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

Pull block layer fixes from Jens Axboe:
 "A round of fixes/updates for the current series.

  This looks a little bigger than it is, but that's mainly because we
  pushed the lightnvm enabled null_blk change out of the merge window so
  it could be updated a bit.  The rest of the volume is also mostly
  lightnvm.  In particular:

   - Lightnvm.  Various fixes, additions, updates from Matias and
     Javier, as well as from Wenwei Tao.

   - NVMe:
        - Fix for potential arithmetic overflow from Keith.
        - Also from Keith, ensure that we reap pending completions from
          a completion queue before deleting it.  Fixes kernel crashes
          when resetting a device with IO pending.
        - Various little lightnvm related tweaks from Matias.

   - Fixup flushes to go through the IO scheduler, for the cases where a
     flush is not required.  Fixes a case in CFQ where we would be
     idling and not see this request, hence not break the idling.  From
     Jan Kara.

   - Use list_{first,prev,next} in elevator.c for cleaner code.  From
     Gelian Tang.

   - Fix for a warning trigger on btrfs and raid on single queue blk-mq
     devices, where we would flush plug callbacks with preemption
     disabled.  From me.

   - A mac partition validation fix from Kees Cook.

   - Two merge fixes from Ming, marked stable.  A third part is adding a
     new warning so we'll notice this quicker in the future, if we screw
     up the accounting.

   - Cleanup of thread name/creation in mtip32xx from Rasmus Villemoes"

* 'for-linus' of git://git.kernel.dk/linux-block: (32 commits)
  blk-merge: warn if figured out segment number is bigger than nr_phys_segments
  blk-merge: fix blk_bio_segment_split
  block: fix segment split
  blk-mq: fix calling unplug callbacks with preempt disabled
  mac: validate mac_partition is within sector
  mtip32xx: use formatting capability of kthread_create_on_node
  NVMe: reap completion entries when deleting queue
  lightnvm: add free and bad lun info to show luns
  lightnvm: keep track of block counts
  nvme: lightnvm: use admin queues for admin cmds
  lightnvm: missing free on init error
  lightnvm: wrong return value and redundant free
  null_blk: do not del gendisk with lightnvm
  null_blk: use device addressing mode
  null_blk: use ppa_cache pool
  NVMe: Fix possible arithmetic overflow for max segments
  blk-flush: Queue through IO scheduler when flush not required
  null_blk: register as a LightNVM device
  elevator: use list_{first,prev,next}_entry
  lightnvm: cleanup queue before target removal
  ...

Documentation/block/null_blk.txt

@@ -70,3 +70,6 @@ use_per_node_hctx=[0/1]: Default: 0
      parameter.
   1: The multi-queue block layer is instantiated with a hardware dispatch
      queue for each CPU node in the system.
+
+use_lightnvm=[0/1]: Default: 0
+  Register device with LightNVM. Requires blk-mq to be used.

MAINTAINERS

@@ -6366,6 +6366,7 @@ F:	arch/*/include/asm/pmem.h
 LIGHTNVM PLATFORM SUPPORT
 M:	Matias Bjorling <mb@lightnvm.io>
 W:	http://github/OpenChannelSSD
+L:	linux-block@vger.kernel.org
 S:	Maintained
 F:	drivers/lightnvm/
 F:	include/linux/lightnvm.h

block/blk-flush.c

@@ -422,7 +422,7 @@ void blk_insert_flush(struct request *rq)
 		if (q->mq_ops) {
 			blk_mq_insert_request(rq, false, false, true);
 		} else
-			list_add_tail(&rq->queuelist, &q->queue_head);
+			q->elevator->type->ops.elevator_add_req_fn(q, rq);
 		return;
 	}
 

block/blk-merge.c

@@ -76,6 +76,9 @@ static struct bio *blk_bio_segment_split(struct request_queue *q,
 	struct bio_vec bv, bvprv, *bvprvp = NULL;
 	struct bvec_iter iter;
 	unsigned seg_size = 0, nsegs = 0, sectors = 0;
+	unsigned front_seg_size = bio->bi_seg_front_size;
+	bool do_split = true;
+	struct bio *new = NULL;
 
 	bio_for_each_segment(bv, bio, iter) {
 		if (sectors + (bv.bv_len >> 9) > queue_max_sectors(q))
@@ -98,7 +101,7 @@ static struct bio *blk_bio_segment_split(struct request_queue *q,
 
 			seg_size += bv.bv_len;
 			bvprv = bv;
-			bvprvp = &bv;
+			bvprvp = &bvprv;
 			sectors += bv.bv_len >> 9;
 			continue;
 		}
@@ -108,16 +111,29 @@ new_segment:
 
 		nsegs++;
 		bvprv = bv;
-		bvprvp = &bv;
+		bvprvp = &bvprv;
 		seg_size = bv.bv_len;
 		sectors += bv.bv_len >> 9;
+
+		if (nsegs == 1 && seg_size > front_seg_size)
+			front_seg_size = seg_size;
 	}
 
-	*segs = nsegs;
-	return NULL;
+	do_split = false;
 split:
 	*segs = nsegs;
-	return bio_split(bio, sectors, GFP_NOIO, bs);
+
+	if (do_split) {
+		new = bio_split(bio, sectors, GFP_NOIO, bs);
+		if (new)
+			bio = new;
+	}
+
+	bio->bi_seg_front_size = front_seg_size;
+	if (seg_size > bio->bi_seg_back_size)
+		bio->bi_seg_back_size = seg_size;
+
+	return do_split ? new : NULL;
 }
 
 void blk_queue_split(struct request_queue *q, struct bio **bio,
@@ -412,6 +428,12 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
 	if (sg)
 		sg_mark_end(sg);
 
+	/*
+	 * Something must have been wrong if the figured number of
+	 * segment is bigger than number of req's physical segments
+	 */
+	WARN_ON(nsegs > rq->nr_phys_segments);
+
 	return nsegs;
 }
 EXPORT_SYMBOL(blk_rq_map_sg);

block/blk-mq.c

@@ -1291,15 +1291,16 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
 		blk_mq_bio_to_request(rq, bio);
 
 		/*
-		 * we do limited pluging. If bio can be merged, do merge.
+		 * We do limited pluging. If the bio can be merged, do that.
 		 * Otherwise the existing request in the plug list will be
 		 * issued. So the plug list will have one request at most
 		 */
 		if (plug) {
 			/*
 			 * The plug list might get flushed before this. If that
-			 * happens, same_queue_rq is invalid and plug list is empty
-			 **/
+			 * happens, same_queue_rq is invalid and plug list is
+			 * empty
+			 */
 			if (same_queue_rq && !list_empty(&plug->mq_list)) {
 				old_rq = same_queue_rq;
 				list_del_init(&old_rq->queuelist);
@@ -1380,12 +1381,15 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio)
 		blk_mq_bio_to_request(rq, bio);
 		if (!request_count)
 			trace_block_plug(q);
-		else if (request_count >= BLK_MAX_REQUEST_COUNT) {
+
+		blk_mq_put_ctx(data.ctx);
+
+		if (request_count >= BLK_MAX_REQUEST_COUNT) {
 			blk_flush_plug_list(plug, false);
 			trace_block_plug(q);
 		}
+
 		list_add_tail(&rq->queuelist, &plug->mq_list);
-		blk_mq_put_ctx(data.ctx);
 		return cookie;
 	}
 

block/noop-iosched.c

@@ -21,10 +21,10 @@ static void noop_merged_requests(struct request_queue *q, struct request *rq,
 static int noop_dispatch(struct request_queue *q, int force)
 {
 	struct noop_data *nd = q->elevator->elevator_data;
+	struct request *rq;
 
