Merge tag 'for-4.21/block-20181221' of git://git.kernel.dk/linux-block

Pull block updates from Jens Axboe:
 "This is the main pull request for block/storage for 4.21.

  Larger than usual, it was a busy round with lots of goodies queued up.
  Most notable is the removal of the old IO stack, which has been a long
  time coming. No new features for a while, everything coming in this
  week has all been fixes for things that were previously merged.

  This contains:

   - Use atomic counters instead of semaphores for mtip32xx (Arnd)

   - Cleanup of the mtip32xx request setup (Christoph)

   - Fix for circular locking dependency in loop (Jan, Tetsuo)

   - bcache (Coly, Guoju, Shenghui)
      * Optimizations for writeback caching
      * Various fixes and improvements

   - nvme (Chaitanya, Christoph, Sagi, Jay, me, Keith)
      * host and target support for NVMe over TCP
      * Error log page support
      * Support for separate read/write/poll queues
      * Much improved polling
      * discard OOM fallback
      * Tracepoint improvements

   - lightnvm (Hans, Hua, Igor, Matias, Javier)
      * Igor added packed metadata to pblk. Now drives without metadata
        per LBA can be used as well.
      * Fix from Geert on uninitialized value on chunk metadata reads.
      * Fixes from Hans and Javier to pblk recovery and write path.
      * Fix from Hua Su to fix a race condition in the pblk recovery
        code.
      * Scan optimization added to pblk recovery from Zhoujie.
      * Small geometry cleanup from me.

   - Conversion of the last few drivers that used the legacy path to
     blk-mq (me)

   - Removal of legacy IO path in SCSI (me, Christoph)

   - Removal of legacy IO stack and schedulers (me)

   - Support for much better polling, now without interrupts at all.
     blk-mq adds support for multiple queue maps, which enables us to
     have a map per type. This in turn enables nvme to have separate
     completion queues for polling, which can then be interrupt-less.
     Also means we're ready for async polled IO, which is hopefully
     coming in the next release.

   - Killing of (now) unused block exports (Christoph)

   - Unification of the blk-rq-qos and blk-wbt wait handling (Josef)

   - Support for zoned testing with null_blk (Masato)

   - sx8 conversion to per-host tag sets (Christoph)

   - IO priority improvements (Damien)

   - mq-deadline zoned fix (Damien)

   - Ref count blkcg series (Dennis)

   - Lots of blk-mq improvements and speedups (me)

   - sbitmap scalability improvements (me)

   - Make core inflight IO accounting per-cpu (Mikulas)

   - Export timeout setting in sysfs (Weiping)

   - Cleanup the direct issue path (Jianchao)

   - Export blk-wbt internals in block debugfs for easier debugging
     (Ming)

   - Lots of other fixes and improvements"

* tag 'for-4.21/block-20181221' of git://git.kernel.dk/linux-block: (364 commits)
  kyber: use sbitmap add_wait_queue/list_del wait helpers
  sbitmap: add helpers for add/del wait queue handling
  block: save irq state in blkg_lookup_create()
  dm: don't reuse bio for flushes
  nvme-pci: trace SQ status on completions
  nvme-rdma: implement polling queue map
  nvme-fabrics: allow user to pass in nr_poll_queues
  nvme-fabrics: allow nvmf_connect_io_queue to poll
  nvme-core: optionally poll sync commands
  block: make request_to_qc_t public
  nvme-tcp: fix spelling mistake "attepmpt" -> "attempt"
  nvme-tcp: fix endianess annotations
  nvmet-tcp: fix endianess annotations
  nvme-pci: refactor nvme_poll_irqdisable to make sparse happy
  nvme-pci: only set nr_maps to 2 if poll queues are supported
  nvmet: use a macro for default error location
  nvmet: fix comparison of a u16 with -1
  blk-mq: enable IO poll if .nr_queues of type poll > 0
  blk-mq: change blk_mq_queue_busy() to blk_mq_queue_inflight()
  blk-mq: skip zero-queue maps in blk_mq_map_swqueue
  ...

