X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=block%2Fll_rw_blk.c;h=0f82e12f7b678553b51189374bfb562c676de919;hb=4549df891a31b9a05b7d183106c09049b79327be;hp=4b7b4461e8d6032d23e669de5a83f4b15a7d0c50;hpb=1ea25ecb7256978947c258f08a30c878eebe9edb;p=linux-2.6

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 4b7b4461e8..0f82e12f7b 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -39,6 +39,7 @@ static void blk_unplug_timeout(unsigned long data);
 static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io);
 static void init_request_from_bio(struct request *req, struct bio *bio);
 static int __make_request(request_queue_t *q, struct bio *bio);
+static struct io_context *current_io_context(gfp_t gfp_flags, int node);
 
 /*
  * For the allocated request tables
@@ -55,11 +56,6 @@ static kmem_cache_t *requestq_cachep;
  */
 static kmem_cache_t *iocontext_cachep;
 
-static wait_queue_head_t congestion_wqh[2] = {
-		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
-		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
-	};
-
 /*
  * Controlling structure to kblockd
  */
@@ -111,35 +107,6 @@ static void blk_queue_congestion_threshold(struct request_queue *q)
 	q->nr_congestion_off = nr;
 }
 
-/*
- * A queue has just exitted congestion.  Note this in the global counter of
- * congested queues, and wake up anyone who was waiting for requests to be
- * put back.
- */
-static void clear_queue_congested(request_queue_t *q, int rw)
-{
-	enum bdi_state bit;
-	wait_queue_head_t *wqh = &congestion_wqh[rw];
-
-	bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested;
-	clear_bit(bit, &q->backing_dev_info.state);
-	smp_mb__after_clear_bit();
-	if (waitqueue_active(wqh))
-		wake_up(wqh);
-}
-
-/*
- * A queue has just entered congestion.  Flag that in the queue's VM-visible
- * state flags and increment the global gounter of congested queues.
- */
-static void set_queue_congested(request_queue_t *q, int rw)
-{
-	enum bdi_state bit;
-
-	bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested;
-	set_bit(bit, &q->backing_dev_info.state);
-}
-
 /**
  * blk_get_backing_dev_info - get the address of a queue's backing_dev_info
  * @bdev:	device
@@ -158,7 +125,6 @@ struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev)
 		ret = &q->backing_dev_info;
 	return ret;
 }
-
 EXPORT_SYMBOL(blk_get_backing_dev_info);
 
 void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data)
@@ -166,7 +132,6 @@ void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data)
 	q->activity_fn = fn;
 	q->activity_data = data;
 }
-
 EXPORT_SYMBOL(blk_queue_activity_fn);
 
 /**
@@ -839,12 +804,7 @@ EXPORT_SYMBOL(blk_queue_dma_alignment);
  **/
 struct request *blk_queue_find_tag(request_queue_t *q, int tag)
 {
-	struct blk_queue_tag *bqt = q->queue_tags;
-
-	if (unlikely(bqt == NULL || tag >= bqt->real_max_depth))
-		return NULL;
-
-	return bqt->tag_index[tag];
+	return blk_map_queue_find_tag(q->queue_tags, tag);
 }
 
 EXPORT_SYMBOL(blk_queue_find_tag);
@@ -1170,11 +1130,16 @@ int blk_queue_start_tag(request_queue_t *q, struct request *rq)
 		BUG();
 	}
 
-	tag = find_first_zero_bit(bqt->tag_map, bqt->max_depth);
-	if (tag >= bqt->max_depth)
-		return 1;
+	/*
+	 * 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;
 
-	__set_bit(tag, bqt->tag_map);
+	} while (test_and_set_bit(tag, bqt->tag_map));
 
 	rq->cmd_flags |= REQ_QUEUED;
 	rq->tag = tag;
@@ -2066,7 +2031,7 @@ static void __freed_request(request_queue_t *q, int rw)
 	struct request_list *rl = &q->rq;
 
 	if (rl->count[rw] < queue_congestion_off_threshold(q))
-		clear_queue_congested(q, rw);
+		blk_clear_queue_congested(q, rw);
 
 	if (rl->count[rw] + 1 <= q->nr_requests) {
 		if (waitqueue_active(&rl->wait[rw]))
@@ -2114,7 +2079,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 
 	if (rl->count[rw]+1 >= queue_congestion_on_threshold(q)) {
 		if (rl->count[rw]+1 >= q->nr_requests) {
-			ioc = current_io_context(GFP_ATOMIC);
+			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".
@@ -2136,7 +2101,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 				}
 			}
 		}
-		set_queue_congested(q, rw);
+		blk_set_queue_congested(q, rw);
 	}
 
