Merge tag 'for-linus-20180623' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - Further timeout fixes. We aren't quite there yet, so expect another
   round of fixes for that to completely close some of the IRQ vs
   completion races. (Christoph/Bart)

 - Set of NVMe fixes from the usual suspects, mostly error handling

 - Two off-by-one fixes (Dan)

 - Another bdi race fix (Jan)

 - Fix nbd reconfigure with NBD_DISCONNECT_ON_CLOSE (Doron)

* tag 'for-linus-20180623' of git://git.kernel.dk/linux-block:
  blk-mq: Fix timeout handling in case the timeout handler returns BLK_EH_DONE
  bdi: Fix another oops in wb_workfn()
  lightnvm: Remove depends on HAS_DMA in case of platform dependency
  nvme-pci: limit max IO size and segments to avoid high order allocations
  nvme-pci: move nvme_kill_queues to nvme_remove_dead_ctrl
  nvme-fc: release io queues to allow fast fail
  nbd: Add the nbd NBD_DISCONNECT_ON_CLOSE config flag.
  block: sed-opal: Fix a couple off by one bugs
  blk-mq-debugfs: Off by one in blk_mq_rq_state_name()
  nvmet: reset keep alive timer in controller enable
  nvme-rdma: don't override opts->queue_size
  nvme-rdma: Fix command completion race at error recovery
  nvme-rdma: fix possible free of a non-allocated async event buffer
  nvme-rdma: fix possible double free condition when failing to create a controller
  Revert "block: Add warning for bi_next not NULL in bio_endio()"
  block: fix timeout changes for legacy request drivers

block/bio.c

@@ -1807,9 +1807,6 @@ void bio_endio(struct bio *bio)
 	if (!bio_integrity_endio(bio))
 		return;
 
-	if (WARN_ONCE(bio->bi_next, "driver left bi_next not NULL"))
-		bio->bi_next = NULL;
-
 	/*
 	 * Need to have a real endio function for chained bios, otherwise
 	 * various corner cases will break (like stacking block devices that

block/blk-core.c

@@ -273,10 +273,6 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 	bio_advance(bio, nbytes);
 
 	/* don't actually finish bio if it's part of flush sequence */
-	/*
-	 * XXX this code looks suspicious - it's not consistent with advancing
-	 * req->bio in caller
-	 */
 	if (bio->bi_iter.bi_size == 0 && !(rq->rq_flags & RQF_FLUSH_SEQ))
 		bio_endio(bio);
 }
@@ -3081,10 +3077,8 @@ bool blk_update_request(struct request *req, blk_status_t error,
 		struct bio *bio = req->bio;
 		unsigned bio_bytes = min(bio->bi_iter.bi_size, nr_bytes);
 
-		if (bio_bytes == bio->bi_iter.bi_size) {
+		if (bio_bytes == bio->bi_iter.bi_size)
 			req->bio = bio->bi_next;
-			bio->bi_next = NULL;
-		}
 
 		/* Completion has already been traced */
 		bio_clear_flag(bio, BIO_TRACE_COMPLETION);

block/blk-mq-debugfs.c

@@ -356,7 +356,7 @@ static const char *const blk_mq_rq_state_name_array[] = {
 
 static const char *blk_mq_rq_state_name(enum mq_rq_state rq_state)
 {
-	if (WARN_ON_ONCE((unsigned int)rq_state >
+	if (WARN_ON_ONCE((unsigned int)rq_state >=
 			 ARRAY_SIZE(blk_mq_rq_state_name_array)))
 		return "(?)";
 	return blk_mq_rq_state_name_array[rq_state];

block/blk-mq.c

@@ -781,7 +781,6 @@ static void blk_mq_rq_timed_out(struct request *req, bool reserved)
 		WARN_ON_ONCE(ret != BLK_EH_RESET_TIMER);
 	}
 
-	req->rq_flags &= ~RQF_TIMED_OUT;
 	blk_add_timer(req);
 }
 

block/blk-softirq.c

@@ -144,6 +144,7 @@ void __blk_complete_request(struct request *req)
 
 	local_irq_restore(flags);
 }
+EXPORT_SYMBOL(__blk_complete_request);
 