1 files changed, 79 insertions, 24 deletions

diff --git a/drivers/lightnvm/pblk-init.c b/drivers/lightnvm/pblk-init.c
index 13822594647c..f9a3e47b6a93 100644
--- a/drivers/lightnvm/pblk-init.c
+++ b/drivers/lightnvm/pblk-init.c
@@ -207,9 +207,6 @@ static int pblk_rwb_init(struct pblk *pblk)
 	return pblk_rb_init(&pblk->rwb, buffer_size, threshold, geo->csecs);
 }
 
-/* Minimum pages needed within a lun */
-#define ADDR_POOL_SIZE 64
-
 static int pblk_set_addrf_12(struct pblk *pblk, struct nvm_geo *geo,
 			     struct nvm_addrf_12 *dst)
 {
@@ -350,23 +347,19 @@ fail_destroy_ws:
 
 static int pblk_get_global_caches(void)
 {
-	int ret;
+	int ret = 0;
 
 	mutex_lock(&pblk_caches.mutex);
 
-	if (kref_read(&pblk_caches.kref) > 0) {
-		kref_get(&pblk_caches.kref);
-		mutex_unlock(&pblk_caches.mutex);
-		return 0;
-	}
+	if (kref_get_unless_zero(&pblk_caches.kref))
+		goto out;
 
 	ret = pblk_create_global_caches();
-
 	if (!ret)
-		kref_get(&pblk_caches.kref);
+		kref_init(&pblk_caches.kref);
 
+out:
 	mutex_unlock(&pblk_caches.mutex);
-
 	return ret;
 }
 
@@ -406,12 +399,45 @@ static int pblk_core_init(struct pblk *pblk)
 	pblk->nr_flush_rst = 0;
 
 	pblk->min_write_pgs = geo->ws_opt;
+	pblk->min_write_pgs_data = pblk->min_write_pgs;
 	max_write_ppas = pblk->min_write_pgs * geo->all_luns;
 	pblk->max_write_pgs = min_t(int, max_write_ppas, NVM_MAX_VLBA);
 	pblk->max_write_pgs = min_t(int, pblk->max_write_pgs,
 		queue_max_hw_sectors(dev->q) / (geo->csecs >> SECTOR_SHIFT));
 	pblk_set_sec_per_write(pblk, pblk->min_write_pgs);
 
+	pblk->oob_meta_size = geo->sos;
+	if (!pblk_is_oob_meta_supported(pblk)) {
+		/* For drives which does not have OOB metadata feature
+		 * in order to support recovery feature we need to use
+		 * so called packed metadata. Packed metada will store
+		 * the same information as OOB metadata (l2p table mapping,
+		 * but in the form of the single page at the end of
+		 * every write request.
+		 */
+		if (pblk->min_write_pgs
+			* sizeof(struct pblk_sec_meta) > PAGE_SIZE) {
+			/* We want to keep all the packed metadata on single
+			 * page per write requests. So we need to ensure that
+			 * it will fit.
+			 *
+			 * This is more like sanity check, since there is
+			 * no device with such a big minimal write size
+			 * (above 1 metabytes).
+			 */
+			pblk_err(pblk, "Not supported min write size\n");
+			return -EINVAL;
+		}
+		/* For packed meta approach we do some simplification.
+		 * On read path we always issue requests which size
+		 * equal to max_write_pgs, with all pages filled with
+		 * user payload except of last one page which will be
+		 * filled with packed metadata.
+		 */
+		pblk->max_write_pgs = pblk->min_write_pgs;
+		pblk->min_write_pgs_data = pblk->min_write_pgs - 1;
+	}
+
 	pblk->pad_dist = kcalloc(pblk->min_write_pgs - 1, sizeof(atomic64_t),
 								GFP_KERNEL);
 	if (!pblk->pad_dist)
@@ -635,40 +661,61 @@ static unsigned int calc_emeta_len(struct pblk *pblk)
 	return (lm->emeta_len[1] + lm->emeta_len[2] + lm->emeta_len[3]);
 }
 