-	if (!list_empty(&nd->queue)) {
-		struct request *rq;
-		rq = list_entry(nd->queue.next, struct request, queuelist);
+	rq = list_first_entry_or_null(&nd->queue, struct request, queuelist);
+	if (rq) {
 		list_del_init(&rq->queuelist);
 		elv_dispatch_sort(q, rq);
 		return 1;
@@ -46,7 +46,7 @@ noop_former_request(struct request_queue *q, struct request *rq)
 
 	if (rq->queuelist.prev == &nd->queue)
 		return NULL;
-	return list_entry(rq->queuelist.prev, struct request, queuelist);
+	return list_prev_entry(rq, queuelist);
 }
 
 static struct request *
@@ -56,7 +56,7 @@ noop_latter_request(struct request_queue *q, struct request *rq)
 
 	if (rq->queuelist.next == &nd->queue)
 		return NULL;
-	return list_entry(rq->queuelist.next, struct request, queuelist);
+	return list_next_entry(rq, queuelist);
 }
 
 static int noop_init_queue(struct request_queue *q, struct elevator_type *e)

block/partitions/mac.c

@@ -32,7 +32,7 @@ int mac_partition(struct parsed_partitions *state)
 	Sector sect;
 	unsigned char *data;
 	int slot, blocks_in_map;
-	unsigned secsize;
+	unsigned secsize, datasize, partoffset;
 #ifdef CONFIG_PPC_PMAC
 	int found_root = 0;
 	int found_root_goodness = 0;
@@ -50,10 +50,14 @@ int mac_partition(struct parsed_partitions *state)
 	}
 	secsize = be16_to_cpu(md->block_size);
 	put_dev_sector(sect);
-	data = read_part_sector(state, secsize/512, &sect);
+	datasize = round_down(secsize, 512);
+	data = read_part_sector(state, datasize / 512, &sect);
 	if (!data)
 		return -1;
-	part = (struct mac_partition *) (data + secsize%512);
+	partoffset = secsize % 512;
+	if (partoffset + sizeof(*part) > datasize)
+		return -1;
+	part = (struct mac_partition *) (data + partoffset);
 	if (be16_to_cpu(part->signature) != MAC_PARTITION_MAGIC) {
 		put_dev_sector(sect);
 		return 0;		/* not a MacOS disk */

drivers/Makefile

@@ -63,6 +63,7 @@ obj-$(CONFIG_FB_I810)           += video/fbdev/i810/
 obj-$(CONFIG_FB_INTEL)          += video/fbdev/intelfb/
 
 obj-$(CONFIG_PARPORT)		+= parport/
+obj-$(CONFIG_NVM)		+= lightnvm/
 obj-y				+= base/ block/ misc/ mfd/ nfc/
 obj-$(CONFIG_LIBNVDIMM)		+= nvdimm/
 obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf/
@@ -70,7 +71,6 @@ obj-$(CONFIG_NUBUS)		+= nubus/
 obj-y				+= macintosh/
 obj-$(CONFIG_IDE)		+= ide/
 obj-$(CONFIG_SCSI)		+= scsi/
-obj-$(CONFIG_NVM)		+= lightnvm/
 obj-y				+= nvme/
 obj-$(CONFIG_ATA)		+= ata/
 obj-$(CONFIG_TARGET_CORE)	+= target/

drivers/block/mtip32xx/mtip32xx.c

@@ -3810,7 +3810,6 @@ static int mtip_block_initialize(struct driver_data *dd)
 	sector_t capacity;
 	unsigned int index = 0;
 	struct kobject *kobj;
-	unsigned char thd_name[16];
 
 	if (dd->disk)
 		goto skip_create_disk; /* hw init done, before rebuild */
@@ -3958,10 +3957,9 @@ skip_create_disk:
 	}
 
 start_service_thread:
-	sprintf(thd_name, "mtip_svc_thd_%02d", index);
 	dd->mtip_svc_handler = kthread_create_on_node(mtip_service_thread,
-						dd, dd->numa_node, "%s",
-						thd_name);
+						dd, dd->numa_node,
+						"mtip_svc_thd_%02d", index);
 
 	if (IS_ERR(dd->mtip_svc_handler)) {
 		dev_err(&dd->pdev->dev, "service thread failed to start\n");

drivers/block/null_blk.c

@@ -8,6 +8,7 @@
 #include <linux/slab.h>
 #include <linux/blk-mq.h>
 #include <linux/hrtimer.h>
+#include <linux/lightnvm.h>
 
 struct nullb_cmd {
 	struct list_head list;
@@ -39,12 +40,14 @@ struct nullb {
 
 	struct nullb_queue *queues;
 	unsigned int nr_queues;
+	char disk_name[DISK_NAME_LEN];
 };
 
 static LIST_HEAD(nullb_list);
 static struct mutex lock;
 static int null_major;
 static int nullb_indexes;
+static struct kmem_cache *ppa_cache;
 
 struct completion_queue {
 	struct llist_head list;
@@ -119,6 +122,10 @@ static int nr_devices = 2;
 module_param(nr_devices, int, S_IRUGO);
 MODULE_PARM_DESC(nr_devices, "Number of devices to register");
 
+static bool use_lightnvm;
+module_param(use_lightnvm, bool, S_IRUGO);
+MODULE_PARM_DESC(use_lightnvm, "Register as a LightNVM device");
+
 static int irqmode = NULL_IRQ_SOFTIRQ;
 
 static int null_set_irqmode(const char *str, const struct kernel_param *kp)
@@ -427,15 +434,156 @@ static void null_del_dev(struct nullb *nullb)
 {
 	list_del_init(&nullb->list);
 
-	del_gendisk(nullb->disk);
+	if (use_lightnvm)
+		nvm_unregister(nullb->disk_name);
+	else
+		del_gendisk(nullb->disk);
 	blk_cleanup_queue(nullb->q);
 	if (queue_mode == NULL_Q_MQ)
 		blk_mq_free_tag_set(&nullb->tag_set);
-	put_disk(nullb->disk);
+	if (!use_lightnvm)
+		put_disk(nullb->disk);
 	cleanup_queues(nullb);
 	kfree(nullb);
 }
 