 /**
  * blk_complete_request - end I/O on a request

block/blk-timeout.c

@@ -210,6 +210,7 @@ void blk_add_timer(struct request *req)
 	if (!req->timeout)
 		req->timeout = q->rq_timeout;
 
+	req->rq_flags &= ~RQF_TIMED_OUT;
 	blk_rq_set_deadline(req, jiffies + req->timeout);
 
 	/*

block/sed-opal.c

@@ -877,7 +877,7 @@ static size_t response_get_string(const struct parsed_resp *resp, int n,
 		return 0;
 	}
 
-	if (n > resp->num) {
+	if (n >= resp->num) {
 		pr_debug("Response has %d tokens. Can't access %d\n",
 			 resp->num, n);
 		return 0;
@@ -916,7 +916,7 @@ static u64 response_get_u64(const struct parsed_resp *resp, int n)
 		return 0;
 	}
 
-	if (n > resp->num) {
+	if (n >= resp->num) {
 		pr_debug("Response has %d tokens. Can't access %d\n",
 			 resp->num, n);
 		return 0;

drivers/block/nbd.c

@@ -76,6 +76,7 @@ struct link_dead_args {
 #define NBD_HAS_CONFIG_REF		4
 #define NBD_BOUND			5
 #define NBD_DESTROY_ON_DISCONNECT	6
+#define NBD_DISCONNECT_ON_CLOSE 	7
 
 struct nbd_config {
 	u32 flags;
@@ -138,6 +139,7 @@ static void nbd_config_put(struct nbd_device *nbd);
 static void nbd_connect_reply(struct genl_info *info, int index);
 static int nbd_genl_status(struct sk_buff *skb, struct genl_info *info);
 static void nbd_dead_link_work(struct work_struct *work);
+static void nbd_disconnect_and_put(struct nbd_device *nbd);
 
 static inline struct device *nbd_to_dev(struct nbd_device *nbd)
 {
@@ -1305,6 +1307,12 @@ static int nbd_open(struct block_device *bdev, fmode_t mode)
 static void nbd_release(struct gendisk *disk, fmode_t mode)
 {
 	struct nbd_device *nbd = disk->private_data;
+	struct block_device *bdev = bdget_disk(disk, 0);
+
+	if (test_bit(NBD_DISCONNECT_ON_CLOSE, &nbd->config->runtime_flags) &&
+			bdev->bd_openers == 0)
+		nbd_disconnect_and_put(nbd);
+
 	nbd_config_put(nbd);
 	nbd_put(nbd);
 }
@@ -1705,6 +1713,10 @@ static int nbd_genl_connect(struct sk_buff *skb, struct genl_info *info)
 				&config->runtime_flags);
 			put_dev = true;
 		}
+		if (flags & NBD_CFLAG_DISCONNECT_ON_CLOSE) {
+			set_bit(NBD_DISCONNECT_ON_CLOSE,
+				&config->runtime_flags);
+		}
 	}
 
 	if (info->attrs[NBD_ATTR_SOCKETS]) {
@@ -1749,6 +1761,17 @@ static int nbd_genl_connect(struct sk_buff *skb, struct genl_info *info)
 	return ret;
 }
 
+static void nbd_disconnect_and_put(struct nbd_device *nbd)
+{
+	mutex_lock(&nbd->config_lock);
+	nbd_disconnect(nbd);
+	nbd_clear_sock(nbd);
+	mutex_unlock(&nbd->config_lock);
+	if (test_and_clear_bit(NBD_HAS_CONFIG_REF,
+			       &nbd->config->runtime_flags))
+		nbd_config_put(nbd);
+}
+
 static int nbd_genl_disconnect(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nbd_device *nbd;
@@ -1781,13 +1804,7 @@ static int nbd_genl_disconnect(struct sk_buff *skb, struct genl_info *info)
 		nbd_put(nbd);
 		return 0;
 	}
-	mutex_lock(&nbd->config_lock);
-	nbd_disconnect(nbd);
-	nbd_clear_sock(nbd);
-	mutex_unlock(&nbd->config_lock);
-	if (test_and_clear_bit(NBD_HAS_CONFIG_REF,
-			       &nbd->config->runtime_flags))
-		nbd_config_put(nbd);
+	nbd_disconnect_and_put(nbd);
 	nbd_config_put(nbd);
 	nbd_put(nbd);
 	return 0;
@@ -1798,7 +1815,7 @@ static int nbd_genl_reconfigure(struct sk_buff *skb, struct genl_info *info)
 	struct nbd_device *nbd = NULL;
 	struct nbd_config *config;
 	int index;
-	int ret = -EINVAL;
+	int ret = 0;
 	bool put_dev = false;
 
