fill after this allocation, so set - * it as full, and mark this process as "batching". - * This process will be allowed to complete a batch of - * requests, others will be blocked. - */ - if (!blk_queue_full(q, rw)) { - ioc_set_batching(q, ioc); - blk_set_queue_full(q, rw); - } else { - if (may_queue != ELV_MQUEUE_MUST - && !ioc_batching(q, ioc)) { - /* - * The queue is full and the allocating - * process is not a "batcher", and not - * exempted by the IO scheduler - */ - goto out; - } - } - } - blk_set_queue_congested(q, rw); - } - - /* - * Only allow batching queuers to allocate up to 50% over the defined - * limit of requests, otherwise we could have thousands of requests - * allocated with any setting of ->nr_requests - */ - if (rl->count[rw] >= (3 * q->nr_requests / 2)) - goto out; - - rl->count[rw]++; - rl->starved[rw] = 0; - - priv = !test_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags); - if (priv) - rl->elvpriv++; - - spin_unlock_irq(q->queue_lock); - - rq = blk_alloc_request(q, rw_flags, priv, gfp_mask); - if (unlikely(!rq)) { - /* - * Allocation failed presumably due to memory. Undo anything - * we might have messed up. - * - * Allocating task should really be put onto the front of the - * wait queue, but this is pretty rare. - */ - spin_lock_irq(q->queue_lock); - freed_request(q, rw, priv); - - /* - * in the very unlikely event that allocation failed and no - * requests for this direction was pending, mark us starved - * so that freeing of a request in the other direction will - * notice us. another possible fix would be to split the - * rq mempool into READ and WRITE - */ -rq_starved: - if (unlikely(rl->count[rw] == 0)) - rl->starved[rw] = 1; - - goto out; - } - - /* - * ioc may be NULL here, and ioc_batching will be false. That's - * OK, if the queue is under the request limit then requests need - * not count toward the nr_batch_requests limit. There will always - * be some limit enforced by BLK_BATCH_TIME. - */ - if (ioc_batching(q, ioc)) - ioc->nr_batch_requests--; - - rq_init(q, rq); - - blk_add_trace_generic(q, bio, rw, BLK_TA_GETRQ); -out: - return rq; -} - -/* - * No available requests for this queue, unplug the device and wait for some - * requests to become available. - * - * Called with q->queue_lock held, and returns with it unlocked. - */ -static struct request *get_request_wait(struct request_queue *q, int rw_flags, - struct bio *bio) -{ - const int rw = rw_flags & 0x01; - struct request *rq; - - rq = get_request(q, rw_flags, bio, GFP_NOIO); - while (!rq) { - DEFINE_WAIT(wait); - struct request_list *rl = &q->rq; - - prepare_to_wait_exclusive(&rl->wait[rw], &wait, - TASK_UNINTERRUPTIBLE); - - rq = get_request(q, rw_flags, bio, GFP_NOIO); - - if (!rq) { - struct io_context *ioc; - - blk_add_trace_generic(q, bio, rw, BLK_TA_SLEEPRQ); - - __generic_unplug_device(q); - spin_unlock_irq(q->queue_lock); - io_schedule(); - - /* - * After sleeping, we become a "batching" process and - * will be able to allocate at least one request, and - * up to a big batch of them for a small period time. - * See ioc_batching, ioc_set_batching - */ - ioc = current_io_context(GFP_NOIO, q->node); - ioc_set_batching(q, ioc); - - spin_lock_irq(q->queue_lock); - } - finish_wait(&rl->wait[rw], &wait); - } - - return rq; -} - -struct request *blk_get_request(struct request_queue *q, int rw, gfp_t gfp_mask) -{ - struct request *rq; - - BUG_ON(rw != READ && rw != WRITE); - - spin_lock_irq(q->queue_lock); - if (gfp_mask & __GFP_WAIT) { - rq = get_request_wait(q, rw, NULL); - } else { - rq = get_request(q, rw, NULL, gfp_mask); - if (!rq) - spin_unlock_irq(q->queue_lock); - } - /* q->queue_lock is unlocked at this point */ - - return rq; -} -EXPORT_SYMBOL(blk_get_request); - -/** - * blk_start_queueing - initiate dispatch of requests to device - * @q: request queue to kick into gear - * - * This is basically a helper to remove the need to know whether a queue - * is plugged or not if someone just wants to initiate dispatch of requests - * for this queue. - * - * The queue lock must be held with interrupts disabled. - */ -void blk_start_queueing(struct request_queue *q) -{ - if (!blk_queue_plugged(q)) - q->request_fn(q); - else - __generic_unplug_device(q); -} -EXPORT_SYMBOL(blk_start_queueing); - -/** - * blk_requeue_request - put a request back on queue - * @q: request queue where request should be inserted - * @rq: request to be inserted - * - * Description: - * Drivers often keep queueing requests until the hardware cannot accept - * more, when that condition happens we need to put the request back - * on the queue. Must be called with queue lock held. - */ -void blk_requeue_request(struct request_queue *q, struct request *rq) -{ - blk_add_trace_rq(q, rq, BLK_TA_REQUEUE); - - if (blk_rq_tagged(rq)) - blk_queue_end_tag(q, rq); - - elv_requeue_request(q, rq); -} - -EXPORT_SYMBOL(blk_requeue_request); - -/** - * blk_insert_request - insert a special request in to a request queue - * @q: request queue where request should be inserted - * @rq: request to be inserted - * @at_head: insert request at head or tail of queue - * @data: private data - * - * Description: - * Many block devices need to execute commands asynchronously, so they don't - * block the whole kernel from preemption during request execution. This is - * accomplished normally by inserting aritficial requests tagged as - * REQ_SPECIAL in to the corresponding request queue, and letting them be - * scheduled for actual execution by the request queue. - * - * We have the option of inserting the head or the tail of the queue. - * Typically we use the tail for new ioctls and so forth. We use the head - * of the queue for things like a QUEUE_FULL message from a device, or a - * host that is unable to accept a particular command. - */ -void blk_insert_request(struct request_queue *q, struct request *rq, - int at_head, void *data) -{ - int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK; - unsigned long flags; - - /* - * tell I/O scheduler that this isn't a regular read/write (ie it - * must not attempt merges on this) and that it acts as a soft - * barrier - */ - rq->cmd_type = REQ_TYPE_SPECIAL; - rq->cmd_flags |= REQ_SOFTBARRIER; - - rq->special = data; - - spin_lock_irqsave(q->queue_lock, flags); - - /* - * If command is tagged, release the tag - */ - if (blk_rq_tagged(rq)) - blk_queue_end_tag(q, rq); - - drive_stat_acct(rq, 1); - __elv_add_request(q, rq, where, 0); - blk_start_queueing(q); - spin_unlock_irqrestore(q->queue_lock, flags); -} - -EXPORT_SYMBOL(blk_insert_request); - -static int __blk_rq_unmap_user(struct bio *bio) -{ - int ret = 0; - - if (bio) { - if (bio_flagged(bio, BIO_USER_MAPPED)) - bio_unmap_user(bio); - else - ret = bio_uncopy_user(bio); - } - - return ret; -} - -int blk_rq_append_bio(struct request_queue *q, struct request *rq, - struct bio *bio) -{ - if (!rq->bio) - blk_rq_bio_prep(q, rq, bio); - else if (!ll_back_merge_fn(q, rq, bio)) - return -EINVAL; - else { - rq->biotail->bi_next = bio; - rq->biotail = bio; - - rq->data_len += bio->bi_size; - } - return 0; -} -EXPORT_SYMBOL(blk_rq_append_bio); - -static int __blk_rq_map_user(struct request_queue *q, struct request *rq, - void __user *ubuf, unsigned int len) -{ - unsigned long uaddr; - struct bio *bio, *orig_bio; - int reading, ret; - - reading = rq_data_dir(rq) == READ; - - /* - * if alignment requirement is satisfied, map in user pages for - * direct dma. else, set up kernel bounce buffers - */ - uaddr = (unsigned long) ubuf; - if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q))) - bio = bio_map_user(q, NULL, uaddr, len, reading); - else - bio = bio_copy_user(q, uaddr, len, reading); - - if (IS_ERR(bio)) - return PTR_ERR(bio); - - orig_bio = bio; - blk_queue_bounce(q, &bio); - - /* - * We link the bounce buffer in and could have to traverse it - * later so we have to get a ref to prevent it from being freed - */ - bio_get(bio); - - ret = blk_rq_append_bio(q, rq, bio); - if (!ret) - return bio->bi_size; - - /* if it was boucned we must call the end io function */ - bio_endio(bio, 0); - __blk_rq_unmap_user(orig_bio); - bio_put(bio); - return ret; -} - -/** - * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage - * @q: request queue where request should be inserted - * @rq: request structure to fill - * @ubuf: the user buffer - * @len: length of user data - * - * Description: - * Data will be mapped directly for zero copy io, if possible. Otherwise - * a kernel bounce buffer is used. - * - * A matching blk_rq_unmap_user() must be issued at the end of io, while - * still in process context. - * - * Note: The mapped bio may need to be bounced through blk_queue_bounce() - * before being submitted to the device, as pages mapped may be out of - * reach. It's the callers responsibility to make sure this happens. The - * original bio must be passed back in to blk_rq_unmap_user() for proper - * unmapping. - */ -int blk_rq_map_user(struct request_queue *q, struct request *rq, - void __user *ubuf, unsigned long len) -{ - unsigned long bytes_read = 0; - struct bio *bio = NULL; - int ret; - - if (len > (q->max_hw_sectors << 9)) - return -EINVAL; - if (!len || !ubuf) - return -EINVAL; - - while (bytes_read != len) { - unsigned long map_len, end, start; - - map_len = min_t(unsigned long, len - bytes_read, BIO_MAX_SIZE); - end = ((unsigned long)ubuf + map_len + PAGE_SIZE - 1) - >> PAGE_SHIFT; - start = (unsigned long)ubuf >> PAGE_SHIFT; - - /* - * A bad offset could cause us to require BIO_MAX_PAGES + 1 - * pages. If this happens we just lower the requested - * mapping len by a page so that we can fit - */ - if (end - start > BIO_MAX_PAGES) - map_len -= PAGE_SIZE; - - ret = __blk_rq_map_user(q, rq, ubuf, map_len); - if (ret < 0) - goto unmap_rq; - if (!bio) - bio = rq->bio; - bytes_read += ret; - ubuf += ret; - } - - rq->buffer = rq->data = NULL; - return 0; -unmap_rq: - blk_rq_unmap_user(bio); - return ret; -} - -EXPORT_SYMBOL(blk_rq_map_user); - -/** - * blk_rq_map_user_iov - map user data to a request, for REQ_BLOCK_PC usage - * @q: request queue where request should be inserted - * @rq: request to map data to - * @iov: pointer to the iovec - * @iov_count: number of elements in the iovec - * @len: I/O byte count - * - * Description: - * Data will be mapped directly for zero copy io, if possible. Otherwise - * a kernel bounce buffer is used. - * - * A matching blk_rq_unmap_user() must be issued at the end of io, while - * still in process context. - * - * Note: The mapped bio may need to be bounced through blk_queue_bounce() - * before being submitted to the device, as pages mapped may be out of - * reach. It's the callers responsibility to make sure this happens. The - * original bio must be passed back in to blk_rq_unmap_user() for proper - * unmapping. - */ -int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, - struct sg_iovec *iov, int iov_count, unsigned int len) -{ - struct bio *bio; - - if (!iov || iov_count <= 0) - return -EINVAL; - - /* we don't allow misaligned data like bio_map_user() does. If the - * user is using sg, they're expected to know the alignment constraints - * and respect them accordingly */ - bio = bio_map_user_iov(q, NULL, iov, iov_count, rq_data_dir(rq)== READ); - if (IS_ERR(bio)) - return PTR_ERR(bio); - - if (bio->bi_size != len) { - bio_endio(bio, 0); - bio_unmap_user(bio); - return -EINVAL; - } - - bio_get(bio); - blk_rq_bio_prep(q, rq, bio); - rq->buffer = rq->data = NULL; - return 0; -} - -EXPORT_SYMBOL(blk_rq_map_user_iov); - -/** - * blk_rq_unmap_user - unmap a request with user data - * @bio: start of bio list - * - * Description: - * Unmap a rq previously mapped by blk_rq_map_user(). The caller must - * supply the original rq->bio from the blk_rq_map_user() return, since - * the io completion may have changed rq->bio. - */ -int blk_rq_unmap_user(struct bio *bio) -{ - struct bio *mapped_bio; - int ret = 0, ret2; - - while (bio) { - mapped_bio = bio; - if (unlikely(bio_flagged(bio, BIO_BOUNCED))) - mapped_bio = bio->bi_private; - - ret2 = __blk_rq_unmap_user(mapped_bio); - if (ret2 && !ret) - ret = ret2; - - mapped_bio = bio; - bio = bio->bi_next; - bio_put(mapped_bio); - } - - return ret; -} - -EXPORT_SYMBOL(blk_rq_unmap_user); - -/** - * blk_rq_map_kern - map kernel data to a request, for REQ_BLOCK_PC usage - * @q: request queue where request should be inserted - * @rq: request to fill - * @kbuf: the kernel buffer - * @len: length of user data - * @gfp_mask: memory allocation flags - */ -int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, - unsigned int len, gfp_t gfp_mask) -{ - struct bio *bio; - - if (len > (q->max_hw_sectors << 9)) - return -EINVAL; - if (!len || !kbuf) - return -EINVAL; - - bio = bio_map_kern(q, kbuf, len, gfp_mask); - if (IS_ERR(bio)) - return PTR_ERR(bio); - - if (rq_data_dir(rq) == WRITE) - bio->bi_rw |= (1 << BIO_RW); - - blk_rq_bio_prep(q, rq, bio); - blk_queue_bounce(q, &rq->bio); - rq->buffer = rq->data = NULL; - return 0; -} - -EXPORT_SYMBOL(blk_rq_map_kern); - -/** - * blk_execute_rq_nowait - insert a request into queue for execution - * @q: queue to insert the request in - * @bd_disk: matching gendisk - * @rq: request to insert - * @at_head: insert request at head or tail of queue - * @done: I/O completion handler - * - * Description: - * Insert a fully prepared request at the back of the io scheduler queue - * for execution. Don't wait for completion. - */ -void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk, - struct request *rq, int at_head, - rq_end_io_fn *done) -{ - int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK; - - rq->rq_disk = bd_disk; - rq->cmd_flags |= REQ_NOMERGE; - rq->end_io = done; - WARN_ON(irqs_disabled()); - spin_lock_irq(q->queue_lock); - __elv_add_request(q, rq, where, 