+#ifdef CONFIG_NVM
+
+static void null_lnvm_end_io(struct request *rq, int error)
+{
+	struct nvm_rq *rqd = rq->end_io_data;
+	struct nvm_dev *dev = rqd->dev;
+
+	dev->mt->end_io(rqd, error);
+
+	blk_put_request(rq);
+}
+
+static int null_lnvm_submit_io(struct request_queue *q, struct nvm_rq *rqd)
+{
+	struct request *rq;
+	struct bio *bio = rqd->bio;
+
+	rq = blk_mq_alloc_request(q, bio_rw(bio), GFP_KERNEL, 0);
+	if (IS_ERR(rq))
+		return -ENOMEM;
+
+	rq->cmd_type = REQ_TYPE_DRV_PRIV;
+	rq->__sector = bio->bi_iter.bi_sector;
+	rq->ioprio = bio_prio(bio);
+
+	if (bio_has_data(bio))
+		rq->nr_phys_segments = bio_phys_segments(q, bio);
+
+	rq->__data_len = bio->bi_iter.bi_size;
+	rq->bio = rq->biotail = bio;
+
+	rq->end_io_data = rqd;
+
+	blk_execute_rq_nowait(q, NULL, rq, 0, null_lnvm_end_io);
+
+	return 0;
+}
+
+static int null_lnvm_id(struct request_queue *q, struct nvm_id *id)
+{
+	sector_t size = gb * 1024 * 1024 * 1024ULL;
+	sector_t blksize;
+	struct nvm_id_group *grp;
+
+	id->ver_id = 0x1;
+	id->vmnt = 0;
+	id->cgrps = 1;
+	id->cap = 0x3;
+	id->dom = 0x1;
+
+	id->ppaf.blk_offset = 0;
+	id->ppaf.blk_len = 16;
+	id->ppaf.pg_offset = 16;
+	id->ppaf.pg_len = 16;
+	id->ppaf.sect_offset = 32;
+	id->ppaf.sect_len = 8;
+	id->ppaf.pln_offset = 40;
+	id->ppaf.pln_len = 8;
+	id->ppaf.lun_offset = 48;
+	id->ppaf.lun_len = 8;
+	id->ppaf.ch_offset = 56;
+	id->ppaf.ch_len = 8;
+
+	do_div(size, bs); /* convert size to pages */
+	do_div(size, 256); /* concert size to pgs pr blk */
+	grp = &id->groups[0];
+	grp->mtype = 0;
+	grp->fmtype = 0;
+	grp->num_ch = 1;
+	grp->num_pg = 256;
+	blksize = size;
+	do_div(size, (1 << 16));
+	grp->num_lun = size + 1;
+	do_div(blksize, grp->num_lun);
+	grp->num_blk = blksize;
+	grp->num_pln = 1;
+
+	grp->fpg_sz = bs;
+	grp->csecs = bs;
+	grp->trdt = 25000;
+	grp->trdm = 25000;
+	grp->tprt = 500000;
+	grp->tprm = 500000;
+	grp->tbet = 1500000;
+	grp->tbem = 1500000;
+	grp->mpos = 0x010101; /* single plane rwe */
+	grp->cpar = hw_queue_depth;
+
+	return 0;
+}
+
+static void *null_lnvm_create_dma_pool(struct request_queue *q, char *name)
+{
+	mempool_t *virtmem_pool;
+
+	virtmem_pool = mempool_create_slab_pool(64, ppa_cache);
+	if (!virtmem_pool) {
+		pr_err("null_blk: Unable to create virtual memory pool\n");
+		return NULL;
+	}
+
+	return virtmem_pool;
+}
+
+static void null_lnvm_destroy_dma_pool(void *pool)
+{
+	mempool_destroy(pool);
+}
+
+static void *null_lnvm_dev_dma_alloc(struct request_queue *q, void *pool,
+				gfp_t mem_flags, dma_addr_t *dma_handler)
+{
+	return mempool_alloc(pool, mem_flags);
+}
+
+static void null_lnvm_dev_dma_free(void *pool, void *entry,
+							dma_addr_t dma_handler)
+{
+	mempool_free(entry, pool);
+}
+
+static struct nvm_dev_ops null_lnvm_dev_ops = {
+	.identity		= null_lnvm_id,
+	.submit_io		= null_lnvm_submit_io,
+
+	.create_dma_pool	= null_lnvm_create_dma_pool,
+	.destroy_dma_pool	= null_lnvm_destroy_dma_pool,
+	.dev_dma_alloc		= null_lnvm_dev_dma_alloc,
+	.dev_dma_free		= null_lnvm_dev_dma_free,
+
+	/* Simulate nvme protocol restriction */
+	.max_phys_sect		= 64,
+};
+#else
+static struct nvm_dev_ops null_lnvm_dev_ops;
+#endif /* CONFIG_NVM */
+
 static int null_open(struct block_device *bdev, fmode_t mode)
 {
 	return 0;
@@ -575,11 +723,6 @@ static int null_add_dev(void)
 	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, nullb->q);
 	queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, nullb->q);
 
-	disk = nullb->disk = alloc_disk_node(1, home_node);
-	if (!disk) {
-		rv = -ENOMEM;
-		goto out_cleanup_blk_queue;
-	}
 
 	mutex_lock(&lock);
 	list_add_tail(&nullb->list, &nullb_list);
@@ -589,6 +732,21 @@ static int null_add_dev(void)
 	blk_queue_logical_block_size(nullb->q, bs);
 	blk_queue_physical_block_size(nullb->q, bs);
 
+	sprintf(nullb->disk_name, "nullb%d", nullb->index);
+
+	if (use_lightnvm) {
+		rv = nvm_register(nullb->q, nullb->disk_name,
+							&null_lnvm_dev_ops);
+		if (rv)
+			goto out_cleanup_blk_queue;
+		goto done;
+	}
+
+	disk = nullb->disk = alloc_disk_node(1, home_node);
+	if (!disk) {
+		rv = -ENOMEM;
+		goto out_cleanup_lightnvm;
+	}
 	size = gb * 1024 * 1024 * 1024ULL;
 	set_capacity(disk, size >> 9);
 
@@ -598,10 +756,15 @@ static int null_add_dev(void)
 	disk->fops		= &null_fops;
 	disk->private_data	= nullb;
 	disk->queue		= nullb->q;
-	sprintf(disk->disk_name, "nullb%d", nullb->index);
+	strncpy(disk->disk_name, nullb->disk_name, DISK_NAME_LEN);
+
 	add_disk(disk);
+done:
 	return 0;
 
+out_cleanup_lightnvm:
+	if (use_lightnvm)
+		nvm_unregister(nullb->disk_name);
 out_cleanup_blk_queue:
 	blk_cleanup_queue(nullb->q);
 out_cleanup_tags:
@@ -625,6 +788,18 @@ static int __init null_init(void)
 		bs = PAGE_SIZE;
 	}
 
+	if (use_lightnvm && bs != 4096) {
+		pr_warn("null_blk: LightNVM only supports 4k block size\n");
+		pr_warn("null_blk: defaults block size to 4k\n");
+		bs = 4096;
+	}
+
+	if (use_lightnvm && queue_mode != NULL_Q_MQ) {
+		pr_warn("null_blk: LightNVM only supported for blk-mq\n");
+		pr_warn("null_blk: defaults queue mode to blk-mq\n");
+		queue_mode = NULL_Q_MQ;
+	}
+
 	if (queue_mode == NULL_Q_MQ && use_per_node_hctx) {
 		if (submit_queues < nr_online_nodes) {
 			pr_warn("null_blk: submit_queues param is set to %u.",
@@ -655,15 +830,27 @@ static int __init null_init(void)
 	if (null_major < 0)
 		return null_major;
 
+	if (use_lightnvm) {
+		ppa_cache = kmem_cache_create("ppa_cache", 64 * sizeof(u64),
+								0, 0, NULL);
+		if (!ppa_cache) {
+			pr_err("null_blk: unable to create ppa cache\n");
+			return -ENOMEM;
+		}
+	}
+
 	for (i = 0; i < nr_devices; i++) {
 		if (null_add_dev()) {
 			unregister_blkdev(null_major, "nullb");
-			return -EINVAL;
+			goto err_ppa;
 		}
 	}
 
 	pr_info("null: module loaded\n");
 	return 0;
+err_ppa:
+	kmem_cache_destroy(ppa_cache);
+	return -EINVAL;
 }
 
 static void __exit null_exit(void)
@@ -678,6 +865,8 @@ static void __exit null_exit(void)
 		null_del_dev(nullb);
 	}
 	mutex_unlock(&lock);
+
+	kmem_cache_destroy(ppa_cache);
 }
 
 module_init(null_init);

drivers/lightnvm/core.c

@@ -160,11 +160,6 @@ int nvm_erase_blk(struct nvm_dev *dev, struct nvm_block *blk)
 }
 EXPORT_SYMBOL(nvm_erase_blk);
 
-static void nvm_core_free(struct nvm_dev *dev)
-{
-	kfree(dev);
-}
-
 static int nvm_core_init(struct nvm_dev *dev)
 {
 	struct nvm_id *id = &dev->identity;
@@ -179,12 +174,21 @@ static int nvm_core_init(struct nvm_dev *dev)
 	dev->sec_size = grp->csecs;
 	dev->oob_size = grp->sos;
 	dev->sec_per_pg = grp->fpg_sz / grp->csecs;
-	dev->addr_mode = id->ppat;
-	dev->addr_format = id->ppaf;
+	memcpy(&dev->ppaf, &id->ppaf, sizeof(struct nvm_addr_format));
 
 	dev->plane_mode = NVM_PLANE_SINGLE;
 	dev->max_rq_size = dev->ops->max_phys_sect * dev->sec_size;
 