 	if (!netlink_capable(skb, CAP_SYS_ADMIN))
@@ -1838,6 +1855,7 @@ static int nbd_genl_reconfigure(struct sk_buff *skb, struct genl_info *info)
 	    !nbd->task_recv) {
 		dev_err(nbd_to_dev(nbd),
 			"not configured, cannot reconfigure\n");
+		ret = -EINVAL;
 		goto out;
 	}
 
@@ -1862,6 +1880,14 @@ static int nbd_genl_reconfigure(struct sk_buff *skb, struct genl_info *info)
 					       &config->runtime_flags))
 				refcount_inc(&nbd->refs);
 		}
+
+		if (flags & NBD_CFLAG_DISCONNECT_ON_CLOSE) {
+			set_bit(NBD_DISCONNECT_ON_CLOSE,
+					&config->runtime_flags);
+		} else {
+			clear_bit(NBD_DISCONNECT_ON_CLOSE,
+					&config->runtime_flags);
+		}
 	}
 
 	if (info->attrs[NBD_ATTR_SOCKETS]) {

drivers/block/null_blk.c

@@ -1365,7 +1365,7 @@ static blk_qc_t null_queue_bio(struct request_queue *q, struct bio *bio)
 static enum blk_eh_timer_return null_rq_timed_out_fn(struct request *rq)
 {
 	pr_info("null: rq %p timed out\n", rq);
-	blk_mq_complete_request(rq);
+	__blk_complete_request(rq);
 	return BLK_EH_DONE;
 }
 

drivers/lightnvm/Kconfig

@@ -4,7 +4,7 @@
 
 menuconfig NVM
 	bool "Open-Channel SSD target support"
-	depends on BLOCK && HAS_DMA && PCI
+	depends on BLOCK && PCI
 	select BLK_DEV_NVME
 	help
 	  Say Y here to get to enable Open-channel SSDs.

drivers/nvme/host/core.c

@@ -1808,6 +1808,7 @@ static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
 		u32 max_segments =
 			(ctrl->max_hw_sectors / (ctrl->page_size >> 9)) + 1;
 
+		max_segments = min_not_zero(max_segments, ctrl->max_segments);
 		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
 		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
 	}

drivers/nvme/host/fc.c

@@ -2790,6 +2790,9 @@ nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl)
 	/* re-enable the admin_q so anything new can fast fail */
 	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 
+	/* resume the io queues so that things will fast fail */
+	nvme_start_queues(&ctrl->ctrl);
+
 	nvme_fc_ctlr_inactive_on_rport(ctrl);
 }
 
@@ -2804,9 +2807,6 @@ nvme_fc_delete_ctrl(struct nvme_ctrl *nctrl)
 	 * waiting for io to terminate
 	 */
 	nvme_fc_delete_association(ctrl);
-
-	/* resume the io queues so that things will fast fail */
-	nvme_start_queues(nctrl);
 }
 
 static void

drivers/nvme/host/nvme.h

@@ -170,6 +170,7 @@ struct nvme_ctrl {
 	u64 cap;
 	u32 page_size;
 	u32 max_hw_sectors;
+	u32 max_segments;
 	u16 oncs;
 	u16 oacs;
 	u16 nssa;

drivers/nvme/host/pci.c

@@ -38,6 +38,13 @@
 
 #define SGES_PER_PAGE	(PAGE_SIZE / sizeof(struct nvme_sgl_desc))
 
+/*
+ * These can be higher, but we need to ensure that any command doesn't
+ * require an sg allocation that needs more than a page of data.
+ */
+#define NVME_MAX_KB_SZ	4096
+#define NVME_MAX_SEGS	127
+
 static int use_threaded_interrupts;
 module_param(use_threaded_interrupts, int, 0);
 
@@ -100,6 +107,8 @@ struct nvme_dev {
 	struct nvme_ctrl ctrl;
 	struct completion ioq_wait;
 