1); - __generic_unplug_device(q); - spin_unlock_irq(q->queue_lock); -} -EXPORT_SYMBOL_GPL(blk_execute_rq_nowait); - -/** - * blk_execute_rq - insert a request into queue for execution - * @q: queue to insert the request in - * @bd_disk: matching gendisk - * @rq: request to insert - * @at_head: insert request at head or tail of queue - * - * Description: - * Insert a fully prepared request at the back of the io scheduler queue - * for execution and wait for completion. - */ -int blk_execute_rq(struct request_queue *q, struct gendisk *bd_disk, - struct request *rq, int at_head) -{ - DECLARE_COMPLETION_ONSTACK(wait); - char sense[SCSI_SENSE_BUFFERSIZE]; - int err = 0; - - /* - * we need an extra reference to the request, so we can look at - * it after io completion - */ - rq->ref_count++; - - if (!rq->sense) { - memset(sense, 0, sizeof(sense)); - rq->sense = sense; - rq->sense_len = 0; - } - - rq->end_io_data = &wait; - blk_execute_rq_nowait(q, bd_disk, rq, at_head, blk_end_sync_rq); - wait_for_completion(&wait); - - if (rq->errors) - err = -EIO; - - return err; -} - -EXPORT_SYMBOL(blk_execute_rq); - -static void bio_end_empty_barrier(struct bio *bio, int err) -{ - if (err) - clear_bit(BIO_UPTODATE, &bio->bi_flags); - - complete(bio->bi_private); -} - -/** - * blkdev_issue_flush - queue a flush - * @bdev: blockdev to issue flush for - * @error_sector: error sector - * - * Description: - * Issue a flush for the block device in question. Caller can supply - * room for storing the error offset in case of a flush error, if they - * wish to. Caller must run wait_for_completion() on its own. - */ -int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector) -{ - DECLARE_COMPLETION_ONSTACK(wait); - struct request_queue *q; - struct bio *bio; - int ret; - - if (bdev->bd_disk == NULL) - return -ENXIO; - - q = bdev_get_queue(bdev); - if (!q) - return -ENXIO; - - bio = bio_alloc(GFP_KERNEL, 0); - if (!bio) - return -ENOMEM; - - bio->bi_end_io = bio_end_empty_barrier; - bio->bi_private = &wait; - bio->bi_bdev = bdev; - submit_bio(1 << BIO_RW_BARRIER, bio); - - wait_for_completion(&wait); - - /* - * The driver must store the error location in ->bi_sector, if - * it supports it. For non-stacked drivers, this should be copied - * from rq->sector. - */ - if (error_sector) - *error_sector = bio->bi_sector; - - ret = 0; - if (!bio_flagged(bio, BIO_UPTODATE)) - ret = -EIO; - - bio_put(bio); - return ret; -} - -EXPORT_SYMBOL(blkdev_issue_flush); - -static void drive_stat_acct(struct request *rq, int new_io) -{ - int rw = rq_data_dir(rq); - - if (!blk_fs_request(rq) || !rq->rq_disk) - return; - - if (!new_io) { - __disk_stat_inc(rq->rq_disk, merges[rw]); - } else { - disk_round_stats(rq->rq_disk); - rq->rq_disk->in_flight++; - } -} - -/* - * add-request adds a request to the linked list. - * queue lock is held and interrupts disabled, as we muck with the - * request queue list. - */ -static inline void add_request(struct request_queue * q, struct request * req) -{ - drive_stat_acct(req, 1); - - /* - * elevator indicated where it wants this request to be - * inserted at elevator_merge time - */ - __elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0); -} - -/* - * disk_round_stats() - Round off the performance stats on a struct - * disk_stats. - * - * The average IO queue length and utilisation statistics are maintained - * by observing the current state of the queue length and the amount of - * time it has been in this state for. - * - * Normally, that accounting is done on IO completion, but that can result - * in more than a second's worth of IO being accounted for within any one - * second, leading to >100% utilisation. To deal with that, we call this - * function to do a round-off before returning the results when reading - * /proc/diskstats. This accounts immediately for all queue usage up to - * the current jiffies and restarts the counters again. - */ -void disk_round_stats(struct gendisk *disk) -{ - unsigned long now = jiffies; - - if (now == disk->stamp) - return; - - if (disk->in_flight) { - __disk_stat_add(disk, time_in_queue, - disk->in_flight * (now - disk->stamp)); - __disk_stat_add(disk, io_ticks, (now - disk->stamp)); - } - disk->stamp = now; -} - -EXPORT_SYMBOL_GPL(disk_round_stats); - -/* - * queue lock must be held - */ -void __blk_put_request(struct request_queue *q, struct request *req) -{ - if (unlikely(!q)) - return; - if (unlikely(--req->ref_count)) - return; - - elv_completed_request(q, req); - - /* - * Request may not have originated from ll_rw_blk. if not, - * it didn't come out of our reserved rq pools - */ - if (req->cmd_flags & REQ_ALLOCED) { - int rw = rq_data_dir(req); - int priv = req->cmd_flags & REQ_ELVPRIV; - - BUG_ON(!list_empty(&req->queuelist)); - BUG_ON(!hlist_unhashed(&req->hash)); - - blk_free_request(q, req); - freed_request(q, rw, priv); - } -} - -EXPORT_SYMBOL_GPL(__blk_put_request); - -void blk_put_request(struct request *req) -{ - unsigned long flags; - struct request_queue *q = req->q; - - /* - * Gee, IDE calls in w/ NULL q. Fix IDE and remove the - * following if (q) test. - */ - if (q) { - spin_lock_irqsave(q->queue_lock, flags); - __blk_put_request(q, req); - spin_unlock_irqrestore(q->queue_lock, flags); - } -} - -EXPORT_SYMBOL(blk_put_request); - -/** - * blk_end_sync_rq - executes a completion event on a request - * @rq: request to complete - * @error: end io status of the request - */ -void blk_end_sync_rq(struct request *rq, int error) -{ - struct completion *waiting = rq->end_io_data; - - rq->end_io_data = NULL; - __blk_put_request(rq->q, rq); - - /* - * complete last, if this is a stack request the process (and thus - * the rq pointer) could be invalid right after this complete() - */ - complete(waiting); -} -EXPORT_SYMBOL(blk_end_sync_rq); - -/* - * Has to be called with the request spinlock acquired - */ -static int attempt_merge(struct request_queue *q, struct request *req, - struct request *next) -{ - if (!rq_mergeable(req) || !rq_mergeable(next)) - return 0; - - /* - * not contiguous - */ - if (req->sector + req->nr_sectors != next->sector) - return 0; - - if (rq_data_dir(req) != rq_data_dir(next) - || req->rq_disk != next->rq_disk - || next->special) - return 0; - - /* - * If we are allowed to merge, then append bio list - * from next to rq and release next. merge_requests_fn - * will have updated segment counts, update sector - * counts here. - */ - if (!ll_merge_requests_fn(q, req, next)) - return 0; - - /* - * At this point we have either done a back merge - * or front merge. We need the smaller start_time of - * the merged requests to be the current request - * for accounting purposes. - */ - if (time_after(req->start_time, next->start_time)) - req->start_time = next->start_time; - - req->biotail->bi_next = next->bio; - req->biotail = next->biotail; - - req->nr_sectors = req->hard_nr_sectors += next->hard_nr_sectors; - - elv_merge_requests(q, req, next); - - if (req->rq_disk) { - disk_round_stats(req->rq_disk); - req->rq_disk->in_flight--; - } - - req->ioprio = ioprio_best(req->ioprio, next->ioprio); - - __blk_put_request(q, next); - return 1; -} - -static inline int attempt_back_merge(struct request_queue *q, - struct request *rq) -{ - struct request *next = elv_latter_request(q, rq); - - if (next) - return attempt_merge(q, rq, next); - - return 0; -} - -static inline int attempt_front_merge(struct request_queue *q, - struct request *rq) -{ - struct request *prev = elv_former_request(q, rq); - - if (prev) - return attempt_merge(q, prev, rq); - - return 0; -} - -static void init_request_from_bio(struct request *req, struct bio *bio) -{ - req->cmd_type = REQ_TYPE_FS; - - /* - * inherit FAILFAST from bio (for read-ahead, and explicit FAILFAST) - */ - if (bio_rw_ahead(bio) || bio_failfast(bio)) - req->cmd_flags |= REQ_FAILFAST; - - /* - * REQ_BARRIER implies no merging, but lets make it explicit - */ - if (unlikely(bio_barrier(bio))) - req->cmd_flags |= (REQ_HARDBARRIER | REQ_NOMERGE); - - if (bio_sync(bio)) - req->cmd_flags |= REQ_RW_SYNC; - if (bio_rw_meta(bio)) - req->cmd_flags |= REQ_RW_META; - - req->errors = 0; - req->hard_sector = req->sector = bio->bi_sector; - req->ioprio = bio_prio(bio); - req->start_time = jiffies; - blk_rq_bio_prep(req->q, req, bio); -} - -static int __make_request(struct request_queue *q, struct bio *bio) -{ - struct request *req; - int el_ret, nr_sectors, barrier, err; - const unsigned short prio = bio_prio(bio); - const int sync = bio_sync(bio); - int rw_flags; - - nr_sectors = bio_sectors(bio); - - /* - * low level driver can indicate that it wants pages above a - * certain limit bounced to low memory (ie for highmem, or even - * ISA dma in theory) - */ - blk_queue_bounce(q, &bio); - - barrier = bio_barrier(bio); - if (unlikely(barrier) && (q->next_ordered == QUEUE_ORDERED_NONE)) { - err = -EOPNOTSUPP; - goto end_io; - } - - spin_lock_irq(q->queue_lock); - - if (unlikely(barrier) || elv_queue_empty(q)) - goto get_rq; - - el_ret = elv_merge(q, &req, bio); - switch (el_ret) { - case ELEVATOR_BACK_MERGE: - BUG_ON(!rq_mergeable(req)); - - if (!ll_back_merge_fn(q, req, bio)) - break; - - blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE); - - req->biotail->bi_next = bio; - req->biotail = bio; - req->nr_sectors = req->hard_nr_sectors += nr_sectors; - req->ioprio = ioprio_best(req->ioprio, prio); - drive_stat_acct(req, 0); - if (!attempt_back_merge(q, req)) - elv_merged_request(q, req, el_ret); - goto out; - - case ELEVATOR_FRONT_MERGE: - BUG_ON(!rq_mergeable(req)); - - if (!ll_front_merge_fn(q, req, bio)) - break; - - blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE); - - bio->bi_next = req->bio; - req->bio = bio; - - /* - * may not be valid. if the low level driver said - * it didn't need a bounce buffer then it better - * not touch req->buffer either... - */ - req->buffer = bio_data(bio); - req->current_nr_sectors = bio_cur_sectors(bio); - req->hard_cur_sectors = req->current_nr_sectors; - req->sector = req->hard_sector = bio->bi_sector; - req->nr_sectors = req->hard_nr_sectors += nr_sectors; - req->ioprio = ioprio_best(req->ioprio, prio); - drive_stat_acct(req, 0); - if (!attempt_front_merge(q, req)) - elv_merged_request(q, req, el_ret); - goto out; - - /* ELV_NO_MERGE: elevator says don't/can't merge. */ - default: - ; - } - -get_rq: - /* - * This sync check and mask will be re-done in init_request_from_bio(), - * but we need to set it earlier to expose the sync flag to the - * rq allocator and io schedulers. - */ - rw_flags = bio_data_dir(bio); - if (sync) - rw_flags |= REQ_RW_SYNC; - - /* - * Grab a free request. This is might sleep but can not fail. - * Returns with the queue unlocked. - */ - req = get_request_wait(q, rw_flags, bio); - - /* - * After dropping the lock and possibly sleeping here, our request - * may now be mergeable after it had proven unmergeable (above). - * We don't worry about that case for efficiency. It won't happen - * often, and the elevators are able to handle it. - */ - init_request_from_bio(req, bio); - - spin_lock_irq(q->queue_lock); - if (elv_queue_empty(q)) - blk_plug_device(q); - add_request(q, req); -out: - if (sync) - __generic_unplug_device(q); - - spin_unlock_irq(q->queue_lock); - return 0; - -end_io: - bio_endio(bio, err); - return 0; -} - -/* - * If bio->bi_dev is a partition, remap the location - */ -static inline void blk_partition_remap(struct bio *bio) -{ - struct block_device *bdev = bio->bi_bdev; - - if (bio_sectors(bio) && bdev != bdev->bd_contains) { - struct hd_struct *p = bdev->bd_part; - const int rw = bio_data_dir(bio); - - p->sectors[rw] += bio_sectors(bio); - p->ios[rw]++; - - bio->bi_sector += p->start_sect; - bio->bi_bdev = bdev->bd_contains; - - blk_add_trace_remap(bdev_get_queue(bio->bi_bdev), bio, - bdev->bd_dev, bio->bi_sector, - bio->bi_sector - p->start_sect); - } -} - -static void handle_bad_sector(struct bio *bio) -{ - char b[BDEVNAME_SIZE]; - - printk(KERN_INFO "attempt to access beyond end of device\n"); - printk(KERN_INFO "%s: rw=%ld, want=%Lu, limit=%Lu\n", - bdevname(bio->bi_bdev, b), - bio->bi_rw, - (unsigned long long)bio->bi_sector + bio_sectors(bio), - (long long)(bio->bi_bdev->bd_inode->i_size >> 9)); - - set_bit(BIO_EOF, &bio->bi_flags); -} - -#ifdef CONFIG_FAIL_MAKE_REQUEST - -static DECLARE_FAULT_ATTR(fail_make_request); - -static int __init setup_fail_make_request(char *str) -{ - return setup_fault_attr(&fail_make_request, str); -} -__setup("fail_make_request=", setup_fail_make_request); - -static int should_fail_request(struct bio *bio) -{ - if ((bio->bi_bdev->bd_disk->flags & GENHD_FL_FAIL) || - (bio->bi_bdev->bd_part && bio->bi_bdev->bd_part->make_it_fail)) - return should_fail(&fail_make_request, bio->bi_size); - - return 0; -} - -static int __init fail_make_request_debugfs(void) -{ - return init_fault_attr_dentries(&fail_make_request, - "fail_make_request"); -} - -late_initcall(fail_make_request_debugfs); - -#else /* CONFIG_FAIL_MAKE_REQUEST */ - -static inline int should_fail_request(struct bio *bio) -{ - return 0; -} - -#endif /* CONFIG_FAIL_MAKE_REQUEST */ - -/* - * Check whether this bio extends beyond the end of the device. - */ -static inline int bio_check_eod(struct bio *bio, unsigned int nr_sectors) -{ - sector_t maxsector; - - if (!nr_sectors) - return 0; - - /* Test device or partition size, when known. */ - maxsector = bio->bi_bdev->bd_inode->i_size >> 9; - if (maxsector) { - sector_t sector = bio->bi_sector; - - if (maxsector < nr_sectors || maxsector - nr_sectors < sector) { - /* - * This may well happen - the kernel calls bread() - * without checking the size of the device, e.g., when - * mounting a device. - */ - handle_bad_sector(bio); - return 