 	/*
@@ -2234,7 +2199,7 @@ static struct request *get_request_wait(request_queue_t *q, int rw,
 			 * up to a big batch of them for a small period time.
 			 * See ioc_batching, ioc_set_batching
 			 */
-			ioc = current_io_context(GFP_NOIO);
+			ioc = current_io_context(GFP_NOIO, q->node);
 			ioc_set_batching(q, ioc);
 
 			spin_lock_irq(q->queue_lock);
@@ -2265,6 +2230,25 @@ struct request *blk_get_request(request_queue_t *q, int rw, gfp_t gfp_mask)
 }
 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(request_queue_t *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
@@ -2332,16 +2316,90 @@ void blk_insert_request(request_queue_t *q, struct request *rq,
 
 	drive_stat_acct(rq, rq->nr_sectors, 1);
 	__elv_add_request(q, rq, where, 0);
-
-	if (blk_queue_plugged(q))
-		__generic_unplug_device(q);
-	else
-		q->request_fn(q);
+	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;
+}
+
+static int __blk_rq_map_user(request_queue_t *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);
+
+	/*
+	 * for most (all? don't know of any) queues we could
+	 * skip grabbing the queue lock here. only drivers with
+	 * funky private ->back_merge_fn() function could be
+	 * problematic.
+	 */
+	spin_lock_irq(q->queue_lock);
+	if (!rq->bio)
+		blk_rq_bio_prep(q, rq, bio);
+	else if (!q->back_merge_fn(q, rq, bio)) {
+		ret = -EINVAL;
+		spin_unlock_irq(q->queue_lock);
+		goto unmap_bio;
+	} else {
+		rq->biotail->bi_next = bio;
+		rq->biotail = bio;
+
+		rq->nr_sectors += bio_sectors(bio);
+		rq->hard_nr_sectors = rq->nr_sectors;
+		rq->data_len += bio->bi_size;
+	}
+	spin_unlock_irq(q->queue_lock);
+
+	return bio->bi_size;
+
+unmap_bio:
+	/* if it was boucned we must call the end io function */
+	bio_endio(bio, bio->bi_size, 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
@@ -2363,42 +2421,44 @@ EXPORT_SYMBOL(blk_insert_request);
  *    unmapping.
  */
 int blk_rq_map_user(request_queue_t *q, struct request *rq, void __user *ubuf,
-		    unsigned int len)
+		    unsigned long len)
 {
-	unsigned long uaddr;
-	struct bio *bio;
-	int reading;
+	unsigned long bytes_read = 0;
+	int ret;
 
 	if (len > (q->max_hw_sectors << 9))
 		return -EINVAL;
 	if (!len || !ubuf)
 		return -EINVAL;
 
-	reading = rq_data_dir(rq) == READ;
+	while (bytes_read != len) {
+		unsigned long map_len, end, start;
 
-	/*
-	 * 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);
+		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;
 
-	if (!IS_ERR(bio)) {
-		rq->bio = rq->biotail = bio;
-		blk_rq_bio_prep(q, rq, bio);
+		/*
+		 * 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;
 
-		rq->buffer = rq->data = NULL;
-		rq->data_len = len;
-		return 0;
+		ret = __blk_rq_map_user(q, rq, ubuf, map_len);
+		if (ret < 0)
+			goto unmap_rq;
+		bytes_read += ret;
+		ubuf += ret;
 	}
 
-	/*
-	 * bio is the err-ptr
-	 */
-	return PTR_ERR(bio);
+	rq->buffer = rq->data = NULL;
+	return 0;
+unmap_rq:
+	blk_rq_unmap_user(rq);
+	return ret;
 }
 