+	if (grp->mtype != 0) {
+		pr_err("nvm: memory type not supported\n");
+		return -EINVAL;
+	}
+
+	if (grp->fmtype != 0 && grp->fmtype != 1) {
+		pr_err("nvm: flash type not supported\n");
+		return -EINVAL;
+	}
+
 	if (grp->mpos & 0x020202)
 		dev->plane_mode = NVM_PLANE_DOUBLE;
 	if (grp->mpos & 0x040404)
@@ -213,21 +217,18 @@ static void nvm_free(struct nvm_dev *dev)
 
 	if (dev->mt)
 		dev->mt->unregister_mgr(dev);
-
-	nvm_core_free(dev);
 }
 
 static int nvm_init(struct nvm_dev *dev)
 {
 	struct nvmm_type *mt;
-	int ret = 0;
+	int ret = -EINVAL;
 
 	if (!dev->q || !dev->ops)
-		return -EINVAL;
+		return ret;
 
 	if (dev->ops->identity(dev->q, &dev->identity)) {
 		pr_err("nvm: device could not be identified\n");
-		ret = -EINVAL;
 		goto err;
 	}
 
@@ -273,7 +274,6 @@ static int nvm_init(struct nvm_dev *dev)
 			dev->nr_chnls);
 	return 0;
 err:
-	nvm_free(dev);
 	pr_err("nvm: failed to initialize nvm\n");
 	return ret;
 }
@@ -308,22 +308,24 @@ int nvm_register(struct request_queue *q, char *disk_name,
 	if (ret)
 		goto err_init;
 
-	down_write(&nvm_lock);
-	list_add(&dev->devices, &nvm_devices);
-	up_write(&nvm_lock);
-
 	if (dev->ops->max_phys_sect > 1) {
 		dev->ppalist_pool = dev->ops->create_dma_pool(dev->q,
 								"ppalist");
 		if (!dev->ppalist_pool) {
 			pr_err("nvm: could not create ppa pool\n");
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto err_init;
 		}
 	} else if (dev->ops->max_phys_sect > 256) {
 		pr_info("nvm: max sectors supported is 256.\n");
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_init;
 	}
 
+	down_write(&nvm_lock);
+	list_add(&dev->devices, &nvm_devices);
+	up_write(&nvm_lock);
+
 	return 0;
 err_init:
 	kfree(dev);
@@ -341,11 +343,12 @@ void nvm_unregister(char *disk_name)
 		return;
 	}
 
-	nvm_exit(dev);
-
 	down_write(&nvm_lock);
 	list_del(&dev->devices);
 	up_write(&nvm_lock);
+
+	nvm_exit(dev);
+	kfree(dev);
 }
 EXPORT_SYMBOL(nvm_unregister);
 
@@ -457,11 +460,11 @@ static void nvm_remove_target(struct nvm_target *t)
 	lockdep_assert_held(&nvm_lock);
 
 	del_gendisk(tdisk);
+	blk_cleanup_queue(q);
+
 	if (tt->exit)
 		tt->exit(tdisk->private_data);
 
-	blk_cleanup_queue(q);
-
 	put_disk(tdisk);
 
 	list_del(&t->list);
@@ -541,7 +544,7 @@ static int nvm_configure_show(const char *val)
 	if (!dev->mt)
 		return 0;
 
-	dev->mt->free_blocks_print(dev);
+	dev->mt->lun_info_print(dev);
 
 	return 0;
 }

drivers/lightnvm/gennvm.c

@@ -60,23 +60,28 @@ static int gennvm_luns_init(struct nvm_dev *dev, struct gen_nvm *gn)
 		lun->vlun.lun_id = i % dev->luns_per_chnl;
 		lun->vlun.chnl_id = i / dev->luns_per_chnl;
 		lun->vlun.nr_free_blocks = dev->blks_per_lun;
+		lun->vlun.nr_inuse_blocks = 0;
+		lun->vlun.nr_bad_blocks = 0;
 	}
 	return 0;
 }
 
-static int gennvm_block_bb(u32 lun_id, void *bb_bitmap, unsigned int nr_blocks,
+static int gennvm_block_bb(struct ppa_addr ppa, int nr_blocks, u8 *blks,
 								void *private)
 {
 	struct gen_nvm *gn = private;
-	struct gen_lun *lun = &gn->luns[lun_id];
+	struct nvm_dev *dev = gn->dev;
+	struct gen_lun *lun;
 	struct nvm_block *blk;
 	int i;
 
-	if (unlikely(bitmap_empty(bb_bitmap, nr_blocks)))
-		return 0;
+	ppa = dev_to_generic_addr(gn->dev, ppa);
+	lun = &gn->luns[(dev->nr_luns * ppa.g.ch) + ppa.g.lun];
+
+	for (i = 0; i < nr_blocks; i++) {
+		if (blks[i] == 0)
+			continue;
 
-	i = -1;
-	while ((i = find_next_bit(bb_bitmap, nr_blocks, i + 1)) < nr_blocks) {
 		blk = &lun->vlun.blocks[i];
 		if (!blk) {
 			pr_err("gennvm: BB data is out of bounds.\n");
@@ -84,6 +89,7 @@ static int gennvm_block_bb(u32 lun_id, void *bb_bitmap, unsigned int nr_blocks,
 		}
 
 		list_move_tail(&blk->list, &lun->bb_list);
+		lun->vlun.nr_bad_blocks++;
 	}
 
 	return 0;
@@ -136,6 +142,7 @@ static int gennvm_block_map(u64 slba, u32 nlb, __le64 *entries, void *private)
 			list_move_tail(&blk->list, &lun->used_list);
 			blk->type = 1;
 			lun->vlun.nr_free_blocks--;
+			lun->vlun.nr_inuse_blocks++;
 		}
 	}
 
@@ -164,15 +171,25 @@ static int gennvm_blocks_init(struct nvm_dev *dev, struct gen_nvm *gn)
 			block->id = cur_block_id++;
 
 			/* First block is reserved for device */
-			if (unlikely(lun_iter == 0 && blk_iter == 0))
+			if (unlikely(lun_iter == 0 && blk_iter == 0)) {
+				lun->vlun.nr_free_blocks--;
 				continue;
+			}
 
 			list_add_tail(&block->list, &lun->free_list);
 		}
 
 		if (dev->ops->get_bb_tbl) {
-			ret = dev->ops->get_bb_tbl(dev->q, lun->vlun.id,
-					dev->blks_per_lun, gennvm_block_bb, gn);
+			struct ppa_addr ppa;
+
+			ppa.ppa = 0;
+			ppa.g.ch = lun->vlun.chnl_id;
+			ppa.g.lun = lun->vlun.id;
+			ppa = generic_to_dev_addr(dev, ppa);
+
+			ret = dev->ops->get_bb_tbl(dev->q, ppa,
+						dev->blks_per_lun,
+						gennvm_block_bb, gn);
 			if (ret)
 				pr_err("gennvm: could not read BB table\n");
 		}
@@ -199,6 +216,7 @@ static int gennvm_register(struct nvm_dev *dev)
 	if (!gn)
 		return -ENOMEM;
 
+	gn->dev = dev;
 	gn->nr_luns = dev->nr_luns;
 	dev->mp = gn;
 