+	mempool_t *iod_mempool;
+
 	/* shadow doorbell buffer support: */
 	u32 *dbbuf_dbs;
 	dma_addr_t dbbuf_dbs_dma_addr;
@@ -477,10 +486,7 @@ static blk_status_t nvme_init_iod(struct request *rq, struct nvme_dev *dev)
 	iod->use_sgl = nvme_pci_use_sgls(dev, rq);
 
 	if (nseg > NVME_INT_PAGES || size > NVME_INT_BYTES(dev)) {
-		size_t alloc_size = nvme_pci_iod_alloc_size(dev, size, nseg,
-				iod->use_sgl);
-
-		iod->sg = kmalloc(alloc_size, GFP_ATOMIC);
+		iod->sg = mempool_alloc(dev->iod_mempool, GFP_ATOMIC);
 		if (!iod->sg)
 			return BLK_STS_RESOURCE;
 	} else {
@@ -526,7 +532,7 @@ static void nvme_free_iod(struct nvme_dev *dev, struct request *req)
 	}
 
 	if (iod->sg != iod->inline_sg)
-		kfree(iod->sg);
+		mempool_free(iod->sg, dev->iod_mempool);
 }
 
 #ifdef CONFIG_BLK_DEV_INTEGRITY
@@ -2280,6 +2286,7 @@ static void nvme_pci_free_ctrl(struct nvme_ctrl *ctrl)
 		blk_put_queue(dev->ctrl.admin_q);
 	kfree(dev->queues);
 	free_opal_dev(dev->ctrl.opal_dev);
+	mempool_destroy(dev->iod_mempool);
 	kfree(dev);
 }
 
@@ -2289,6 +2296,7 @@ static void nvme_remove_dead_ctrl(struct nvme_dev *dev, int status)
 
 	nvme_get_ctrl(&dev->ctrl);
 	nvme_dev_disable(dev, false);
+	nvme_kill_queues(&dev->ctrl);
 	if (!queue_work(nvme_wq, &dev->remove_work))
 		nvme_put_ctrl(&dev->ctrl);
 }
@@ -2333,6 +2341,13 @@ static void nvme_reset_work(struct work_struct *work)
 	if (result)
 		goto out;
 
+	/*
+	 * Limit the max command size to prevent iod->sg allocations going
+	 * over a single page.
+	 */
+	dev->ctrl.max_hw_sectors = NVME_MAX_KB_SZ << 1;
+	dev->ctrl.max_segments = NVME_MAX_SEGS;
+
 	result = nvme_init_identify(&dev->ctrl);
 	if (result)
 		goto out;
@@ -2405,7 +2420,6 @@ static void nvme_remove_dead_ctrl_work(struct work_struct *work)
 	struct nvme_dev *dev = container_of(work, struct nvme_dev, remove_work);
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
 
-	nvme_kill_queues(&dev->ctrl);
 	if (pci_get_drvdata(pdev))
 		device_release_driver(&pdev->dev);
 	nvme_put_ctrl(&dev->ctrl);
@@ -2509,6 +2523,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	int node, result = -ENOMEM;
 	struct nvme_dev *dev;
 	unsigned long quirks = id->driver_data;
+	size_t alloc_size;
 
 	node = dev_to_node(&pdev->dev);
 	if (node == NUMA_NO_NODE)
@@ -2546,6 +2561,23 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (result)
 		goto release_pools;
 
+	/*
+	 * Double check that our mempool alloc size will cover the biggest
+	 * command we support.
+	 */
+	alloc_size = nvme_pci_iod_alloc_size(dev, NVME_MAX_KB_SZ,
+						NVME_MAX_SEGS, true);
+	WARN_ON_ONCE(alloc_size > PAGE_SIZE);
+
+	dev->iod_mempool = mempool_create_node(1, mempool_kmalloc,
+						mempool_kfree,
+						(void *) alloc_size,
+						GFP_KERNEL, node);
+	if (!dev->iod_mempool) {
+		result = -ENOMEM;
+		goto release_pools;
+	}
+
 	dev_info(dev->ctrl.device, "pci function %s\n", dev_name(&pdev->dev));
 
 	nvme_get_ctrl(&dev->ctrl);