 EXPORT_SYMBOL(blk_rq_map_user);
@@ -2424,7 +2484,7 @@ EXPORT_SYMBOL(blk_rq_map_user);
  *    unmapping.
  */
 int blk_rq_map_user_iov(request_queue_t *q, struct request *rq,
-			struct sg_iovec *iov, int iov_count)
+			struct sg_iovec *iov, int iov_count, unsigned int len)
 {
 	struct bio *bio;
 
@@ -2438,10 +2498,15 @@ int blk_rq_map_user_iov(request_queue_t *q, struct request *rq,
 	if (IS_ERR(bio))
 		return PTR_ERR(bio);
 
-	rq->bio = rq->biotail = bio;
+	if (bio->bi_size != len) {
+		bio_endio(bio, bio->bi_size, 0);
+		bio_unmap_user(bio);
+		return -EINVAL;
+	}
+
+	bio_get(bio);
 	blk_rq_bio_prep(q, rq, bio);
 	rq->buffer = rq->data = NULL;
-	rq->data_len = bio->bi_size;
 	return 0;
 }
 
@@ -2449,23 +2514,26 @@ EXPORT_SYMBOL(blk_rq_map_user_iov);
 
 /**
  * blk_rq_unmap_user - unmap a request with user data
- * @bio:	bio to be unmapped
- * @ulen:	length of user buffer
+ * @rq:		rq to be unmapped
  *
  * Description:
- *    Unmap a bio previously mapped by blk_rq_map_user().
+ *    Unmap a rq previously mapped by blk_rq_map_user().
+ *    rq->bio must be set to the original head of the request.
  */
-int blk_rq_unmap_user(struct bio *bio, unsigned int ulen)
+int blk_rq_unmap_user(struct request *rq)
 {
-	int ret = 0;
+	struct bio *bio, *mapped_bio;
 
-	if (bio) {
-		if (bio_flagged(bio, BIO_USER_MAPPED))
-			bio_unmap_user(bio);
+	while ((bio = rq->bio)) {
+		if (bio_flagged(bio, BIO_BOUNCED))
+			mapped_bio = bio->bi_private;
 		else
-			ret = bio_uncopy_user(bio);
-	}
+			mapped_bio = bio;
 
+		__blk_rq_unmap_user(mapped_bio);
+		rq->bio = bio->bi_next;
+		bio_put(bio);
+	}
 	return 0;
 }
 
@@ -2496,11 +2564,8 @@ int blk_rq_map_kern(request_queue_t *q, struct request *rq, void *kbuf,
 	if (rq_data_dir(rq) == WRITE)
 		bio->bi_rw |= (1 << BIO_RW);
 
-	rq->bio = rq->biotail = bio;
 	blk_rq_bio_prep(q, rq, bio);
-
 	rq->buffer = rq->data = NULL;
-	rq->data_len = len;
 	return 0;
 }
 
@@ -2739,41 +2804,6 @@ void blk_end_sync_rq(struct request *rq, int error)
 }
 EXPORT_SYMBOL(blk_end_sync_rq);
 
-/**
- * blk_congestion_wait - wait for a queue to become uncongested
- * @rw: READ or WRITE
- * @timeout: timeout in jiffies
- *
- * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion.
- * If no queues are congested then just wait for the next request to be
- * returned.
- */
-long blk_congestion_wait(int rw, long timeout)
-{
-	long ret;
-	DEFINE_WAIT(wait);
-	wait_queue_head_t *wqh = &congestion_wqh[rw];
-
-	prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
-	ret = io_schedule_timeout(timeout);
-	finish_wait(wqh, &wait);
-	return ret;
-}
-
-EXPORT_SYMBOL(blk_congestion_wait);
-
-/**
- * blk_congestion_end - wake up sleepers on a congestion queue
- * @rw: READ or WRITE
- */
-void blk_congestion_end(int rw)
-{
-	wait_queue_head_t *wqh = &congestion_wqh[rw];
-
-	if (waitqueue_active(wqh))
-		wake_up(wqh);
-}
-
 /*
  * Has to be called with the request spinlock acquired
  */
@@ -2868,6 +2898,8 @@ static void init_request_from_bio(struct request *req, struct bio *bio)
 
 	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;
@@ -3052,6 +3084,7 @@ void generic_make_request(struct bio *bio)
 {
 	request_queue_t *q;
 	sector_t maxsector;
+	sector_t old_sector;
 	int ret, nr_sectors = bio_sectors(bio);
 	dev_t old_dev;
 