@@ -254,6 +272,7 @@ static struct nvm_block *gennvm_get_blk(struct nvm_dev *dev,
 	blk->type = 1;
 
 	lun->vlun.nr_free_blocks--;
+	lun->vlun.nr_inuse_blocks++;
 
 	spin_unlock(&vlun->lock);
 out:
@@ -271,16 +290,21 @@ static void gennvm_put_blk(struct nvm_dev *dev, struct nvm_block *blk)
 	case 1:
 		list_move_tail(&blk->list, &lun->free_list);
 		lun->vlun.nr_free_blocks++;
+		lun->vlun.nr_inuse_blocks--;
 		blk->type = 0;
 		break;
 	case 2:
 		list_move_tail(&blk->list, &lun->bb_list);
+		lun->vlun.nr_bad_blocks++;
+		lun->vlun.nr_inuse_blocks--;
 		break;
 	default:
 		WARN_ON_ONCE(1);
 		pr_err("gennvm: erroneous block type (%lu -> %u)\n",
 							blk->id, blk->type);
 		list_move_tail(&blk->list, &lun->bb_list);
+		lun->vlun.nr_bad_blocks++;
+		lun->vlun.nr_inuse_blocks--;
 	}
 
 	spin_unlock(&vlun->lock);
@@ -292,10 +316,10 @@ static void gennvm_addr_to_generic_mode(struct nvm_dev *dev, struct nvm_rq *rqd)
 
 	if (rqd->nr_pages > 1) {
 		for (i = 0; i < rqd->nr_pages; i++)
-			rqd->ppa_list[i] = addr_to_generic_mode(dev,
+			rqd->ppa_list[i] = dev_to_generic_addr(dev,
 							rqd->ppa_list[i]);
 	} else {
-		rqd->ppa_addr = addr_to_generic_mode(dev, rqd->ppa_addr);
+		rqd->ppa_addr = dev_to_generic_addr(dev, rqd->ppa_addr);
 	}
 }
 
@@ -305,10 +329,10 @@ static void gennvm_generic_to_addr_mode(struct nvm_dev *dev, struct nvm_rq *rqd)
 
 	if (rqd->nr_pages > 1) {
 		for (i = 0; i < rqd->nr_pages; i++)
-			rqd->ppa_list[i] = generic_to_addr_mode(dev,
+			rqd->ppa_list[i] = generic_to_dev_addr(dev,
 							rqd->ppa_list[i]);
 	} else {
-		rqd->ppa_addr = generic_to_addr_mode(dev, rqd->ppa_addr);
+		rqd->ppa_addr = generic_to_dev_addr(dev, rqd->ppa_addr);
 	}
 }
 
@@ -354,10 +378,10 @@ static void gennvm_mark_blk_bad(struct nvm_dev *dev, struct nvm_rq *rqd)
 {
 	int i;
 
-	if (!dev->ops->set_bb)
+	if (!dev->ops->set_bb_tbl)
 		return;
 
-	if (dev->ops->set_bb(dev->q, rqd, 1))
+	if (dev->ops->set_bb_tbl(dev->q, rqd, 1))
 		return;
 
 	gennvm_addr_to_generic_mode(dev, rqd);
@@ -440,15 +464,24 @@ static struct nvm_lun *gennvm_get_lun(struct nvm_dev *dev, int lunid)
 	return &gn->luns[lunid].vlun;
 }
 
-static void gennvm_free_blocks_print(struct nvm_dev *dev)
+static void gennvm_lun_info_print(struct nvm_dev *dev)
 {
 	struct gen_nvm *gn = dev->mp;
 	struct gen_lun *lun;
 	unsigned int i;
 
-	gennvm_for_each_lun(gn, lun, i)
-		pr_info("%s: lun%8u\t%u\n",
-					dev->name, i, lun->vlun.nr_free_blocks);
+
+	gennvm_for_each_lun(gn, lun, i) {
+		spin_lock(&lun->vlun.lock);
+
+		pr_info("%s: lun%8u\t%u\t%u\t%u\n",
+				dev->name, i,
+				lun->vlun.nr_free_blocks,
+				lun->vlun.nr_inuse_blocks,
+				lun->vlun.nr_bad_blocks);
+
+		spin_unlock(&lun->vlun.lock);
+	}
 }
 
 static struct nvmm_type gennvm = {
@@ -466,7 +499,7 @@ static struct nvmm_type gennvm = {
 	.erase_blk	= gennvm_erase_blk,
 
 	.get_lun	= gennvm_get_lun,
-	.free_blocks_print = gennvm_free_blocks_print,
+	.lun_info_print = gennvm_lun_info_print,
 };
 
 static int __init gennvm_module_init(void)

drivers/lightnvm/gennvm.h

@@ -35,6 +35,8 @@ struct gen_lun {
 };
 
 struct gen_nvm {
+	struct nvm_dev *dev;
+
 	int nr_luns;
 	struct gen_lun *luns;
 };

drivers/lightnvm/rrpc.c

@@ -123,12 +123,42 @@ static u64 block_to_addr(struct rrpc *rrpc, struct rrpc_block *rblk)
 	return blk->id * rrpc->dev->pgs_per_blk;
 }
 
+static struct ppa_addr linear_to_generic_addr(struct nvm_dev *dev,
+							struct ppa_addr r)
+{
+	struct ppa_addr l;
+	int secs, pgs, blks, luns;
+	sector_t ppa = r.ppa;
+
+	l.ppa = 0;
+
+	div_u64_rem(ppa, dev->sec_per_pg, &secs);
+	l.g.sec = secs;
+
+	sector_div(ppa, dev->sec_per_pg);
+	div_u64_rem(ppa, dev->sec_per_blk, &pgs);
+	l.g.pg = pgs;
+
+	sector_div(ppa, dev->pgs_per_blk);
+	div_u64_rem(ppa, dev->blks_per_lun, &blks);
+	l.g.blk = blks;
+
+	sector_div(ppa, dev->blks_per_lun);
+	div_u64_rem(ppa, dev->luns_per_chnl, &luns);
+	l.g.lun = luns;
+
+	sector_div(ppa, dev->luns_per_chnl);
+	l.g.ch = ppa;
+
+	return l;
+}
+
 static struct ppa_addr rrpc_ppa_to_gaddr(struct nvm_dev *dev, u64 addr)
 {
 	struct ppa_addr paddr;
 
 	paddr.ppa = addr;
-	return __linear_to_generic_addr(dev, paddr);
+	return linear_to_generic_addr(dev, paddr);
 }
 
 /* requires lun->lock taken */

drivers/nvme/host/lightnvm.c

@@ -93,7 +93,7 @@ struct nvme_nvm_l2ptbl {
 	__le16			cdw14[6];
 };
 
-struct nvme_nvm_bbtbl {
+struct nvme_nvm_getbbtbl {
 	__u8			opcode;
 	__u8			flags;
 	__u16			command_id;
@@ -101,10 +101,23 @@ struct nvme_nvm_bbtbl {
 	__u64			rsvd[2];
 	__le64			prp1;
 	__le64			prp2;
-	__le32			prp1_len;
-	__le32			prp2_len;
-	__le32			lbb;
-	__u32			rsvd11[3];
+	__le64			spba;
+	__u32			rsvd4[4];
+};
+
+struct nvme_nvm_setbbtbl {
+	__u8			opcode;
+	__u8			flags;
+	__u16			command_id;
+	__le32			nsid;
+	__le64			rsvd[2];
+	__le64			prp1;
+	__le64			prp2;
+	__le64			spba;
+	__le16			nlb;
+	__u8			value;
+	__u8			rsvd3;
+	__u32			rsvd4[3];
 };
 
 struct nvme_nvm_erase_blk {
@@ -129,8 +142,8 @@ struct nvme_nvm_command {
 		struct nvme_nvm_hb_rw hb_rw;
 		struct nvme_nvm_ph_rw ph_rw;
 		struct nvme_nvm_l2ptbl l2p;
-		struct nvme_nvm_bbtbl get_bb;
-		struct nvme_nvm_bbtbl set_bb;
+		struct nvme_nvm_getbbtbl get_bb;
+		struct nvme_nvm_setbbtbl set_bb;
 		struct nvme_nvm_erase_blk erase;
 	};
 };
@@ -142,11 +155,13 @@ struct nvme_nvm_id_group {
 	__u8			num_ch;
 	__u8			num_lun;
 	__u8			num_pln;
+	__u8			rsvd1;
 	__le16			num_blk;
 	__le16			num_pg;
 	__le16			fpg_sz;
 	__le16			csecs;
 	__le16			sos;
+	__le16			rsvd2;
 	__le32			trdt;
 	__le32			trdm;
 	__le32			tprt;
@@ -154,8 +169,9 @@ struct nvme_nvm_id_group {
 	__le32			tbet;
 	__le32			tbem;
 	__le32			mpos;
+	__le32			mccap;
 	__le16			cpar;
-	__u8			reserved[913];
+	__u8			reserved[906];
 } __packed;
 
 struct nvme_nvm_addr_format {
@@ -178,15 +194,28 @@ struct nvme_nvm_id {
 	__u8			ver_id;
 	__u8			vmnt;
 	__u8			cgrps;
-	__u8			res[5];
+	__u8			res;
 	__le32			cap;
 	__le32			dom;
 	struct nvme_nvm_addr_format ppaf;
-	__u8			ppat;
-	__u8			resv[223];
+	__u8			resv[228];
 	struct nvme_nvm_id_group groups[4];
 } __packed;
 