1; - } - } - - return 0; -} - -/** - * generic_make_request: hand a buffer to its device driver for I/O - * @bio: The bio describing the location in memory and on the device. - * - * generic_make_request() is used to make I/O requests of block - * devices. It is passed a &struct bio, which describes the I/O that needs - * to be done. - * - * generic_make_request() does not return any status. The - * success/failure status of the request, along with notification of - * completion, is delivered asynchronously through the bio->bi_end_io - * function described (one day) else where. - * - * The caller of generic_make_request must make sure that bi_io_vec - * are set to describe the memory buffer, and that bi_dev and bi_sector are - * set to describe the device address, and the - * bi_end_io and optionally bi_private are set to describe how - * completion notification should be signaled. - * - * generic_make_request and the drivers it calls may use bi_next if this - * bio happens to be merged with someone else, and may change bi_dev and - * bi_sector for remaps as it sees fit. So the values of these fields - * should NOT be depended on after the call to generic_make_request. - */ -static inline void __generic_make_request(struct bio *bio) -{ - struct request_queue *q; - sector_t old_sector; - int ret, nr_sectors = bio_sectors(bio); - dev_t old_dev; - int err = -EIO; - - might_sleep(); - - if (bio_check_eod(bio, nr_sectors)) - goto end_io; - - /* - * Resolve the mapping until finished. (drivers are - * still free to implement/resolve their own stacking - * by explicitly returning 0) - * - * NOTE: we don't repeat the blk_size check for each new device. - * Stacking drivers are expected to know what they are doing. - */ - old_sector = -1; - old_dev = 0; - do { - char b[BDEVNAME_SIZE]; - - q = bdev_get_queue(bio->bi_bdev); - if (!q) { - printk(KERN_ERR - "generic_make_request: Trying to access " - "nonexistent block-device %s (%Lu)\n", - bdevname(bio->bi_bdev, b), - (long long) bio->bi_sector); -end_io: - bio_endio(bio, err); - break; - } - - if (unlikely(nr_sectors > q->max_hw_sectors)) { - printk("bio too big device %s (%u > %u)\n", - bdevname(bio->bi_bdev, b), - bio_sectors(bio), - q->max_hw_sectors); - goto end_io; - } - - if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) - goto end_io; - - if (should_fail_request(bio)) - goto end_io; - - /* - * If this device has partitions, remap block n - * of partition p to block n+start(p) of the disk. - */ - blk_partition_remap(bio); - - if (old_sector != -1) - blk_add_trace_remap(q, bio, old_dev, bio->bi_sector, - old_sector); - - blk_add_trace_bio(q, bio, BLK_TA_QUEUE); - - old_sector = bio->bi_sector; - old_dev = bio->bi_bdev->bd_dev; - - if (bio_check_eod(bio, nr_sectors)) - goto end_io; - if (bio_empty_barrier(bio) && !q->prepare_flush_fn) { - err = -EOPNOTSUPP; - goto end_io; - } - - ret = q->make_request_fn(q, bio); - } while (ret); -} - -/* - * We only want one ->make_request_fn to be active at a time, - * else stack usage with stacked devices could be a problem. - * So use current->bio_{list,tail} to keep a list of requests - * submited by a make_request_fn function. - * current->bio_tail is also used as a flag to say if - * generic_make_request is currently active in this task or not. - * If it is NULL, then no make_request is active. If it is non-NULL, - * then a make_request is active, and new requests should be added - * at the tail - */ -void generic_make_request(struct bio *bio) -{ - if (current->bio_tail) { - /* make_request is active */ - *(current->bio_tail) = bio; - bio->bi_next = NULL; - current->bio_tail = &bio->bi_next; - return; - } - /* following loop may be a bit non-obvious, and so deserves some - * explanation. - * Before entering the loop, bio->bi_next is NULL (as all callers - * ensure that) so we have a list with a single bio. - * We pretend that we have just taken it off a longer list, so - * we assign bio_list to the next (which is NULL) and bio_tail - * to &bio_list, thus initialising the bio_list of new bios to be - * added. __generic_make_request may indeed add some more bios - * through a recursive call to generic_make_request. If it - * did, we find a non-NULL value in bio_list and re-enter the loop - * from the top. In this case we really did just take the bio - * of the top of the list (no pretending) and so fixup bio_list and - * bio_tail or bi_next, and call into __generic_make_request again. - * - * The loop was structured like this to make only one call to - * __generic_make_request (which is important as it is large and - * inlined) and to keep the structure simple. - */ - BUG_ON(bio->bi_next); - do { - current->bio_list = bio->bi_next; - if (bio->bi_next == NULL) - current->bio_tail = ¤t->bio_list; - else - bio->bi_next = NULL; - __generic_make_request(bio); - bio = current->bio_list; - } while (bio); - current->bio_tail = NULL; /* deactivate */ -} - -EXPORT_SYMBOL(generic_make_request); - -/** - * submit_bio: submit a bio to the block device layer for I/O - * @rw: whether to %READ or %WRITE, or maybe to %READA (read ahead) - * @bio: The &struct bio which describes the I/O - * - * submit_bio() is very similar in purpose to generic_make_request(), and - * uses that function to do most of the work. Both are fairly rough - * interfaces, @bio must be presetup and ready for I/O. - * - */ -void submit_bio(int rw, struct bio *bio) -{ - int count = bio_sectors(bio); - - bio->bi_rw |= rw; - - /* - * If it's a regular read/write or a barrier with data attached, - * go through the normal accounting stuff before submission. - */ - if (!bio_empty_barrier(bio)) { - - BIO_BUG_ON(!bio->bi_size); - BIO_BUG_ON(!bio->bi_io_vec); - - if (rw & WRITE) { - count_vm_events(PGPGOUT, count); - } else { - task_io_account_read(bio->bi_size); - count_vm_events(PGPGIN, count); - } - - if (unlikely(block_dump)) { - char b[BDEVNAME_SIZE]; - printk(KERN_DEBUG "%s(%d): %s block %Lu on %s\n", - current->comm, task_pid_nr(current), - (rw & WRITE) ? "WRITE" : "READ", - (unsigned long long)bio->bi_sector, - bdevname(bio->bi_bdev,b)); - } - } - - generic_make_request(bio); -} - -EXPORT_SYMBOL(submit_bio); - -static void blk_recalc_rq_sectors(struct request *rq, int nsect) -{ - if (blk_fs_request(rq)) { - rq->hard_sector += nsect; - rq->hard_nr_sectors -= nsect; - - /* - * Move the I/O submission pointers ahead if required. - */ - if ((rq->nr_sectors >= rq->hard_nr_sectors) && - (rq->sector <= rq->hard_sector)) { - rq->sector = rq->hard_sector; - rq->nr_sectors = rq->hard_nr_sectors; - rq->hard_cur_sectors = bio_cur_sectors(rq->bio); - rq->current_nr_sectors = rq->hard_cur_sectors; - rq->buffer = bio_data(rq->bio); - } - - /* - * if total number of sectors is less than the first segment - * size, something has gone terribly wrong - */ - if (rq->nr_sectors < rq->current_nr_sectors) { - printk("blk: request botched\n"); - rq->nr_sectors = rq->current_nr_sectors; - } - } -} - -static int __end_that_request_first(struct request *req, int uptodate, - int nr_bytes) -{ - int total_bytes, bio_nbytes, error, next_idx = 0; - struct bio *bio; - - blk_add_trace_rq(req->q, req, BLK_TA_COMPLETE); - - /* - * extend uptodate bool to allow < 0 value to be direct io error - */ - error = 0; - if (end_io_error(uptodate)) - error = !uptodate ? -EIO : uptodate; - - /* - * for a REQ_BLOCK_PC request, we want to carry any eventual - * sense key with us all the way through - */ - if (!blk_pc_request(req)) - req->errors = 0; - - if (!uptodate) { - if (blk_fs_request(req) && !