@@ -3080,7 +3113,7 @@ void generic_make_request(struct bio *bio)
 	 * NOTE: we don't repeat the blk_size check for each new device.
 	 * Stacking drivers are expected to know what they are doing.
 	 */
-	maxsector = -1;
+	old_sector = -1;
 	old_dev = 0;
 	do {
 		char b[BDEVNAME_SIZE];
@@ -3114,15 +3147,31 @@ end_io:
 		 */
 		blk_partition_remap(bio);
 
-		if (maxsector != -1)
+		if (old_sector != -1)
 			blk_add_trace_remap(q, bio, old_dev, bio->bi_sector, 
-					    maxsector);
+					    old_sector);
 
 		blk_add_trace_bio(q, bio, BLK_TA_QUEUE);
 
-		maxsector = bio->bi_sector;
+		old_sector = bio->bi_sector;
 		old_dev = bio->bi_bdev->bd_dev;
 
+		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 - partitions are not
+				 * checked to make sure they are within the size
+				 * of the whole device.
+				 */
+				handle_bad_sector(bio);
+				goto end_io;
+			}
+		}
+
 		ret = q->make_request_fn(q, bio);
 	} while (ret);
 }
@@ -3531,6 +3580,7 @@ void blk_rq_bio_prep(request_queue_t *q, struct request *rq, struct bio *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;
 }
@@ -3611,25 +3661,22 @@ EXPORT_SYMBOL(put_io_context);
 /* Called by the exitting task */
 void exit_io_context(void)
 {
-	unsigned long flags;
 	struct io_context *ioc;
 	struct cfq_io_context *cic;
 
-	local_irq_save(flags);
 	task_lock(current);
 	ioc = current->io_context;
 	current->io_context = NULL;
-	ioc->task = NULL;
 	task_unlock(current);
-	local_irq_restore(flags);
 
+	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);
 }
 
@@ -3641,7 +3688,7 @@ void exit_io_context(void)
  * 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)
+static struct io_context *current_io_context(gfp_t gfp_flags, int node)
 {
 	struct task_struct *tsk = current;
 	struct io_context *ret;
@@ -3650,7 +3697,7 @@ struct io_context *current_io_context(gfp_t gfp_flags)
 	if (likely(ret))
 		return ret;
 
-	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
+	ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node);
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
 		ret->task = current;
@@ -3674,10 +3721,10 @@ EXPORT_SYMBOL(current_io_context);
  *
  * 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)
+struct io_context *get_io_context(gfp_t gfp_flags, int node)
 {
 	struct io_context *ret;
-	ret = current_io_context(gfp_flags);
+	ret = current_io_context(gfp_flags, node);
 	if (likely(ret))
 		atomic_inc(&ret->refcount);
 	return ret;
@@ -3750,14 +3797,14 @@ queue_requests_store(struct request_queue *q, const char *page, size_t count)
 	blk_queue_congestion_threshold(q);
 
 	if (rl->count[READ] >= queue_congestion_on_threshold(q))
-		set_queue_congested(q, READ);
+		blk_set_queue_congested(q, READ);
 	else if (rl->count[READ] < queue_congestion_off_threshold(q))
-		clear_queue_congested(q, READ);
+		blk_clear_queue_congested(q, READ);
 
 	if (rl->count[WRITE] >= queue_congestion_on_threshold(q))
-		set_queue_congested(q, WRITE);
+		blk_set_queue_congested(q, WRITE);
 	else if (rl->count[WRITE] < queue_congestion_off_threshold(q))
-		clear_queue_congested(q, WRITE);
+		blk_clear_queue_congested(q, WRITE);
 
 	if (rl->count[READ] >= q->nr_requests) {
 		blk_set_queue_full(q, READ);
@@ -3790,9 +3837,6 @@ queue_ra_store(struct request_queue *q, const char *page, size_t count)
 	ssize_t ret = queue_var_store(&ra_kb, page, count);
 
 	spin_lock_irq(q->queue_lock);
-	if (ra_kb > (q->max_sectors >> 1))
-		ra_kb = (q->max_sectors >> 1);
-
 	q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10);
 	spin_unlock_irq(q->queue_lock);