+struct nvme_nvm_bb_tbl {
+	__u8	tblid[4];
+	__le16	verid;
+	__le16	revid;
+	__le32	rvsd1;
+	__le32	tblks;
+	__le32	tfact;
+	__le32	tgrown;
+	__le32	tdresv;
+	__le32	thresv;
+	__le32	rsvd2[8];
+	__u8	blk[0];
+};
+
 /*
  * Check we didn't inadvertently grow the command struct
  */
@@ -195,12 +224,14 @@ static inline void _nvme_nvm_check_size(void)
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_identity) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_hb_rw) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_ph_rw) != 64);
-	BUILD_BUG_ON(sizeof(struct nvme_nvm_bbtbl) != 64);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_getbbtbl) != 64);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_setbbtbl) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_l2ptbl) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_erase_blk) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_id_group) != 960);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_addr_format) != 128);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_id) != 4096);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_bb_tbl) != 512);
 }
 
 static int init_grps(struct nvm_id *nvm_id, struct nvme_nvm_id *nvme_nvm_id)
@@ -234,6 +265,7 @@ static int init_grps(struct nvm_id *nvm_id, struct nvme_nvm_id *nvme_nvm_id)
 		dst->tbet = le32_to_cpu(src->tbet);
 		dst->tbem = le32_to_cpu(src->tbem);
 		dst->mpos = le32_to_cpu(src->mpos);
+		dst->mccap = le32_to_cpu(src->mccap);
 
 		dst->cpar = le16_to_cpu(src->cpar);
 	}
@@ -244,6 +276,7 @@ static int init_grps(struct nvm_id *nvm_id, struct nvme_nvm_id *nvme_nvm_id)
 static int nvme_nvm_identity(struct request_queue *q, struct nvm_id *nvm_id)
 {
 	struct nvme_ns *ns = q->queuedata;
+	struct nvme_dev *dev = ns->dev;
 	struct nvme_nvm_id *nvme_nvm_id;
 	struct nvme_nvm_command c = {};
 	int ret;
@@ -256,8 +289,8 @@ static int nvme_nvm_identity(struct request_queue *q, struct nvm_id *nvm_id)
 	if (!nvme_nvm_id)
 		return -ENOMEM;
 
-	ret = nvme_submit_sync_cmd(q, (struct nvme_command *)&c, nvme_nvm_id,
-						sizeof(struct nvme_nvm_id));
+	ret = nvme_submit_sync_cmd(dev->admin_q, (struct nvme_command *)&c,
+				nvme_nvm_id, sizeof(struct nvme_nvm_id));
 	if (ret) {
 		ret = -EIO;
 		goto out;
@@ -268,6 +301,8 @@ static int nvme_nvm_identity(struct request_queue *q, struct nvm_id *nvm_id)
 	nvm_id->cgrps = nvme_nvm_id->cgrps;
 	nvm_id->cap = le32_to_cpu(nvme_nvm_id->cap);
 	nvm_id->dom = le32_to_cpu(nvme_nvm_id->dom);
+	memcpy(&nvm_id->ppaf, &nvme_nvm_id->ppaf,
+					sizeof(struct nvme_nvm_addr_format));
 
 	ret = init_grps(nvm_id, nvme_nvm_id);
 out:
@@ -281,7 +316,7 @@ static int nvme_nvm_get_l2p_tbl(struct request_queue *q, u64 slba, u32 nlb,
 	struct nvme_ns *ns = q->queuedata;
 	struct nvme_dev *dev = ns->dev;
 	struct nvme_nvm_command c = {};
-	u32 len = queue_max_hw_sectors(q) << 9;
+	u32 len = queue_max_hw_sectors(dev->admin_q) << 9;
 	u32 nlb_pr_rq = len / sizeof(u64);
 	u64 cmd_slba = slba;
 	void *entries;
@@ -299,8 +334,8 @@ static int nvme_nvm_get_l2p_tbl(struct request_queue *q, u64 slba, u32 nlb,
 		c.l2p.slba = cpu_to_le64(cmd_slba);
 		c.l2p.nlb = cpu_to_le32(cmd_nlb);
 
-		ret = nvme_submit_sync_cmd(q, (struct nvme_command *)&c,
-								entries, len);
+		ret = nvme_submit_sync_cmd(dev->admin_q,
+				(struct nvme_command *)&c, entries, len);
 		if (ret) {
 			dev_err(dev->dev, "L2P table transfer failed (%d)\n",
 									ret);
@@ -322,43 +357,82 @@ out:
 	return ret;
 }
 
-static int nvme_nvm_get_bb_tbl(struct request_queue *q, int lunid,
-				unsigned int nr_blocks,
-				nvm_bb_update_fn *update_bbtbl, void *priv)
+static int nvme_nvm_get_bb_tbl(struct request_queue *q, struct ppa_addr ppa,
+				int nr_blocks, nvm_bb_update_fn *update_bbtbl,
+				void *priv)
 {
 	struct nvme_ns *ns = q->queuedata;
 	struct nvme_dev *dev = ns->dev;
 	struct nvme_nvm_command c = {};
-	void *bb_bitmap;
-	u16 bb_bitmap_size;
+	struct nvme_nvm_bb_tbl *bb_tbl;
+	int tblsz = sizeof(struct nvme_nvm_bb_tbl) + nr_blocks;
 	int ret = 0;
 
 	c.get_bb.opcode = nvme_nvm_admin_get_bb_tbl;
 	c.get_bb.nsid = cpu_to_le32(ns->ns_id);
-	c.get_bb.lbb = cpu_to_le32(lunid);
-	bb_bitmap_size = ((nr_blocks >> 15) + 1) * PAGE_SIZE;
-	bb_bitmap = kmalloc(bb_bitmap_size, GFP_KERNEL);
-	if (!bb_bitmap)
-		return -ENOMEM;
+	c.get_bb.spba = cpu_to_le64(ppa.ppa);
 
-	bitmap_zero(bb_bitmap, nr_blocks);
+	bb_tbl = kzalloc(tblsz, GFP_KERNEL);
+	if (!bb_tbl)
+		return -ENOMEM;
 
-	ret = nvme_submit_sync_cmd(q, (struct nvme_command *)&c, bb_bitmap,
-								bb_bitmap_size);
+	ret = nvme_submit_sync_cmd(dev->admin_q, (struct nvme_command *)&c,
+								bb_tbl, tblsz);
 	if (ret) {
 		dev_err(dev->dev, "get bad block table failed (%d)\n", ret);
 		ret = -EIO;
 		goto out;
 	}
 