-static void pblk_set_provision(struct pblk *pblk, long nr_free_blks)
+static int pblk_set_provision(struct pblk *pblk, int nr_free_chks)
 {
 	struct nvm_tgt_dev *dev = pblk->dev;
 	struct pblk_line_mgmt *l_mg = &pblk->l_mg;
 	struct pblk_line_meta *lm = &pblk->lm;
 	struct nvm_geo *geo = &dev->geo;
 	sector_t provisioned;
-	int sec_meta, blk_meta;
+	int sec_meta, blk_meta, clba;
+	int minimum;
 
 	if (geo->op == NVM_TARGET_DEFAULT_OP)
 		pblk->op = PBLK_DEFAULT_OP;
 	else
 		pblk->op = geo->op;
 
-	provisioned = nr_free_blks;
+	minimum = pblk_get_min_chks(pblk);
+	provisioned = nr_free_chks;
 	provisioned *= (100 - pblk->op);
 	sector_div(provisioned, 100);
 
-	pblk->op_blks = nr_free_blks - provisioned;
+	if ((nr_free_chks - provisioned) < minimum) {
+		if (geo->op != NVM_TARGET_DEFAULT_OP) {
+			pblk_err(pblk, "OP too small to create a sane instance\n");
+			return -EINTR;
+		}
+
+		/* If the user did not specify an OP value, and PBLK_DEFAULT_OP
+		 * is not enough, calculate and set sane value
+		 */
+
+		provisioned = nr_free_chks - minimum;
+		pblk->op =  (100 * minimum) / nr_free_chks;
+		pblk_info(pblk, "Default OP insufficient, adjusting OP to %d\n",
+				pblk->op);
+	}
+
+	pblk->op_blks = nr_free_chks - provisioned;
 
 	/* Internally pblk manages all free blocks, but all calculations based
 	 * on user capacity consider only provisioned blocks
 	 */
-	pblk->rl.total_blocks = nr_free_blks;
-	pblk->rl.nr_secs = nr_free_blks * geo->clba;
+	pblk->rl.total_blocks = nr_free_chks;
+	pblk->rl.nr_secs = nr_free_chks * geo->clba;
 
 	/* Consider sectors used for metadata */
 	sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
 	blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
 
-	pblk->capacity = (provisioned - blk_meta) * geo->clba;
+	clba = (geo->clba / pblk->min_write_pgs) * pblk->min_write_pgs_data;
+	pblk->capacity = (provisioned - blk_meta) * clba;
 
-	atomic_set(&pblk->rl.free_blocks, nr_free_blks);
-	atomic_set(&pblk->rl.free_user_blocks, nr_free_blks);
+	atomic_set(&pblk->rl.free_blocks, nr_free_chks);
+	atomic_set(&pblk->rl.free_user_blocks, nr_free_chks);
+
+	return 0;
 }
 
 static int pblk_setup_line_meta_chk(struct pblk *pblk, struct pblk_line *line,
@@ -984,7 +1031,7 @@ static int pblk_lines_init(struct pblk *pblk)
 	struct pblk_line_mgmt *l_mg = &pblk->l_mg;
 	struct pblk_line *line;
 	void *chunk_meta;
-	long nr_free_chks = 0;
+	int nr_free_chks = 0;
 	int i, ret;
 
 	ret = pblk_line_meta_init(pblk);
@@ -1031,7 +1078,9 @@ static int pblk_lines_init(struct pblk *pblk)
 		goto fail_free_lines;
 	}
 
-	pblk_set_provision(pblk, nr_free_chks);
+	ret = pblk_set_provision(pblk, nr_free_chks);
+	if (ret)
+		goto fail_free_lines;
 
 	vfree(chunk_meta);
 	return 0;
@@ -1041,7 +1090,7 @@ fail_free_lines:
 		pblk_line_meta_free(l_mg, &pblk->lines[i]);
 	kfree(pblk->lines);
 fail_free_chunk_meta:
-	kfree(chunk_meta);
+	vfree(chunk_meta);
 fail_free_luns:
 	kfree(pblk->luns);
 fail_free_meta:
@@ -1154,6 +1203,12 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk,
 		return ERR_PTR(-EINVAL);
 	}
 
+	if (geo->ext) {
+		pblk_err(pblk, "extended metadata not supported\n");
+		kfree(pblk);
+		return ERR_PTR(-EINVAL);
+	}
+
 	spin_lock_init(&pblk->resubmit_lock);
 	spin_lock_init(&pblk->trans_lock);
 	spin_lock_init(&pblk->lock);