(req->cmd_flags & REQ_QUIET)) - printk("end_request: I/O error, dev %s, sector %llu\n", - req->rq_disk ? req->rq_disk->disk_name : "?", - (unsigned long long)req->sector); - } - - if (blk_fs_request(req) && req->rq_disk) { - const int rw = rq_data_dir(req); - - disk_stat_add(req->rq_disk, sectors[rw], nr_bytes >> 9); - } - - total_bytes = bio_nbytes = 0; - while ((bio = req->bio) != NULL) { - int nbytes; - - /* - * For an empty barrier request, the low level driver must - * store a potential error location in ->sector. We pass - * that back up in ->bi_sector. - */ - if (blk_empty_barrier(req)) - bio->bi_sector = req->sector; - - if (nr_bytes >= bio->bi_size) { - req->bio = bio->bi_next; - nbytes = bio->bi_size; - req_bio_endio(req, bio, nbytes, error); - next_idx = 0; - bio_nbytes = 0; - } else { - int idx = bio->bi_idx + next_idx; - - if (unlikely(bio->bi_idx >= bio->bi_vcnt)) { - blk_dump_rq_flags(req, "__end_that"); - printk("%s: bio idx %d >= vcnt %d\n", - __FUNCTION__, - bio->bi_idx, bio->bi_vcnt); - break; - } - - nbytes = bio_iovec_idx(bio, idx)->bv_len; - BIO_BUG_ON(nbytes > bio->bi_size); - - /* - * not a complete bvec done - */ - if (unlikely(nbytes > nr_bytes)) { - bio_nbytes += nr_bytes; - total_bytes += nr_bytes; - break; - } - - /* - * advance to the next vector - */ - next_idx++; - bio_nbytes += nbytes; - } - - total_bytes += nbytes; - nr_bytes -= nbytes; - - if ((bio = req->bio)) { - /* - * end more in this run, or just return 'not-done' - */ - if (unlikely(nr_bytes <= 0)) - break; - } - } - - /* - * completely done - */ - if (!req->bio) - return 0; - - /* - * if the request wasn't completed, update state - */ - if (bio_nbytes) { - req_bio_endio(req, bio, bio_nbytes, error); - bio->bi_idx += next_idx; - bio_iovec(bio)->bv_offset += nr_bytes; - bio_iovec(bio)->bv_len -= nr_bytes; - } - - blk_recalc_rq_sectors(req, total_bytes >> 9); - blk_recalc_rq_segments(req); - return 1; -} - -/** - * end_that_request_first - end I/O on a request - * @req: the request being processed - * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error - * @nr_sectors: number of sectors to end I/O on - * - * Description: - * Ends I/O on a number of sectors attached to @req, and sets it up - * for the next range of segments (if any) in the cluster. - * - * Return: - * 0 - we are done with this request, call end_that_request_last() - * 1 - still buffers pending for this request - **/ -int end_that_request_first(struct request *req, int uptodate, int nr_sectors) -{ - return __end_that_request_first(req, uptodate, nr_sectors << 9); -} - -EXPORT_SYMBOL(end_that_request_first); - -/** - * end_that_request_chunk - end I/O on a request - * @req: the request being processed - * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error - * @nr_bytes: number of bytes to complete - * - * Description: - * Ends I/O on a number of bytes attached to @req, and sets it up - * for the next range of segments (if any). Like end_that_request_first(), - * but deals with bytes instead of sectors. - * - * Return: - * 0 - we are done with this request, call end_that_request_last() - * 1 - still buffers pending for this request - **/ -int end_that_request_chunk(struct request *req, int uptodate, int nr_bytes) -{ - return __end_that_request_first(req, uptodate, nr_bytes); -} - -EXPORT_SYMBOL(end_that_request_chunk); - -/* - * splice the completion data to a local structure and hand off to - * process_completion_queue() to complete the requests - */ -static void blk_done_softirq(struct softirq_action *h) -{ - struct list_head *cpu_list, local_list; - - local_irq_disable(); - cpu_list = &__get_cpu_var(blk_cpu_done); - list_replace_init(cpu_list, &local_list); - local_irq_enable(); - - while (!list_empty(&local_list)) { - struct request *rq = list_entry(local_list.next, struct request, donelist); - - list_del_init(&rq->donelist); - rq->q->softirq_done_fn(rq); - } -} - -static int __cpuinit blk_cpu_notify(struct notifier_block *self, unsigned long action, - void *hcpu) -{ - /* - * If a CPU goes away, splice its entries to the current CPU - * and trigger a run of the softirq - */ - if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) { - int cpu = (unsigned long) hcpu; - - local_irq_disable(); - list_splice_init(&per_cpu(blk_cpu_done, cpu), - &__get_cpu_var(blk_cpu_done)); - raise_softirq_irqoff(BLOCK_SOFTIRQ); - local_irq_enable(); - } - - return NOTIFY_OK; -} - - -static struct notifier_block blk_cpu_notifier __cpuinitdata = { - .notifier_call = blk_cpu_notify, -}; - -/** - * blk_complete_request - end I/O on a request - * @req: the request being processed - * - * Description: - * Ends all I/O on a request. It does not handle partial completions, - * unless the driver actually implements this in its completion callback - * through requeueing. The actual completion happens out-of-order, - * through a softirq handler. The user must have registered a completion - * callback through blk_queue_softirq_done(). - **/ - -void blk_complete_request(struct request *req) -{ - struct list_head *cpu_list; - unsigned long flags; - - BUG_ON(!req->q->softirq_done_fn); - - local_irq_save(flags); - - cpu_list = &__get_cpu_var(blk_cpu_done); - list_add_tail(&req->donelist, cpu_list); - raise_softirq_irqoff(BLOCK_SOFTIRQ); - - local_irq_restore(flags); -} - -EXPORT_SYMBOL(blk_complete_request); - -/* - * queue lock must be held - */ -void end_that_request_last(struct request *req, int uptodate) -{ - struct gendisk *disk = req->rq_disk; - int error; - - /* - * extend uptodate bool to allow < 0 value to be direct io error - */ - error = 0; - if (end_io_error(uptodate)) - error = !uptodate ? -EIO : uptodate; - - if (unlikely(laptop_mode) && blk_fs_request(req)) - laptop_io_completion(); - - /* - * Account IO completion. bar_rq isn't accounted as a normal - * IO on queueing nor completion. Accounting the containing - * request is enough. - */ - if (disk && blk_fs_request(req) && req != &req->q->bar_rq) { - unsigned long duration = jiffies - req->start_time; - const int rw = rq_data_dir(req); - - __disk_stat_inc(disk, ios[rw]); - __disk_stat_add(disk, ticks[rw], duration); - disk_round_stats(disk); - disk->in_flight--; - } - if (req->end_io) - req->end_io(req, error); - else - __blk_put_request(req->q, req); -} - -EXPORT_SYMBOL(end_that_request_last); - -static inline void __end_request(struct request *rq, int uptodate, - unsigned int nr_bytes, int dequeue) -{ - if (!end_that_request_chunk(rq, uptodate, nr_bytes)) { - if (dequeue) - blkdev_dequeue_request(rq); - add_disk_randomness(rq->rq_disk); - end_that_request_last(rq, uptodate); - } -} - -static unsigned int rq_byte_size(struct request *rq) -{ - if (blk_fs_request(rq)) - return rq->hard_nr_sectors << 9; - - return rq->data_len; -} - -/** - * end_queued_request - end all I/O on a queued request - * @rq: the request being processed - * @uptodate: error value or 0/1 uptodate flag - * - * Description: - * Ends all I/O on a request, and removes it from the block layer queues. - * Not suitable for normal IO completion, unless the driver still has - * the request attached to the block layer. - * - **/ -void end_queued_request(struct request *rq, int uptodate) -{ - __end_request(rq, uptodate, rq_byte_size(rq), 1); -} -EXPORT_SYMBOL(end_queued_request); - -/** - * end_dequeued_request - end all I/O on a dequeued request - * @rq: the request being processed - * @uptodate: error value or 0/1 uptodate flag - * - * Description: - * Ends all I/O on a request. The request must already have been - * dequeued using blkdev_dequeue_request(), as is normally the case - * for most drivers. - * - **/ -void end_dequeued_request(struct request *rq, int uptodate) -{ - __end_request(rq, uptodate, rq_byte_size(rq), 0); -} -EXPORT_SYMBOL(end_dequeued_request); - - -/** - * end_request - end I/O on the current segment of the request - * @req: the request being processed - * @uptodate: error value or 0/1 uptodate flag - * - * Description: - * Ends I/O on the current segment of a request. If that is the only - * remaining segment, the request is also completed and freed. - * - * This is a remnant of how older block drivers handled IO completions. - * Modern drivers typically end IO on the full request in one go, unless - * they have a residual value to account for. For that case this function - * isn't really useful, unless the residual just happens to be the - * full current segment. In other words, don't use this function in new - * code. Either use end_request_completely(), or the - * end_that_request_chunk() (along with end_that_request_last()) for - * partial completions. - * - **/ -void end_request(struct request *req, int uptodate) -{ - __end_request(req, uptodate, req->hard_cur_sectors << 9, 1); -} -EXPORT_SYMBOL(end_request); - -static void blk_rq_bio_prep(struct request_queue *q, struct request *rq, - struct bio *bio) -{ - /* first two bits are identical in rq->cmd_flags and bio->bi_rw */ - rq->cmd_flags |= (bio->bi_rw & 3); - - rq->nr_phys_segments = bio_phys_segments(q, bio); - rq->nr_hw_segments = bio_hw_segments(q, bio); - rq->current_nr_sectors = bio_cur_sectors(bio); - rq->hard_cur_sectors = rq->current_nr_sectors; - rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio); - rq->buffer = bio_data(bio); - rq->data_len = bio->bi_size; - - rq->bio = rq->biotail = bio; - - if (bio->bi_bdev) - rq->rq_disk = bio->bi_bdev->bd_disk; -} - -int kblockd_schedule_work(struct work_struct *work) -{ - return queue_work(kblockd_workqueue, work); -} - -EXPORT_SYMBOL(kblockd_schedule_work); - -void kblockd_flush_work(struct work_struct *work) -{ - cancel_work_sync(work); -} -EXPORT_SYMBOL(kblockd_flush_work); - -int __init blk_dev_init(void) -{ - int i; - - kblockd_workqueue = create_workqueue("kblockd"); - if (!kblockd_workqueue) - panic("Failed to create kblockd\n"); - - request_cachep = kmem_cache_create("blkdev_requests", - sizeof(struct request), 0, SLAB_PANIC, NULL); - - requestq_cachep = kmem_cache_create("blkdev_queue", - sizeof(struct request_queue), 0, SLAB_PANIC, NULL); - - iocontext_cachep = kmem_cache_create("blkdev_ioc", - sizeof(struct io_context), 0, SLAB_PANIC, NULL); - - for_each_possible_cpu(i) - INIT_LIST_HEAD(&per_cpu(blk_cpu_done, i)); - - open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL); - register_hotcpu_notifier(&blk_cpu_notifier); - - blk_max_low_pfn = max_low_pfn - 1; - blk_max_pfn = max_pfn - 1; - - return 0; -} - -/* - * IO Context helper functions - */ -void put_io_context(struct io_context *ioc) -{ - if (ioc == NULL) - return; - - BUG_ON(atomic_read(&ioc->refcount) == 0); - - if (atomic_dec_and_test(&ioc->refcount)) { - struct cfq_io_context *cic; - - rcu_read_lock(); - if (ioc->aic && ioc->aic->dtor) - ioc->aic->dtor(ioc->aic); - if (ioc->cic_root.rb_node != NULL) { - struct rb_node *n = rb_first(&ioc->cic_root); - - cic = rb_entry(n, struct cfq_io_context, rb_node); - cic->dtor(ioc); - } - rcu_read_unlock(); - - kmem_cache_free(iocontext_cachep, ioc); - } -} -EXPORT_SYMBOL(put_io_context); - -/* Called by the exitting task */ -void exit_io_context(void) -{ - struct io_context *ioc; - struct cfq_io_context *cic; - - task_lock(current); - ioc = current->io_context; - current->io_context = NULL; - task_unlock(current); - - ioc->task = NULL; - if (ioc->aic && ioc->aic->exit) - ioc->aic->exit(ioc->aic); - if (ioc->cic_root.rb_node != NULL) { - cic = rb_entry(rb_first(&ioc->cic_root), struct cfq_io_context, rb_node); - cic->exit(ioc); - } - - put_io_context(ioc); -} - -/* - * If the current task has no IO context then create one and initialise it. - * Otherwise, return its existing IO context. - * - * This returned IO context doesn't have a specifically elevated refcount, - * but since the current task itself holds a reference, the context can be - * used in general code, so long as it stays within `current` context. - */ -static struct io_context *current_io_context(gfp_t gfp_flags, int node) -{ - struct task_struct *tsk = current; - struct io_context *ret; - - ret = tsk->io_context; - if (likely(ret)) - return ret; - - ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node); - if (ret) { - atomic_set(&ret->refcount, 1); - ret->task = current; - ret->ioprio_changed = 0; - ret->last_waited = jiffies; /* doesn't matter... */ - ret->nr_batch_requests = 0; /* because this is 0 */ - ret->aic = NULL; - ret->cic_root.rb_node = NULL; - ret->ioc_data = NULL; - /* make sure set_task_ioprio() sees the settings above */ - smp_wmb(); - tsk->io_context = ret; - } - - return ret; -} - -/* - * If the current task has no IO context then create one and initialise it. - * If it does have a context, take a ref on it. - * - * This is always called in the context of the task which submitted the I/O. - */ -struct io_context *get_io_context(gfp_t gfp_flags, int node) -{ - struct io_context *ret; - ret = current_io_context(gfp_flags, node); - if (likely(ret)) - atomic_inc(&ret->refcount); - return ret; -} -EXPORT_SYMBOL(get_io_context); - -void copy_io_context(struct io_context **pdst, struct io_context **psrc) -{ - struct io_context *src = *psrc; - struct io_context *dst = *pdst; - - if (src) { - BUG_ON(atomic_read(&src->refcount) == 0); - atomic_inc(&src->refcount); - put_io_context(dst); - *pdst = src; - } -} -EXPORT_SYMBOL(copy_io_context); - -void swap_io_context(struct io_context **ioc1, struct io_context **ioc2) -{ - struct io_context *temp; - temp = *ioc1; - *ioc1 = *ioc2; - *ioc2 = temp; -} -EXPORT_SYMBOL(swap_io_context); - -/* - * sysfs parts below - */ -struct queue_sysfs_entry { - struct attribute attr; - ssize_t (*show)(struct request_queue *, char *); - ssize_t (*store)(struct request_queue *, const char *, size_t); -}; - -static ssize_t -queue_var_show(unsigned int var, char *page) -{ - return sprintf(page, "%d\n", var); -} - -static ssize_t -queue_var_store(unsigned long *var, const char *page, size_t count) -{ - char *p = (char *) page; - - *var = simple_strtoul(p, &p, 10); - return count; -} - -static ssize_t queue_requests_show(struct request_queue *q, char *page) -{ - return queue_var_show(q->nr_requests, (page)); -} - -static ssize_t -queue_requests_store(struct request_queue *q, const char *page, size_t count) -{ - struct request_list *rl = &q->rq; - unsigned long nr; - int ret = queue_var_store(&nr, page, count); - if (nr < BLKDEV_MIN_RQ) - nr = BLKDEV_MIN_RQ; - - spin_lock_irq(q->queue_lock); - q->nr_requests = nr; - blk_queue_congestion_threshold(q); - - if (rl->count[READ] >= queue_congestion_on_threshold(q)) - blk_set_queue_congested(q, READ); - else if (rl->count[READ] < queue_congestion_off_threshold(q)) - blk_clear_queue_congested(q, READ); - - if (rl->count[WRITE] >= queue_congestion_on_threshold(q)) - blk_set_queue_congested(q, WRITE); - else if (rl->count[WRITE] < queue_congestion_off_threshold(q)) - blk_clear_queue_congested(q, WRITE); - - if (rl->count[READ] >= q->nr_requests) { - blk_set_queue_full(q, READ); - } else if (rl->count[READ]+1 <= q->nr_requests) { - blk_clear_queue_full(q, READ); - wake_up(&rl->wait[READ]); - } - - if (rl->count[WRITE] >= q->nr_requests) { - blk_set_queue_full(q, WRITE); - } else if (rl->count[WRITE]+1 <= q->nr_requests) { - blk_clear_queue_full(q, WRITE); - wake_up(&rl->wait[WRITE]); - } - spin_unlock_irq(q->queue_lock); - return ret; -} - -static ssize_t queue_ra_show(struct request_queue *q, char *page) -{ - int ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10); - - return queue_var_show(ra_kb, (page)); -} - -static ssize_t -queue_ra_store(struct request_queue *q, const char *page, size_t count) -{ - unsigned long ra_kb; - ssize_t ret = queue_var_store(&ra_kb, page, count); - - spin_lock_irq(q->queue_lock); - q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10); - spin_unlock_irq(q->queue_lock); - - return ret; -} - -static ssize_t queue_max_sectors_show(struct request_queue *q, char *page) -{ - int max_sectors_kb = q->max_sectors >> 1; - - return queue_var_show(max_sectors_kb, (page)); -} - -static ssize_t -queue_max_sectors_store(struct request_queue *q, const char *page, size_t count) -{ - unsigned long max_sectors_kb, - max_hw_sectors_kb = q->max_hw_sectors >> 1, - page_kb = 1 << (PAGE_CACHE_SHIFT - 10); - ssize_t ret = queue_var_store(&max_sectors_kb, page, count); - - if (max_sectors_kb > max_hw_sectors_kb || max_sectors_kb < page_kb) - return -EINVAL; - /* - * Take the queue lock to update the readahead and max_sectors - * values synchronously: - */ - spin_lock_irq(q->queue_lock); - q->max_sectors = max_sectors_kb << 1; - spin_unlock_irq(q->queue_lock); - - return ret; -} - -static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page) -{ - int max_hw_sectors_kb = q->max_hw_sectors >> 1; - - return queue_var_show(max_hw_sectors_kb, (page)); -} - - -static struct queue_sysfs_entry queue_requests_entry = { - .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR }, - .show = queue_requests_show, - .store = queue_requests_store, -}; - -static struct queue_sysfs_entry queue_ra_entry = { - .attr = {.name = "read_ahead_kb", .mode = S_IRUGO | S_IWUSR }, - .show = queue_ra_show, - .store = queue_ra_store, -}; - -static struct queue_sysfs_entry queue_max_sectors_entry = { - .attr = {.name = "max_sectors_kb", .mode = S_IRUGO | S_IWUSR }, - .show = queue_max_sectors_show, - .store = queue_max_sectors_store, -}; - -static struct queue_sysfs_entry queue_max_hw_sectors_entry = { - .attr = {.name = "max_hw_sectors_kb", .mode = S_IRUGO }, - .show = queue_max_hw_sectors_show, -}; - -static struct queue_sysfs_entry queue_iosched_entry = { - .attr = {.name = "scheduler", .mode = S_IRUGO | S_IWUSR }, - .show = elv_iosched_show, - .store = elv_iosched_store, -}; - -static struct attribute *default_attrs[] = { - &queue_requests_entry.attr, - &queue_ra_entry.attr, - &queue_max_hw_sectors_entry.attr, - &queue_max_sectors_entry.attr, - &queue_iosched_entry.attr, - NULL, -}; - -#define to_queue(atr) container_of((atr), struct queue_sysfs_entry, attr) - -static ssize_t -queue_attr_show(struct kobject *kobj, struct attribute *attr, char *page) -{ - struct queue_sysfs_entry *entry = to_queue(attr); - struct request_queue *q = - container_of(kobj, struct request_queue, kobj); - ssize_t res; - - if (!entry->show) - return -EIO; - mutex_lock(&q->sysfs_lock); - if (test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)) { - mutex_unlock(&q->sysfs_lock); - return -ENOENT; - } - res = entry->show(q, page); - mutex_unlock(&q->sysfs_lock); - return res; -} - -static ssize_t -queue_attr_store(struct kobject *kobj, struct attribute *attr, - const char *page, size_t length) -{ - struct queue_sysfs_entry *entry = to_queue(attr); - struct request_queue *q = container_of(kobj, struct request_queue, kobj); - - ssize_t res; - - if (!entry->store) - return -EIO; - mutex_lock(&q->sysfs_lock); - if (test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)) { - mutex_unlock(&q->sysfs_lock); - return -ENOENT; - } - res = entry->store(q, page, length); - mutex_unlock(&q->sysfs_lock); - return res; -} - -static struct sysfs_ops queue_sysfs_ops = { - .show = queue_attr_show, - .store = queue_attr_store, -}; - -static struct kobj_type queue_ktype = { - .sysfs_ops = &queue_sysfs_ops, - .default_attrs = default_attrs, - .release = blk_release_queue, -}; - -int blk_register_queue(struct gendisk *disk) -{ - int ret; - - struct request_queue *q = disk->queue; - - if (!q || !q->request_fn) - return -ENXIO; - - q->kobj.parent = kobject_get(&disk->kobj); - - ret = kobject_add(&q->kobj); - if (ret < 0) - return ret; - - kobject_uevent(&q->kobj, KOBJ_ADD); - - ret = elv_register_queue(q); - if (ret) { - kobject_uevent(&q->kobj, KOBJ_REMOVE); - kobject_del(&q->kobj); - return ret; - } - - return 0; -} - -void blk_unregister_queue(struct gendisk *disk) -{ - struct request_queue *q = disk->queue; - - if (q && q->request_fn) { - elv_unregister_queue(q); - - kobject_uevent(&q->kobj, KOBJ_REMOVE); - kobject_del(&q->kobj); - kobject_put(&disk->kobj); - } -} |