-	ret = update_bbtbl(lunid, bb_bitmap, nr_blocks, priv);
+	if (bb_tbl->tblid[0] != 'B' || bb_tbl->tblid[1] != 'B' ||
+		bb_tbl->tblid[2] != 'L' || bb_tbl->tblid[3] != 'T') {
+		dev_err(dev->dev, "bbt format mismatch\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (le16_to_cpu(bb_tbl->verid) != 1) {
+		ret = -EINVAL;
+		dev_err(dev->dev, "bbt version not supported\n");
+		goto out;
+	}
+
+	if (le32_to_cpu(bb_tbl->tblks) != nr_blocks) {
+		ret = -EINVAL;
+		dev_err(dev->dev, "bbt unsuspected blocks returned (%u!=%u)",
+					le32_to_cpu(bb_tbl->tblks), nr_blocks);
+		goto out;
+	}
+
+	ret = update_bbtbl(ppa, nr_blocks, bb_tbl->blk, priv);
 	if (ret) {
 		ret = -EINTR;
 		goto out;
 	}
 
 out:
-	kfree(bb_bitmap);
+	kfree(bb_tbl);
+	return ret;
+}
+
+static int nvme_nvm_set_bb_tbl(struct request_queue *q, struct nvm_rq *rqd,
+								int type)
+{
+	struct nvme_ns *ns = q->queuedata;
+	struct nvme_dev *dev = ns->dev;
+	struct nvme_nvm_command c = {};
+	int ret = 0;
+
+	c.set_bb.opcode = nvme_nvm_admin_set_bb_tbl;
+	c.set_bb.nsid = cpu_to_le32(ns->ns_id);
+	c.set_bb.spba = cpu_to_le64(rqd->ppa_addr.ppa);
+	c.set_bb.nlb = cpu_to_le16(rqd->nr_pages - 1);
+	c.set_bb.value = type;
+
+	ret = nvme_submit_sync_cmd(dev->admin_q, (struct nvme_command *)&c,
+								NULL, 0);
+	if (ret)
+		dev_err(dev->dev, "set bad block table failed (%d)\n", ret);
 	return ret;
 }
 
@@ -474,6 +548,7 @@ static struct nvm_dev_ops nvme_nvm_dev_ops = {
 	.get_l2p_tbl		= nvme_nvm_get_l2p_tbl,
 
 	.get_bb_tbl		= nvme_nvm_get_bb_tbl,
+	.set_bb_tbl		= nvme_nvm_set_bb_tbl,
 
 	.submit_io		= nvme_nvm_submit_io,
 	.erase_block		= nvme_nvm_erase_block,

drivers/nvme/host/pci.c

@@ -968,7 +968,8 @@ static void __nvme_process_cq(struct nvme_queue *nvmeq, unsigned int *tag)
 	if (head == nvmeq->cq_head && phase == nvmeq->cq_phase)
 		return;
 
-	writel(head, nvmeq->q_db + nvmeq->dev->db_stride);
+	if (likely(nvmeq->cq_vector >= 0))
+		writel(head, nvmeq->q_db + nvmeq->dev->db_stride);
 	nvmeq->cq_head = head;
 	nvmeq->cq_phase = phase;
 
@@ -2268,7 +2269,7 @@ static void nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid)
 	if (dev->max_hw_sectors) {
 		blk_queue_max_hw_sectors(ns->queue, dev->max_hw_sectors);
 		blk_queue_max_segments(ns->queue,
-			((dev->max_hw_sectors << 9) / dev->page_size) + 1);
+			(dev->max_hw_sectors / (dev->page_size >> 9)) + 1);
 	}
 	if (dev->stripe_size)
 		blk_queue_chunk_sectors(ns->queue, dev->stripe_size >> 9);
@@ -2787,6 +2788,10 @@ static void nvme_del_queue_end(struct nvme_queue *nvmeq)
 {
 	struct nvme_delq_ctx *dq = nvmeq->cmdinfo.ctx;
 	nvme_put_dq(dq);
+
+	spin_lock_irq(&nvmeq->q_lock);
+	nvme_process_cq(nvmeq);
+	spin_unlock_irq(&nvmeq->q_lock);
 }
 
 static int adapter_async_del_queue(struct nvme_queue *nvmeq, u8 opcode,

include/linux/lightnvm.h

@@ -58,7 +58,6 @@ enum {
 struct nvm_id_group {
 	u8	mtype;
 	u8	fmtype;
-	u16	res16;
 	u8	num_ch;
 	u8	num_lun;
 	u8	num_pln;
@@ -74,9 +73,9 @@ struct nvm_id_group {
 	u32	tbet;
 	u32	tbem;
 	u32	mpos;
+	u32	mccap;
 	u16	cpar;
-	u8	res[913];
-} __packed;
+};
 
 struct nvm_addr_format {
 	u8	ch_offset;
@@ -91,19 +90,15 @@ struct nvm_addr_format {
 	u8	pg_len;
 	u8	sect_offset;
 	u8	sect_len;
-	u8	res[4];
 };
 
 struct nvm_id {
 	u8	ver_id;
 	u8	vmnt;
 	u8	cgrps;
-	u8	res[5];
 	u32	cap;
 	u32	dom;
 	struct nvm_addr_format ppaf;
-	u8	ppat;
-	u8	resv[224];
 	struct nvm_id_group groups[4];
 } __packed;
 
@@ -123,39 +118,28 @@ struct nvm_tgt_instance {
 #define NVM_VERSION_MINOR 0
 #define NVM_VERSION_PATCH 0
 
-#define NVM_SEC_BITS (8)
-#define NVM_PL_BITS  (6)
-#define NVM_PG_BITS  (16)
 #define NVM_BLK_BITS (16)
-#define NVM_LUN_BITS (10)
+#define NVM_PG_BITS  (16)
+#define NVM_SEC_BITS (8)
+#define NVM_PL_BITS  (8)
+#define NVM_LUN_BITS (8)
 #define NVM_CH_BITS  (8)
 
 struct ppa_addr {
+	/* Generic structure for all addresses */
 	union {
-		/* Channel-based PPA format in nand 4x2x2x2x8x10 */
-		struct {
-			u64 ch		: 4;
-			u64 sec		: 2; /* 4 sectors per page */
-			u64 pl		: 2; /* 4 planes per LUN */
-			u64 lun		: 2; /* 4 LUNs per channel */
-			u64 pg		: 8; /* 256 pages per block */
-			u64 blk		: 10;/* 1024 blocks per plane */
-			u64 resved		: 36;
-		} chnl;
-
-		/* Generic structure for all addresses */
 		struct {
+			u64 blk		: NVM_BLK_BITS;
+			u64 pg		: NVM_PG_BITS;
 			u64 sec		: NVM_SEC_BITS;
 			u64 pl		: NVM_PL_BITS;
-			u64 pg		: NVM_PG_BITS;
-			u64 blk		: NVM_BLK_BITS;
 			u64 lun		: NVM_LUN_BITS;
 			u64 ch		: NVM_CH_BITS;
 		} g;
 
 		u64 ppa;
 	};
-} __packed;
+};
 
 struct nvm_rq {
 	struct nvm_tgt_instance *ins;
@@ -191,11 +175,11 @@ static inline void *nvm_rq_to_pdu(struct nvm_rq *rqdata)
 struct nvm_block;
 
 typedef int (nvm_l2p_update_fn)(u64, u32, __le64 *, void *);
-typedef int (nvm_bb_update_fn)(u32, void *, unsigned int, void *);
+typedef int (nvm_bb_update_fn)(struct ppa_addr, int, u8 *, void *);
 typedef int (nvm_id_fn)(struct request_queue *, struct nvm_id *);
 typedef int (nvm_get_l2p_tbl_fn)(struct request_queue *, u64, u32,
 				nvm_l2p_update_fn *, void *);
-typedef int (nvm_op_bb_tbl_fn)(struct request_queue *, int, unsigned int,
+typedef int (nvm_op_bb_tbl_fn)(struct request_queue *, struct ppa_addr, int,
 				nvm_bb_update_fn *, void *);
 typedef int (nvm_op_set_bb_fn)(struct request_queue *, struct nvm_rq *, int);
 typedef int (nvm_submit_io_fn)(struct request_queue *, struct nvm_rq *);
@@ -210,7 +194,7 @@ struct nvm_dev_ops {
 	nvm_id_fn		*identity;
 	nvm_get_l2p_tbl_fn	*get_l2p_tbl;
 	nvm_op_bb_tbl_fn	*get_bb_tbl;
-	nvm_op_set_bb_fn	*set_bb;
+	nvm_op_set_bb_fn	*set_bb_tbl;
 
 	nvm_submit_io_fn	*submit_io;
 	nvm_erase_blk_fn	*erase_block;
@@ -220,7 +204,7 @@ struct nvm_dev_ops {
 	nvm_dev_dma_alloc_fn	*dev_dma_alloc;
 	nvm_dev_dma_free_fn	*dev_dma_free;
 
-	uint8_t			max_phys_sect;
+	unsigned int		max_phys_sect;
 };
 
 struct nvm_lun {
@@ -229,7 +213,9 @@ struct nvm_lun {
 	int lun_id;
 	int chnl_id;
 
+	unsigned int nr_inuse_blocks;	/* Number of used blocks */
 	unsigned int nr_free_blocks;	/* Number of unused blocks */
+	unsigned int nr_bad_blocks;	/* Number of bad blocks */
 	struct nvm_block *blocks;
 
 	spinlock_t lock;
@@ -263,8 +249,7 @@ struct nvm_dev {
 	int blks_per_lun;
 	int sec_size;
 	int oob_size;
-	int addr_mode;
-	struct nvm_addr_format addr_format;
+	struct nvm_addr_format ppaf;
 
 	/* Calculated/Cached values. These do not reflect the actual usable
 	 * blocks at run-time.
@@ -290,118 +275,45 @@ struct nvm_dev {
 	char name[DISK_NAME_LEN];
 };
 
-/* fallback conversion */
-static struct ppa_addr __generic_to_linear_addr(struct nvm_dev *dev,
-							struct ppa_addr r)
-{
-	struct ppa_addr l;
-
-	l.ppa = r.g.sec +
-		r.g.pg  * dev->sec_per_pg +
-		r.g.blk * (dev->pgs_per_blk *
-				dev->sec_per_pg) +
-		r.g.lun * (dev->blks_per_lun *
-				dev->pgs_per_blk *
-				dev->sec_per_pg) +
-		r.g.ch * (dev->blks_per_lun *
-				dev->pgs_per_blk *
-				dev->luns_per_chnl *
-				dev->sec_per_pg);
-
-	return l;
-}
-
-/* fallback conversion */
-static struct ppa_addr __linear_to_generic_addr(struct nvm_dev *dev,
-							struct ppa_addr r)
+static inline struct ppa_addr generic_to_dev_addr(struct nvm_dev *dev,
+						struct ppa_addr r)
 {
 	struct ppa_addr l;
-	int secs, pgs, blks, luns;
-	sector_t ppa = r.ppa;
 
-	l.ppa = 0;
-
-	div_u64_rem(ppa, dev->sec_per_pg, &secs);
-	l.g.sec = secs;
-
-	sector_div(ppa, dev->sec_per_pg);
-	div_u64_rem(ppa, dev->sec_per_blk, &pgs);
-	l.g.pg = pgs;
-
-	sector_div(ppa, dev->pgs_per_blk);
-	div_u64_rem(ppa, dev->blks_per_lun, &blks);
-	l.g.blk = blks;
-
-	sector_div(ppa, dev->blks_per_lun);
-	div_u64_rem(ppa, dev->luns_per_chnl, &luns);
-	l.g.lun = luns;
-
-	sector_div(ppa, dev->luns_per_chnl);
-	l.g.ch = ppa;
+	l.ppa = ((u64)r.g.blk) << dev->ppaf.blk_offset;
+	l.ppa |= ((u64)r.g.pg) << dev->ppaf.pg_offset;
+	l.ppa |= ((u64)r.g.sec) << dev->ppaf.sect_offset;
+	l.ppa |= ((u64)r.g.pl) << dev->ppaf.pln_offset;
+	l.ppa |= ((u64)r.g.lun) << dev->ppaf.lun_offset;
+	l.ppa |= ((u64)r.g.ch) << dev->ppaf.ch_offset;
 
 	return l;
 }
 
-static struct ppa_addr __generic_to_chnl_addr(struct ppa_addr r)
+static inline struct ppa_addr dev_to_generic_addr(struct nvm_dev *dev,
+						struct ppa_addr r)
 {
 	struct ppa_addr l;
 
-	l.ppa = 0;
-
-	l.chnl.sec = r.g.sec;
-	l.chnl.pl = r.g.pl;
-	l.chnl.pg = r.g.pg;
-	l.chnl.blk = r.g.blk;
-	l.chnl.lun = r.g.lun;
-	l.chnl.ch = r.g.ch;
-
-	return l;
-}
-
-static struct ppa_addr __chnl_to_generic_addr(struct ppa_addr r)
-{
-	struct ppa_addr l;
-
-	l.ppa = 0;
-
-	l.g.sec = r.chnl.sec;
-	l.g.pl = r.chnl.pl;
-	l.g.pg = r.chnl.pg;
-	l.g.blk = r.chnl.blk;
-	l.g.lun = r.chnl.lun;
-	l.g.ch = r.chnl.ch;
+	/*
+	 * (r.ppa << X offset) & X len bitmask. X eq. blk, pg, etc.
+	 */
+	l.g.blk = (r.ppa >> dev->ppaf.blk_offset) &
+					(((1 << dev->ppaf.blk_len) - 1));
+	l.g.pg |= (r.ppa >> dev->ppaf.pg_offset) &
+					(((1 << dev->ppaf.pg_len) - 1));
+	l.g.sec |= (r.ppa >> dev->ppaf.sect_offset) &
+					(((1 << dev->ppaf.sect_len) - 1));
+	l.g.pl |= (r.ppa >> dev->ppaf.pln_offset) &
+					(((1 << dev->ppaf.pln_len) - 1));
+	l.g.lun |= (r.ppa >> dev->ppaf.lun_offset) &
+					(((1 << dev->ppaf.lun_len) - 1));
+	l.g.ch |= (r.ppa >> dev->ppaf.ch_offset) &
+					(((1 << dev->ppaf.ch_len) - 1));
 
 	return l;
 }
 
-static inline struct ppa_addr addr_to_generic_mode(struct nvm_dev *dev,
-						struct ppa_addr gppa)
-{
-	switch (dev->addr_mode) {
-	case NVM_ADDRMODE_LINEAR:
-		return __linear_to_generic_addr(dev, gppa);
-	case NVM_ADDRMODE_CHANNEL:
-		return __chnl_to_generic_addr(gppa);
-	default:
-		BUG();
-	}
-	return gppa;
-}
-
-static inline struct ppa_addr generic_to_addr_mode(struct nvm_dev *dev,
-						struct ppa_addr gppa)
-{
-	switch (dev->addr_mode) {
-	case NVM_ADDRMODE_LINEAR:
-		return __generic_to_linear_addr(dev, gppa);
-	case NVM_ADDRMODE_CHANNEL:
-		return __generic_to_chnl_addr(gppa);
-	default:
-		BUG();
-	}
-	return gppa;
-}
-
 static inline int ppa_empty(struct ppa_addr ppa_addr)
 {
 	return (ppa_addr.ppa == ADDR_EMPTY);
@@ -468,7 +380,7 @@ typedef int (nvmm_end_io_fn)(struct nvm_rq *, int);
 typedef int (nvmm_erase_blk_fn)(struct nvm_dev *, struct nvm_block *,
 								unsigned long);
 typedef struct nvm_lun *(nvmm_get_lun_fn)(struct nvm_dev *, int);
-typedef void (nvmm_free_blocks_print_fn)(struct nvm_dev *);
+typedef void (nvmm_lun_info_print_fn)(struct nvm_dev *);
 
 struct nvmm_type {
 	const char *name;
@@ -492,7 +404,7 @@ struct nvmm_type {
 	nvmm_get_lun_fn *get_lun;
 
 	/* Statistics */
-	nvmm_free_blocks_print_fn *free_blocks_print;
+	nvmm_lun_info_print_fn *lun_info_print;
 	struct list_head list;
 };